1

The Organizing Committee gratefully acknowledges the Symposium Program Partners for their generous support of WCBP 2012

Sustaining Platinum Program Partner Agilent Technologies

Diamond Program Partner Genentech, a Member of the Roche Group

Gold Program Partners Biogen Idec MedImmune

Silver Program Partners Abbott Laboratories

Amgen Inc. Bristol-Myers Squibb Company

Eli Lilly and Company Merck

Novartis Pharma AG Pfizer, Inc.

Waters Corporation Bronze Program Partners

BioMarin Pharmaceutical Inc. Novo Nordisk A/S

Friend of CASSS Program Partner Genzyme, A Sanofi Company

Janssen Research & Development, LLC

2

Exhibitor Partners

AAIPharma Services NanoSight ABC Laboratories, Inc. Postnova Analytics

Affinity Biosensors PPD Agilent Technologies Protein Metrics, Inc.

Aptuit ProteinSimple ARTEL ProZyme, Inc.

Bay Bioanalytical Laboratory, Inc.

Sepax Technologies, Inc.

Beckman Coulter SGS M-Scan Inc. Bruker Daltonics, Inc. Thermo Scientific

Catalent Pharma Solutions Waters Corporation Fluid Imaging Technologies, Inc. Wyatt Technology Corporation

Lancaster Laboratories, Inc YMC America, Inc. Molecular Devices, LLC Zhejiang Teruisi Pharmaceutical

Inc. NanoImaging Services

Media Partners

BioPharm International International Pharmaceutical Quality

BioProcess International LCGC North America

Genetic Engineering & Biotechnology News

3

Acknowledgements Symposium Co-Chairs Workshop Committee Co-Chairs Roman Drews, CBER, FDA Laura Bass, Pfizer, Inc. Peter Johnson, 3M Drug Delivery Systems Kathleen Francissen, Genentech, a Member of the

Roche Group Susan Kirshner, CDER, FDA Arifa Khan, CBER, FDA Marjorie Shapiro, CDER, FDA

Permanent Committee Robert Cunico, Bay Bioanalytical Laboratory, Inc. Michael Kunitani, Marin Analytical Consulting John Dougherty, Eli Lilly and Company (Chair) Thomas Layloff, Supply Chain Management

System John Frenz, GlobeImmune, Inc. Robert Sitrin, Sitrin Solutions Christopher Joneckis, CBER, FDA

Special thanks to all the Workshop Session Co-Leaders Program Committee Sid Advant, ImClone Systems Corporation Lotte K. McNamara, LKM CMC Consulting Mehrshid Alai-Safar, Baxter Healthcare Corporation

Edwin Moore, Baxter Healthcare Corporation

Robert Baffi, BioMarin Pharmaceutical Inc. Brian K. Nunnally, Pfizer, Inc. Markus Blümel, Novartis Pharma AG Stefanie Pluschkell, Pfizer, Inc. Ilse Blumentals, GlaxoSmithKline Nadine Ritter, Biologics Consulting Group, Inc. John Champagne, Wyatt Technology Corporation Reb Russell, Bristol-Myers Squibb Company John (JR) Dobbins, Eli Lilly and Company Mark Schenerman, MedImmune Elizabeth Fowler, CMC Consulting Services Timothy Schofield, Arlenda, Inc. Michelle Frazier-Jessen, Micromet, Inc. Sally Seaver, Seaver Associates LLC John Hennessey, NovaDigm Therapeutics, Inc. Joseph Siemiatkoski, Percivia, LLC Chulani Karunatilake, Nektar Therapeutics Arne Staby, Novo Nordisk A/S Jon Kauffman, Lancaster Laboratories, Inc. Jeffrey Staecker, PPD, Inc. Robert Kuhn, Amgen Inc. Garry Takle, Merck, Sharp and Dohme Joseph Kutza, MedImmune Ziping Wei, MedImmune Anthony Lubiniecki, Janssen R & D, LLC Andrew Weiskopf, Biogen Idec Robert Mattaliano, Genzyme, A Sanofi Company Yuan Xu, Novartis Lori McNamara, Medtronic

Collaborating Organizations American Association of Pharmaceutical Scientists (AAPS)

Fédération of Internationale Pharmaceutique (FIP)

American Chemical Society, Division of Analytical Chemistry (ACS, DAC)

International Association for Biologicals (IABs)

Association of Biomolecular Resource Facilities (ABRF)

National Institute for Biological Standards and Control (NIBSC)

Central New England Chromatography Council (CNECC)

PDA, West Coast Chapter

4

Acknowledgements Audio-Visual: Michael Johnstone, MJ Audio-Visual Productions

CASSS Staff: Karen A. Bertani, CMP, Symposium Manager Renee Olson, Manager of Meetings and Events Stephanie L. Flores, CAE, Executive Director Mikaela Sanford, Administrative Coordinator Linda Mansouria, CMP, CMM, Manager of Meetings and Events

Catherine Stewart,

5

General Information Name Badges Please wear your name badge in order to gain admittance to the meetings, poster sessions, exhibit hall and social functions. A name badge with a red circle represents a one-day registration, which does not include the cost of the Welcome Reception for the “Luck be a Lady” Floating Casino Night. Tickets to the Welcome Reception may be purchased at the registration desk at a cost of $95.00. Business Center The Intercontinental San Francisco Hotel has a Business Center that is located on the Lobby Level. The Business Center is open 24 hours a day, seven days a week and is accessible with your room key. Photographic Equipment The use of cameras is not permitted during the lecture program, workshops or poster sessions. Cameras are permitted on the exhibit floor. However, permission from the vendors involved must be obtained before photographs can be taken. Poster Sessions All posters will be set up the entire three days of the Symposium in the Grand Ballroom Foyer on the Third Floor and the Pacific Terrace Foyer on the Fourth Floor. Posters in the P-100-M series will be presented on Monday from 15:45 – 16:45. Posters in the P-200-T series will be presented on 14:45 – 15:45. Poster abstracts can be found beginning on page XXX in this book. Registration Registration will be set up in the Marina Room on the Third Floor from Monday, January 23 through Wednesday, January 25. Social Program Monday, January 23, 2012 12:30 – 14:15 Hosted Lunch Grand Ballroom Foyer

Pacific Terrace – Fourth Floor 18:30 – 22:30 Welcome Reception – “Luck be a Lady” Floating Casino Night NOTE: Transportation will be provided. You should be on the bus no later than 6:30 pm. Tuesday, January 24, 2012 11:45 – 13:30 Hosted Lunch Grand Ballroom Foyer

Pacific Terrace – Fourth Floor 18:15 – 19:30 Exhibit and Poster Reception Grand Ballroom Foyer – Third Floor Pacific Terrace – Fourth Floor If you have special dietary needs, please notify the Symposium Manager Monday morning so that an alternate menu may be prepared. Also, please note that tickets to the Welcome Reception for non-registered guests or one-day registrants may be purchased at the Symposium registration desk for $95.00. Technical Seminars Two technical seminars will be held on Monday, January 23 from 13:00 – 14:00 (lunch is provided). Two technical seminars will be held on Tuesday, January 24 from 12:15 – 13:15 (lunch is provided). Three technical seminars will be held on Tuesday, January 24 from 17:15 – 18:15. Please refer to the program summary for more details. Technical Seminar abstracts may be found beginning on page XXX.

6

Meeting Room Layout

07

:00

07

:00

07

:15

07

:15

07

:30

07

:30

07

:45

07

:45

08

:00

08

:00

08

:15

08

:15

08

:30

08

:30

08

:45

08

:45

09

:00

09

:00

09

:15

09

:15

09

:30

09

:30

09

:45

09

:45

10

:00

10

:00

10

:15

10

:15

10

:30

10

:30

10

:45

Ba

llro

om

AB

allro

om

BB

allro

om

CU

nio

n S

q R

m1

0:4

5

11

:00

Ad

ve

ntitio

us

CM

C S

um

ma

ry:

Ste

m C

ells,

11

:00

11

:15

Ag

en

tS

imila

rity

Se

qu

en

ce

Ge

ne

& C

ell

Ba

llro

om

AB

allro

om

BB

allro

om

CU

nio

n S

q R

m1

1:1

5

11

:30

Te

stin

go

f B

iolo

gic

sV

ari

an

tsT

he

rap

yC

MC

Su

mm

ary

New

an

dQ

bD

fo

rV

accin

e1

1:3

0

11

:45

Rap

id P

rod

uct

Em

erg

ing

Ea

rly

Qu

ality

&1

1:4

5

12

:00

Deve

lop

me

nt

Te

ch

no

log

ies

Deve

lop

me

nt

Su

pp

ly1

2:0

0

12

:15

Ba

llro

om

AB

allro

om

C1

2:1

5

12

:30

Ag

ile

nt

Bru

ke

r1

2:3

0

12

:45

Te

ch

no

log

ies

Dalt

ron

ics

12

:45

13

:00

Ba

llro

om

AB

allro

om

Clu

nch

fo

r 1

00

lun

ch

fo

r 1

00

13

:00

13

:15

Wa

ters

Pro

tein

-1

3:1

5

13

:30

Co

rpo

rati

on

Sim

ple

13

:30

13

:45

lun

ch

fo

r 1

00

lun

ch

fo

r 1

00

13

:45

14

:00

14

:00

14

:15

Ba

llro

om

AB

allro

om

BB

allro

om

CU

nio

n S

q R

m1

4:1

5

14

:30

Assa

yV

isib

le1

4:3

0

14

:45

Life

cycle

Com

pa

rab

ility

Refe

ren

ce

Ap

pe

ara

nce

14

:45

15

:00

Ma

na

ge

me

nt

Sta

nd

ard

sT

estin

g1

5:0

0

15

:15

15

:15

15

:30

15

:30

15

:45

15

:45

16

:00

16

:00

16

:15

16

:15

16

:30

16

:30

16

:45

16

:45

17

:00

Ba

llro

om

AB

allro

om

BB

all

roo

m C

Un

ion

Sq

Rm

17

:00

17

:15

Sp

ecific

atio

ns

New

An

aly

tica

lD

elive

ry D

evic

es

17

:15

17

:30

an

d C

on

tro

lT

ech

no

log

yA

ntib

od

y D

rug

as C

om

bin

atio

nB

allro

om

AB

allro

om

BB

allro

om

C1

7:3

0

17

:45

Str

ate

gie

sIm

ple

me

nta

tio

nC

on

jug

ate

sP

rod

ucts

Bec

km

an

Cata

len

tW

ya

tt1

7:4

5

18

:00

Co

ult

er

Ph

arm

So

lns

Te

ch

no

log

ies

18

:00

18

:15

18

:15

18

:30

18

:30

18

:45

18

:45

19

:00

19

:00

19

:15

19

:15

19

:30

19

:30

Co

nti

ne

nta

l B

rea

kfa

st

(8:2

0 -

8:3

0)

M

ini-

Bre

ak

(8:3

0 -

8:4

0)

A

nn

ou

nc

em

en

ts

PM

Bre

ak

(1

5:1

5 -

15

:45

)

Wo

rks

ho

p S

ess

ion

3

(1

1:0

0 -

12:1

5)

Wo

rks

ho

p S

ess

ion

4

(1

4:0

0 -

15:1

5)

Wo

rks

ho

p S

ess

ion

2

(1

0:3

0 -

11:4

5)

(8:0

0 -

9:0

0)

(14

:45

- 1

5:4

5)

Po

ste

r S

es

sio

n-2

(8:4

0 -

10

:00

)

Co

nti

ne

nta

l B

rea

kfa

st

AM

B

rea

k

(10

:30

- 1

1:0

0)

(12

:15

- 1

3:1

5)

7:1

5 -

8:1

5

C

on

tin

en

tal

Bre

ak

fas

tP

acific

Te

rra

ce

-

4

th F

loo

r

8

:15

C

AS

SS

We

lco

me

an

d I

ntr

od

uc

tory

Co

mm

en

ts

AM

B

rea

k

(10

:45

- 1

1:1

5)

Su

nri

se

Se

ss

ion

-

G

ene

an

d C

ell

Th

era

py C

om

ple

x

Bio

logic

al P

roducts

(9:0

0 -

9:1

0)

A

nn

ou

nc

em

en

ts

Pa

cific

Te

rra

ce

-

4

th F

loo

r

Assa

sy L

ife

cycle

Ma

na

ge

me

nt

-

(15

:45

- 1

7:0

0)

Pa

cific

Te

rra

ce

Fo

ye

r -

4

th F

loo

r

It's

no

t a

Mis

take

- I

t's a

De

sig

n F

ea

ture

Ple

na

ry S

es

sio

n:

5

(12

:15

- 1

4:0

0)

(13

:30

- 1

4:4

5)

Gra

nd

Ba

llro

om

Fo

ye

r -

3

rd F

loo

r

Min

i-B

rea

k

(1

7:0

0 -

17

:15

)

Wo

rks

ho

p S

ess

ion

1

(1

6:4

5 -

18:0

0)

Ple

na

ry S

es

sio

n:

7

Se

qu

en

ce

Va

ria

nts

- W

ha

t, W

hy,

Wh

en

, H

ow

Gra

nd

Ba

llro

om

Fo

ye

r -

3

rd F

loo

r

Lu

nc

h B

rea

k

WC

BP

20

12

-

PR

OG

RA

M O

VE

RV

IEW

We

dn

es

da

y

Clo

sin

g R

em

ark

s

(17

:00

- 1

7:1

5)

Ple

na

ry S

es

sio

n:

1

Ple

na

ry S

es

sio

n:

2

(11

:15

- 1

2:3

0)

(8:4

5 -

10

:45

)

8

:35

W

CB

P W

elc

om

e a

nd

In

tro

du

cto

ry C

om

me

nts

(14

:15

- 1

5:4

5)

Tu

es

da

y

AM

B

rea

k

(10

:00

- 1

0:3

0)

Pa

cific

Te

rra

ce

(7:0

0 -

8:4

5)

(7:1

5 -

8:2

0) Ple

na

ry S

es

sio

n:

4

an

d t

he

n W

ha

t?

4th

Flo

or

An

tib

od

y D

rug

Co

nju

ga

tes

(9:1

0 -

9:3

0)

A

ck

no

wle

dg

em

en

ts

(9:3

0 -

10

:30

)

Mo

nd

ay

8

:25

1

st

An

nu

al

Wil

lia

m H

an

co

ck

Aw

ard

Re

gu

latio

n o

f B

iop

ha

rma

ce

utica

l P

rod

ucts

-

Go

ve

rnm

en

t &

In

du

str

y

Pe

rsp

ective

s

(15

:45

- 1

6:4

5)

Po

ste

r S

es

sio

n-1

Ple

na

ry S

es

sio

n:

3

Co

mm

erc

ializ

atio

n o

f C

om

bin

atio

n P

rod

ucts

me

et

in t

he

lo

bb

y -

tra

nsp

ort

atio

n p

rovid

ed

We

lco

me

Re

ce

pti

on

(18

:30

- 2

2:3

0)

(12

:30

- 1

4:1

5)

Pe

rsp

ective

s o

n t

he

De

ve

lop

me

nt

an

d

Pa

cific

Te

rra

ce

Fo

ye

r -

4

th F

loo

r

Ex

hib

it &

Po

ste

r R

ec

ep

tio

n

Te

ch

nic

al

Se

min

ars

(

17

:15

- 1

8:1

5)

Tra

ns

itio

n

(1

8:0

0 -

18

:30

)

4th

Flo

or

Te

ch

nic

al

Se

min

ars

(13

:00

- 1

4:0

0)

Ad

va

nce

s in

th

e D

eve

lop

me

nt

an

d C

ha

racte

riza

tio

n

of

Va

ccin

es

Pa

cific

Te

rra

ce

Fo

ye

r -

4

th F

loo

r

Gra

nd

Ba

llro

om

Fo

ye

r -

3

rd F

loo

r

(18

:15

- 1

9:3

0)

Be

st

Pra

ctice

s &

Em

erg

ing

Dire

ctio

ns

(11

:45

- 1

3:3

0)

Ho

ste

d L

un

ch

Pa

cific

Te

rra

ce

Te

ch

nic

al

Se

min

ars

4th

Flo

or

No

t Ju

st

Be

ca

use

We

Ca

n

Ple

na

ry S

es

sio

n:

6

Ne

w a

nd

Em

erg

ing

Te

ch

no

log

ies:

Pa

cific

Te

rra

ce

Ho

ste

d L

un

ch

Re

fere

nce

s /

Sta

nd

ard

s:

(15

:45

- 1

7:0

0)

Ple

na

ry S

es

sio

n:

8

8

WCBP 2012 Preliminary Scientific Program Summary

MONDAY, JANUARY 23, 2012 07:15 – 17:30 Registration in the Marina Room 07:15 – 08:15 Continental Breakfast in the Pacific Terrace – Fourth Floor 08:15 – 08:25 CASSS Welcome and Introductory Comments in the Grand Ballroom John Dougherty, Eli Lilly and Company 08:25 – 08:35 1st Annual William S. Hancock Award Announcement in the Grand Ballroom Sponsored by CASSS Wassim Nashabeh, Genentech, a Member of the Roche Group 08:35 – 08:45 WCPB 2012 Welcome and Introductory Comments in the Grand Ballroom

Peter Johnson, 3M Drug Delivery Systems 08:45 – 10:45

Regulation of Biopharmaceutical Products: Government and Industry Perspectives Fundamental Questions regarding Regulation of Biotherapeutics and Novel Therapeutic

Approaches Roundtable Discussion in the Grand Ballroom Session Chair: Mark Schenerman, MedImmune

Facilitator: Kenneth Seamon, University of Cambridge

Panel Members: Laura Gomes Castanheira, National Health Surveillance Agency (ANVISA), Brasilia, Brasil Christopher Joneckis, CBER, FDA, Rockville, MD USA Steven Kozlowski, CDER, FDA, Bethesda, MD USA Wassim Nashabeh, Genentech, a Member of the Roche Group Peter Richardson, European Medicines Agency, London, United Kingdom Anthony Ridgway, Health Canada, Ottawa, Canada

10:45 – 11:15 AM Break – Visit the Exhibits in the Pacific Terrace – Fourth Floor

Antibody Drug Conjugates: Balancing Large and Small Molecule Control Strategies from Development to Commercialization

Plenary Session in the Grand Ballroom Session Chairs: Laura Bass, Pfizer, Inc. and Lotte K. McNamara, LKM CMC Consulting

11:15 – 11:35 Regulation of Antibody-Drug Conjugates: 1993 - 2012 and Beyond Xiao-Hong Chen, CDER, FDA, Bethesda, MD USA Marjorie Shapiro, CDER, FDA, Bethesda, MD USA

9

Monday, January 23 continued… 11:35 – 11:55 Control Strategies for Small Molecule Components of Antibody-Drug

Conjugates Nathan Ihle, Seattle Genetics, Inc., Seattle, WA USA 11:55 – 12:15 Enhanced Analytical Control Strategy and Product/Process Characterization

for Inotuzumab Ozogamicin April Xu, Pfizer, Inc., Pearl River, NY USA 12:15 – 12:30 Questions and Answers 12:30 – 14:15 Hosted Lunch Break in the Pacific Terrace – Fourth Floor 13:00 – 14:00 Technical Seminars Simplifying Glycoprotein Characterization with Released Glycan UPLC Analysis and a Customized Biotherapeutic Glycan Database Sponsored by Waters Corporation Grand Ballroom A NOTE: Lunch is provided for first 100 attendees in Grand Ballroom A Protein Particles and Purity – Now Simple, Now Automated Sponsored by ProteinSimple Grand Ballroom C NOTE: Lunch is provided for first 100 attendees in Grand Ballroom C

Perspectives on the Development and Commercialization of Combination Products Plenary Session in the Grand Ballroom

Session Chairs: John (JR) Dobbins, Eli Lilly and Company and Edwin Moore, Baxter Healthcare Corporation

14:15 – 14:35 Regulatory Challenges for Pre-Filled Syringes and Other Medical Devices Kathy Lee, CDER, FDA, Bethesda, MD USA 14:35 – 14:55 Regulatory Requirements and Functionality Tests for Pre-filled Syringes and

Autoinjectors Doug Mead, Janssen Research & Development, LLC, Malvern, PA USA 14:55 – 15:15 Stability Evaluation and Delivery Device Considerations to Support a

Container Closure Change for a Fusion Protein Angela Blake-Haskins, Human Genome Sciences, Inc., Rockville, MD USA 15:15 – 15:45 Questions and Answers 15:45 – 16:45 Poster Session One in the Grand Ballroom Foyer – Third Floor and the Pacific Terrace Foyer – Fourth Floor

10

Monday, January 23 continued… 16:45 – 18:00 Workshop Session 1 Specifications for Biotech Products: Still an Upward Journey? Grand Ballroom A John Dougherty, Eli Lilly and Company; Sarah Kennett, CDER, FDA; Philip Krause, CBER, FDA; Anthony Mire-Sluis, Amgen Inc. To Boldly Go Where QC Has Never Gone Before...Tales from the Frontier of New Analytical Technology Implementation Grand Ballroom B Lokesh Bhattacharyya, CBER, FDA; Drew Kelner, Amgen Inc.; Jackie Larew, Eli Lilly and Company; Joao Pedras-Vasconcelos, CDER, FDA Antibody Drug Conjugates Grand Ballroom C Kevin Anderson, Seattle Genetics, Inc.; Kathleen Clouse, CDER, FDA; Jonathan Harris, Genentech, a Member of the Roche Group Drug Delivery Devices as Combination Products – Understanding the Additional Challenges: Is It Worth All the Added Effort? Union Square Room Julia Ding, PPD, Inc.; Suzanne Kiani, MedImmune; Maria Lugo Gutierrez, CDER, FDA; Deborah Trout, CBER, FDA 18:30 – 22:30 Welcome Reception at the “Luck be a Lady” Floating Casino Night Transportation will be provided – meet in the lobby. You should be on the bus no later than 6:30 pm.

11

TUESDAY, JANUARY 24, 2012

07:00 – 17:00 Registration in the Marina Room 07:00 – 08:45 Continental Breakfast in the Pacific Terrace – Fourth Floor

Cell and Gene Therapy Complex Biological Products – What Will It Take to Make Them Well-characterized?

Sunrise Plenary Session in the Grand Ballroom Session Chair: Steven Bauer, CBER, FDA and Yuan Xu, Novartis

07:15 – 07:40 Stem Cell Discoveries in Translation and Clinical Trials

Alan Trounson, California Institute for Regenerative Medicine (CIRM), San Francisco, CA USA

07:40 – 08:05 Implementing the Principles of Quality by Design for Early Stage Gene

Therapy Products Michael Kelly, Genzyme, A Sanofi Company, Framingham, MA USA 08:05 – 08:20 Questions and Answers 08:20 – 08:30 Mini-Break 08:30 – 08:40 Announcements by Peter Johnson, 3M Drug Delivery Systems

Sequence Variants: What, Why, When, How and Then What? Plenary Session in the Grand Ballroom

Session Chairs: Anthony Lubiniecki, Janssen Research & Development, LLC and Arne Staby, Novo Nordisk A/S

08:40 – 08:45 History of Sequence Variants

Anthony Lubiniecki, Janssen Research & Development, LLC, Radnor, PA USA 08:45 – 09:05 Detection and Mitigation of a Sequence Variant in a Recombinant Protein

Expressed in Mammalian Cells Haimanti Dorai, Janssen Research &Development, LLC, Radnor, PA USA 09:05 – 09:25 Analysis and Characterisation of Sequence Variants in Recombinant

Haemostasis Products Anne Kroll Kristensen, Novo Nordisk A/S, Måløv, Denmark 09:25 – 09:45 Detection of Sequence Variants in Recombinant Proteins: Sensitivity, Timing

and Evaluation of Risks Kathleen Francissen, Genentech, a Member of the Roche Group, South San

Francisco, CA USA

12

Tuesday, January 24 continued… 09:45 – 10:00 Questions and Answers 10:00 – 10:30 AM Break – Visit the Exhibits in the Pacific Terrace – Fourth Floor 10:30 – 11:45 Workshop Session 2 Testing Quality Requires Quality Testing: Managing the Introduction of New Technologies for Adventitious Agent Testing Grand Ballroom A Arifa Khan, CBER, FDA; Richard Ledwidge, CDER, FDA; Mark Plavsic, Genzyme, A Sanofi Company; Timothy Schofield, Arlenda, Inc. Achieving and Interpreting Similarity of Biologics: Implications for Product Development Strategies and Product Claims Grand Ballroom B Christopher Joneckis, CBER, FDA; Peter Richardson, European Medicines Agency; Nadine Ritter, Biologics Consulting Group, Inc.; Emily Shacter, CDER, FDA Sequence Variants: The How, What, When, Why and What Now? Grand Ballroom C Ralph Bernstein, CDER, FDA; John Bishop, CBER, FDA; Reed Harris, Genentech, a Member of the Roche Group; Ziping Wei, MedImmune Linking Product Quality Attributes to Safety and Efficacy of Cell and Gene Therapy Products Union Square Room Pakwai (Patrick) Au, CBER, FDA; Steven Bauer, CBER, FDA; Joann Parker, Pfizer, Inc.; Darin Weber, Mesoblast 11:45 – 13:30 Hosted Lunch Break in the Pacific Terrace – Fourth Floor 12:15 – 13:15 Technical Seminars Accelerate mAb Characterization Using Automated Sample Prep Sponsored by Agilent Technologies Grand Ballroom A NOTE: Lunch is provided for first 100 attendees in Grand Ballroom A So You Think You Know Me?...Using Mass Spectrometry to Understand Your Biotherapeutic Sponsored by Bruker Daltonics, Inc. Grand Ballroom C NOTE: Lunch is provided for first 100 attendees in Grand Ballroom C

Advances in the Development and Characterization of Vaccines Plenary Session in the Grand Ballroom

Session Chairs: Arifa Khan, CBER, FDA; Stefanie Pluschkell, Pfizer, Inc.; Garry Takle, Merck 13:30 – 13:50 A-Vax—QbD Vaccine Case Study

Mark Schenerman, MedImmune, Gaithersburg, MD USA

13

Tuesday, January 24 continued… 13:50 – 14:10 Life Cycle Management of Vaccine Products: Adapting Production Processes

to Better Meet Tomorrow's Global Needs Colette Ranucci, Merck & Co., Inc., West Point, PA USA

14:10 – 14:30 Challenges for the Production of a Well-characterized Vaccine Robin Levis, CBER, FDA, Rockville, MD USA 14:30 – 14:45 Questions and Answers 14:45 – 15:45 Poster Session Two in the Grand Ballroom Foyer – Third Floor and the Pacific Terrace Foyer – Fourth Floor

New & Emerging Technologies: Not Just Because We Can Plenary Session in the Grand Ballroom

Session Chairs: Roman Drews, CBER, FDA; John Hennessey, NovaDigm Therapeutics, Inc.; Susan Kirshner, CDER, FDA

15:45 – 16:05 GlycoPEGylation – A Robust Technology for Modification of Coagulation

Factors Are Bogsnes, Novo Nordisk A/S, Gentofte, Denmark 16:05 – 16:25 High Throughput Biophysical Approaches to Preformulation

Characterization and Comparability Analysis of Protein Drugs and Macromolecular Vaccines David Volkin, Department of Pharmaceutical Chemistry, University of Kansas, Lawrence, KS USA

16:25 – 16:45 Applications of State-of-the-Art Glycan Analysis to Biopharmaceuticals

Pauline Rudd, National Institute for Bioprocessing Research & Training (NIBRT), Dublin, Ireland

16:45 – 17:00 Questions and Answers 17:00 – 17:15 Mini-Break

17:15 – 18:15 Technical Seminars Aggregate and Particle Analysis in Biologics Production Assessing Objects of All Sizes with Physical Techniques Sponsored by Beckman Coulter Grand Ballroom A Virological Safety of Biopharmaceuticals: Regulatory Considerations and Pragmatic Approaches Sponsored by Catalent Pharma Solutions Grand Ballroom B Absolute Macromolecular Characterization of Proteins Using Light Scattering and Related Techniques Sponsored by Wyatt Technology Corporation Grand Ballroom C 18:15 – 19:30 Exhibit and Poster Reception Grand Ballroom Foyer – Third Floor and the Pacific Terrace Foyer – Fourth Floor

14

WEDNESDAY, JANUARY 25, 2012 08:00 – 17:00 Registration in the Marina Room 08:00 – 09:00 Continental Breakfast in the Pacific Terrace – Fourth Floor 09:00 – 09:10 Announcements in the Grand Ballroom

Peter Johnson, 3M Drug Delivery Systems 09:10 – 09:30 Acknowledgments in the Grand Ballroom

John Dougherty, Eli Lilly and Company

Assay Lifecycle Management: It’s Not a Mistake, It’s a Design Feature Plenary Session in the Grand Ballroom

Session Chairs: Marjorie Shapiro, CDER, FDA and Robert Sitrin, Sitrin Solutions 09:30 – 09:50 Management of the Analytical Method Life Cycle for Biotechnology

Products: A Regulatory Perspective Rashmi Rawat, CDER, FDA, Bethesda, MD USA 09:50 – 10:10 Analytical Lifecycle for Marketed Protein Therapeutics

John Stults, Genentech, a Member of the Roche Group, South San Francisco, CA USA

10:15 – 10:30 Questions and Answers 10:30 – 11:00 AM Break – Visit the Exhibits in the Pacific Terrace – Fourth Floor 11:00 – 12:15 Workshop Session 3 Rapid Pharmaceutical Product Development: Getting Off the Critical Path and onto the Right Path Grand Ballroom A Doug Cecchini, Biogen Idec; Roman Drews, CBER, FDA; Ilona Reischl, AGES PharmMed; Barbara Rellahan, CDER, FDA; James Stonecypher, Micromet, Inc. New and Emerging Technologies – Timing is Everything? Grand Ballroom B Lori McNamara, Medtronic; Ashutosh Rao, CDER, FDA; Joseph Siemiatkoski, Percivia, LLC Quality by Design for Early Phase Development Grand Ballroom C Audrey Jia, CDER, FDA; Elisabeth Kirchisner, F. Hoffmann – La Roche Ltd.; Laurie Norwood, CBER, FDA; Victor Vinci, Eli Lilly and Company Impact of Improved Scientific & Technical Understanding on Vaccine Quality and Supply Union Square Room Collette Ranucci, Merck Manufacturing Division; Steven Rubin, CBER, FDA; Michael Washabaugh, MedImmune

15

Wednesday, January 25 continued…

12:15 – 14:00 Lunch Break – Participants on their own 14:00 – 15:15 Workshop Session 4 Assay Lifecycle Management: How, What, When and Why? Grand Ballroom A Alfred Del-Grosso, CBER, FDA; David Good, Eli Lilly and Company; Joseph Molon, Biogen Idec; Jennifer Swisher, CDER, FDA Comparability Evaluations of Well-Characterized Biologics Grand Ballroom B Sid Advant, ImClone Systems Corporation; Karen Lee, Genzyme, A Sanofi Company; Robin Levis, CBER, FDA; Lixin Xu, CDER, FDA Spotlight on Reference Standards - What is Required to Become “The Standard”? Grand Ballroom C Markus Blümel, Novartis Pharma AG; Barry Cherney, CDER, FDA; Rajesh Gupta, CBER, FDA; Dieter Schmalzing, Genentech, a Member of the Roche Group Sorting Out the Best Methods and Techniques for Visible Appearance Testing Union Square Room Pat Cash, MedImmune; Erwin Freund, Amgen Inc.; Laurie Graham, CDER, FDA; Deborah Trout, CBER, FDA 15:15 – 15:45 PM Break – Visit the Exhibits in the Pacific Terrace – Fourth Floor

References/Standards: Best Practices and Emerging Directions Plenary Session in the Grand Ballroom

Session Chairs: Brian K. Nunnally, Pfizer, Inc. and Sally Seaver, Seaver Associates LLC 15:45 – 16:05 Reference Standards for Monoclonal Antibodies: Key Challenges Addressed Matthew Borer, Eli Lilly and Company, Indianapolis, IN USA 16:05 – 16:25 Standardization of Optical Particle Counters

Dean Ripple, National Institute of Standards and Technology (NIST), Gaithersburg, MD USA

16:25 – 16:45 Novel Standards for Novel Applications Chris Jones, Laboratory for Molecular Structure, NIBSC, Hertsfordshire, United

Kingdom 16:45 – 17:00 Questions and Answers 17:00 – 17:15 Closing Remarks by Kathleen Francissen, Genentech a Member of the Roche

Group

16

Plenary Session Abstracts

Regulation of Biopharmaceutical Products: Government and Industry Perspectives Fundamental Questions regarding Regulation of Biotherapeutics and Novel Therapeutic Approaches Session Chair: Mark Schenerman, MedImmune Facilitator: Kenneth Seamon, University of Cambridge Panel Members: Laura Gomes Castanheira, National Health Surveillance Agency (ANVISA), Brasilia, Brasil Christopher Joneckis, CBER, FDA, Rockville, MD USA Steven Kozlowski, CDER, FDA, Bethesda, MD USA Wassim Nashabeh, Genentech, a Member of the Roche Group Peter Richardson, European Medicines Agency, London, United Kingdom Anthony Ridgway, Health Canada, Ottawa, Canada There has been a significant investment in the understanding of biotherapeutics and their manufacturing over the years. This investment has led to the discovery of novel approaches for prevention, detection, and treatment of many severe and life threatening diseases. Investment in product and manufacturing science has lead to greater process efficiencies and more rational and controlled systems. However, there continues to be significant pressure on all stakeholders to do more to harness the potential of these investments. In addition, these may provide an opportunity to enhance access and facilitate global health goals in developing countries. Supporting innovation is a cornerstone of many strategic endeavors by both industry and government. Reaping the benefits of innovation usually requires changes in business practices or different approaches for development and regulation. This applies for well-characterized drugs and biologics as well as other complex biopharmaceuticals. For example, it was necessary to reevaluate cell substrate issues for rDNA products and how product safety could be assured from an inherently unsafe cell substrate. In a similar manner, it will be necessary to reevaluate the basis of safety for cell therapy products in terms of cell substrate safety, viral safety (and how to achieve it in terms of procedural and environmental controls and confirmation by testing). The understanding of manufacturing controls and regulation is also evolving with the development of QbD via many different industry and regulatory initiatives. The enhanced understanding of products and processes has also led to the evolution of frameworks for the development and approval of Biosimilars. The discussion today will focus on three areas of innovation that can be viewed as examples to stimulate discussion for how best to leverage these investments in biotechnology across a globalized industry. We will ask for a brief statement (5 minutes) about the year’s strategic priorities from the panel participants and then will focus the discussion on the following topics:

1. Globalization of Development and Manufacturing 2. Biosimilars and Determination of Equivalence 3. Harvesting the Benefits of Enhanced Production and Process Understanding through QbD

NOTES:

17

NOTES:

18

Antibody Drug Conjugates: Balancing Large and Small Molecule Control Strategies from Development to Commercialization Antibody-Drug Conjugates (ADCs) are monoclonal antibodies conjugated to small cytotoxic molecules using chemical linkers. ADCs are complex molecules with multiple sources of heterogeneity derived from the incoming mAb (size, charge, glycan variants) and small molecule (charge, chemical variants) components as well as from the conjugation reaction (drug distribution, sites of attachment, conjugation side-products). Although traditional analytical methodologies can be utilized for the intermediates (mAb and small molecule), modification of these technologies are often required to ensure sufficient control of the final drug conjugate in terms of detecting changes in both the ADC and the cytotoxin, linker and/or mAb in the presence of the ADC. To develop an appropriate control strategy for the ADC and its intermediates it is important to identify the ADC quality attributes that impact safety and efficacy. The ADC quality attributes will dictate the mAb and cytotoxin quality attributes and thus the control strategy for these intermediates. The testing strategy and associated specifications for the ADC are thus defined by the combination of control strategies which in addition to release and stability testing of the ADC includes testing of raw materials and intermediates (monoclonal antibody and drug / drug linker molecules) as well as identification of appropriate process and procedural controls to ensure delivery of a consistent, safe and efficacious product that meets regulatory expectations. This plenary will focus on industry approaches to and regulatory perspectives on ADC control strategies and how these strategies have evolved during the past decade from the approval of Mylotarg to the latest ADC BLA filing in 2011. In addition, an update of the regulatory review process for ADC will be provided. NOTES:

19

Regulation of Antibody-Drug Conjugates: 1993 – 2011 and beyond Marjorie Shapiro and Xiao-Hong Chen CDER, FDA, Bethesda, MD USA It has been almost 20 years since the first IND for an antibody-drug conjugate (ADC) was submitted to the FDA. Since that time, two products have made it to the market: Mylotarg was approved in 2000 following the NDA pathway and Adcetris was approved in 2011 following the BLA pathway. In addition to differences in the regulatory pathway for these products, improvements in the conjugation chemistry and enhanced knowledge regarding different IgG isotypes enable application of a rational design to ADCs. Furthermore, improved or new analytical methods have been developed that allow a more precise characterization of ADCs. These advancements have facilitated our regulatory decision making. We will discuss the regulatory approach to ADCs from the small drug molecule and biologics reviewer perspectives. NOTES:

20

Control Strategies for Small Molecule Components of Antibody-Drug Conjugates Nathan Ihle Seattle Genetics, Inc., Seattle, WA USA Antibody-Drug Conjugates (ADCs) are monoclonal antibodies with cytotoxic molecules covalently attached. As such, control strategies for ADCs must address quality attributes of both the therapeutic proteins and small molecule drugs. In addition, some quality attributes are unique to ADCs. A coordinated approach that provides robust control across this diverse set of quality attributes is required. In this presentation, elements of the control strategy related to the small molecule components of ADCs will be discussed. This will include a discussion of representative attributes and factors that should be considered in the determination of appropriate points of control. Test methods and other controls applied before and after the covalent attachment of the antibody, drug and linker will be illustrated. Examples will be taken from the development of ADCETRIS™ (brentuximab vedotin), a CD30-directed ADC approved by FDA in 2011. NOTES:

21

Enhanced Analytical Control Strategy and Product/Process Characterization for Inotuzumab Ozogamicin April Xu1; Lawrence Chen1; Jim Mo2; Laura Bass2 1Pfizer BioTherapeutics Pharmaceutical Sciences, Pear River, NY USA; 2 Pfizer BioTherapeutics Pharmaceutical Sciences, St. Louis, MO USA Inotuzumab ozogamicin (CMC-544) is an antibody-drug conjugate indicated for CD22+ B-cell malignancies (ex. NHL, ALL). It is comprised of a human IgG4 anti-CD22 antibody linked to the cytotoxic agent Calicheamicin. One of the critical quality attributes for an antibody-drug conjugate is the drug load distribution. Obtaining this information has presented some technical challenges due to the specific structure of toxin and conjugation chemistry used for CMC-544. With the advancement of new analytical technologies, we have developed several methods for monitoring drug load distribution in CMC-544, such as HIC, iCE and LC/MS. The advantages and caveats of each method will be compared and discussed. In addition, data will be presented from the application of these methods to support conjugation process understanding. Other updates proposed for the CMC-544 analytical control strategy will also be discussed. NOTES:

22

Perspectives on the Development and Commercialization of Combination Products Pre-filled syringes (PFS) are a preferred delivery system for many biopharmaceuticals. Developing biopharmaceutical products in syringes present a number of challenges including the choice of container (including plastic versus glass), choice of closure, compatibility of the protein and formulation with the syringe materials, stability of the protein in the syringe, and integrity of the container and closure system. Evolution of analytical methods to differentiate and quantify protein aggregates from silicone particles presents new means of characterizing PFS products. In addition to the quality aspects related to the container closure functionality of the PFS, the classification of a PFS as a combination product brings to the forefront additional considerations such as the potential applicability of device standards, and the potential need for user and human factors studies. Furthermore, transitioning a product from a PFS presentation to an auto-injector during development and/or post-licensure presents challenges and could result in the need to conduct additional clinical studies. This plenary session will address the above challenges faced by industry and by regulatory authorities throughout the development of biopharmaceuticals combination products. NOTES:

23

Regulatory Challenges for Pre-filled Syringes and Other Medical Devices Kathy Lee CDER, FDA, Bethesda, MD USA Today more and more products are being developed in prefilled syringes for ease of use for the patients and accurate dosing. Prefilled syringes are regulated by the Food and Drug Administration as combination products, either as drug/device or as biologic/device combination. Because there are no specific manufacturing regulations for combinations products, the expectation is that the manufacturer will follow both the biologic/drug regulations and the device regulations for CMC. This talk will focus on the Agency’s current and evolving thinking on regulatory CMC expectations for prefilled syringes. NOTES:

24

Regulatory Requirements and Functionality Tests for Pre-filled Syringes and Autoinjectors Doug Mead Janssen Research & Development, LLC, Malvern, PA USA Biopharmaceuticals formulated in prefilled syringes (PFS) and autoinjectors for subcutaneous administration by patients and healthcare professionals are considered to be drug-device combination products. While this may be a regulatory definition, there is no question that these products have device characteristics and are used as devices to perform the injection. Biopharmaceutical product development plans should, therefore, include device development and testing elements as well as usability assessments. Indeed, ICH and FDA drug guidances have always required that the “suitability” of devices for administration be fully characterized and reported in regulatory filings. However, FDA has begun to expect that Design Controls, based on medical device regulations, should be used as a process to establish specifications, assess functionality and suitability, and to validate that user needs are met. Functionally of a PFS is often reported in terms of piston plunger break loose and travel forces and these parameters are highly dependent on protein solution viscosity, needle gauge, syringe barrel silicone quantity and distribution, and protein aggregation of the formulation on the syringe barrel over time. Measured force values are also dependent on the piston plunger velocity selected for the test method but, ultimately, PFS functionality in actual use is highly user dependent. Users can, within limits, simply choose to inject slower or faster to accommodate the forces encountered. Likewise, PFS functionality impacts autoinjector performance but piston plunger velocity, under a fixed spring force, and the potential for stiction are also of interest. Finally, other autoinjector functionality characteristics such as actuation forces and needle safety features will need to be assessed. This presentation will explore the functionality aspects of PFS and autoinjector performance and what regulators may be looking for in the characterization of functionality, the validation of delivery device designs in human factors studies, and the assessment of usability in pivotal clinical trials. Because delivery device designs can and should evolve over a biopharmaceutical’s commercial life, strategies for approval of post-licensure combination product design changes will also be presented. NOTES:

25

Stability Evaluation and Delivery Device Considerations to Support a Container Closure Change for a Fusion Protein Angela Blake-Haskins Human Genome Sciences, Inc., Rockville, MD USA Primary container closure changes (eg, vial to prefilled syringe) are sometimes considered for protein therapeutics to enhance patient convenience when administering the drug. Integration of prefilled syringes and cartridges with drug delivery systems are also common and consideration of delivery device factors on both device performance and protein stability is prudent. This presentation describes stability and delivery device characterization studies performed to support a change in the final container closure for albinterferon alfa-2b, a genetic fusion protein consisting of human serum albumin (HSA) and interferon alpha-2b (rIFNa-2b). The effects of silicone oil, needle size, transportation stress (vibration, reduced pressure, etc.), and patient in-use conditions on device performance and protein stability were investigated. Injection forces were characterized using a compression gauge. Protein stability was assessed using HPLC and biophysical methods capable of detecting differences in chemical and physical stability and higher order structure. Subvisible particle levels were also investigated using flow imaging microscopy and photon correlation spectroscopy. An increase in applied silicone oil resulted in a decrease in injection forces. Stability data indicated that high silicone oil levels in combination with intended and stress storage conditions had no impact on albinterferon alfa-2b product quality. Solution turbidity and subvisible particle levels were observed to increase in both placebo and protein containing product due to increases in solution silicone oil level and not protein aggregation. Needle shear had no effect on albinterferon alfa-2b product quality, thus, warranting needle size selection based on injection force studies. Albinterferon alfa-2b was shown to be robust to mechanical stress conditions expected during routine transportation. Data from reduced pressure studies showing plunger movement was utilized in the design of the delivery device and assembly process. Potential device use errors identified through a risk assessment and determined to impact albinterferon alfa-2b product quality through laboratory simulation were also considered in the design of the delivery system. Characterization studies evaluating the effects of delivery device parameters on device performance and protein stability were instrumental in supporting a change in container closure for albinterferon alfa-2b fusion protein. NOTES:

26

Cell and Gene Therapy Complex Biological Products – What Will It Take to Make Them Well-characterized? Cell and gene therapy products are frequently classified as “complex” based on their methods of manufacture, their intricate molecular composition and structure, and the challenge of relating their analytical characteristics to their functional properties. Many chemical, physical, and biological properties can be measured but thorough structural characterization is limited due to their complexity. Incomplete structural characterization means there is a lack of comprehensive product and process knowledge, so the challenge becomes a need to identify and clearly link specific quality attributes with clinical performance. Biological/ functional assays may provide links to clinical performance, but these assays are often challenging to implement as validatable in-process or lot release assays due to factors such as high inherent analytical variability, length of time, and the amount of product needed for assay performance. The goal of this session is to further our understanding of characterization strategies and explore the challenges faced in the manufacture and licensure of these products. The session will include case-studies illustrating strategies used for gene transfer vectors products and for cell-based products. Concepts of Quality by Design paradigm and Risk-based approaches will be discussed when applicable. NOTES:

27

Stem Cell Discoveries in Translation and Clinical Trials Alan Trounson California Institute for Regenerative Medicine (CIRM), San Francisco, CA USA CIRM has initiated a strong translational program to begin ushering the discoveries of basic stem cell research into clinical application1. In order to do this, we have funded early translational projects that are focused on finding candidate products suitable and likely to progress to clinical trials. We have also initiated a program of preclinical development – Disease Teams – where we encourage researchers and biotechnology companies to create public-private partnerships2 to undertake comprehensive studies of candidate products (cell therapies, small molecules and biologics) that could obtain FDA regulatory registration approval for early clinical trials within 4 years. We have also initiated clinical trial work for pluripotential stem cells3 and expect these and other programs to continue to evolve with our support. We are exploring the possibility of developing a network of stem cell clinics in California for providing clinical trial expertise and cell therapies for proven applications. We recognized that important contributions are being made outside California and the desire by Californian scientists, clinicians and companies to collaborate with the best researchers globally. To date CIRM has 11 agreements with international governments and also agreements with a number of US States, Foundations and the NIH to collaborate in research and translation. We also have a partnership agreement with NINDS for iPSC Banking for neurodegenerative diseases. These agreements mean that the research has become a global initiative with California as a central core component. CIRM’s strategy will be to continue to fund the entire discovery to clinic pipeline of research in a focused outcome driven approach. As we enter the second half of the current fund-life of CIRM, it is important to demonstrate how effective the bond funded strategy is for accelerating benefit for Californians, and indeed the entire world. 1. Trounson A (2011) CIRM the way ahead. Regenerative Med. May;6(3):285-90. 2. Trounson A, Baum E, Gibbons D, Tekamp-Olson P (2010) Developing a model for successful

translation of stem cell therapies. Cell Stem Cell 6(6):513-6. 3. Trounson A, Thaker R, Lomax G, Gibbons D (2011) Clinical trials for stem cell therapies. BMC

Med. 2011 May 10;9(1):52. 4. Trounson A, DeWitt ND, Feigel EG (2012) The Alpha Stem Cell Clinic: A Model for Evaluating

and Delivering Stem Cell Based Therapies. Stem Cells Transl Med. (in press). NOTES:

28

Implementing the Principles of Quality by Design for Early Stage Gene Therapy Products Michael Kelly Genzyme, A Sanofi Company, Framingham, MA USA Gene therapy products are highly complex biologic entities manufactured and characterized using a variety of inherently complex manufacturing and testing methods. Historically, in many cases, the primary focus of early clinical development was on establishing improvements in vector design and safety, gene expression, and target validation and as a result, the challenges associated with their production and characterization of the investigational product was of secondary concern. Typically, this resulted “black box” manufacturing processes in conjunction with a Quality by Testing (QbT) approach to determine the suitability of investigational material for clinical use. Recent success in the clinic have led to a renewed optimism for the commercial value of gene therapy as a therapeutic platform and has redirected process development and manufacturing teams to think about the eventual challenges of establishing and maintaining a reliable and predictable supply of products to the market post approval, in accordance with regulatory expectations and in ways that are sustainable and cost-effective and the application of those issues to strategic process development approaches. Accordingly, the focus on process and analytical development is shifting, and many scientists are looking for ways to manage the risks associated with gene therapy manufacturing and characterization and to further enhance their product and process understanding. Many scientists are endorsing the principles of Quality by Design (QbD) to inform their process development strategies and to establish this knowledge base. This presentation will explore the general approach envisioned for the implementation of QbD to the development of early phase gene therapy products, and in particular, establishing a strategy to ensure that the clinical performance characteristics of those products can be understood and maintained throughout the product lifecycle while simplifying the implementation of process modifications and improvements and the regulatory path forward to product approval. NOTES:

29

Sequence Variants: What, Why, When, How, and Then What? Protein sequence variation is an inevitable consequence of the imperfections inherent in replication, transcription, and translation processes within living cells. Although some sequence variants are product-related substances, not impurities, they should be avoided and/or adequately controlled. This session will address what modern science can show about how frequently these events can occur, and what their effects may be on biopharmaceutical product quality attributes and on product development. The session will present the technologies commonly used to detect sequence variant presence, what are the options are to remove them or to mitigate their presence, and the implications they may have on regulatory filings. NOTES:

30

Detection and Mitigation of a Sequence Variant in a Recombinant Protein Expressed in Mammalian Cells Haimanti Dorai Janssen Research &Development, LLC, Radnor, PA USA New pharmaceutical products under development must undergo rigorous clinical testing to demonstrate safety and efficacy before being approved. To characterize the chemical structure of these complex protein products and help ensure product consistency through all phases of clinical testing, an arsenal of sophisticated technologies is routinely being used. One powerful technology that is available for biochemical and biophysical characterization of the protein drug products is peptide mapping. This is an effective method for evaluating the amino acid sequence and posttranslational modifications of proteins. If a product is determined to be overly heterogeneous, cell line development efforts may need to be renewed. Modern molecular biology methods, including quantitative PCR (QPCR), can be used to rapidly identify the underlying mechanism of heterogeneity and subsequently rectify it. As analytical technologies become more sensitive, it is likely that products previously believed to be homogenous will be revealed to contain microheterogeneity, as described here. This issue now can be addressed rapidly and accurately, expediting the product development process while maintaining superior product quality. A case study illustrating this evolution with a peptide–antibody fusion protein (FP, drug product) under development is presented below. NOTES:

31

Analysis and Characterisation of Sequence Variants in Recombinant Haemostasis Products Anne Kroll Kristensen; Niels Kristian Klausen; Marianne K. Hansen Novo Nordisk A/S, Måløv, Denmark Sequence variants of recombinant proteins are variants of the desired product that differ in terms of the primary sequence (i.e. truncation of N- and/or C-terminals, or incorrect processing in the cell). During development, it is essential to monitor and control sequence variants present in drug substance in order to gain a deeper understanding and hereby improved control of the final pharmaceutical product. Vatreptacog alfa is a recombinant human coagulation factor VII analogue with 3 amino acid substitutions allowing for higher in vivo activity. Analogous to eptacog alfa (NovoSeven®), vatreptacog alfa is expressed as a single chain molecule, which upon activation is cleaved into a two-chain molecule. With both O-linked and N-linked glycosylations present in addition to ten gammacarboxylated glutamic acid residues, several analytical separation techniques are necessary in order to fully characterise drug substance. Although many of the analytical techniques employed for separation of eptacog alfa sequence variants were found to be suitable for analysis of vatreptacog alfa, development of new methods were necessary in order to fully document a product in control. Analysis and characterisation of sequence variants present in vatreptacog alfa, will be presented along with data to support the efficient monitoring and control of these sequence variants. NOTES:

32

Detection of Sequence Variants in Recombinant Proteins: Sensitivity, Timing and Evaluation of Risks Kathleen Francissen Genentech, a Member of the Roche Group, South San Francisco, CA USA Sequence variants are unintended amino acid substitutions that can occur in recombinant proteins as a result of mutation, mistranslation, or improper post-translational processing. Detection of sequence variants has become an important aspect of cell line development. Sequence variants at levels above 5-10% are readily screened by a number of methods. However, sequence variants at low levels (< 1%) pose a greater analytical challenge, and raise new questions. As the detection sensitivity is improved, lower levels of apparent sequence variants can be observed. Error rates for transcription and translation in eukaryotes are impressively stringent, yet can be observed if one has sufficiently sensitive analytical methods. State-of-the-art mass spectrometry can now probe levels sufficiently low to observe the biological limits of amino acid sequence fidelity. Approaches for detecting low-level sequence variants will be described, as well as when the analyses are performed at different stages of development. In addition, examples will be given of sequence variants observed during monoclonal antibody cell line development. Approaches to assessing the risk posed by sequence variants will be outlined, based on the potential impact to safety, immunogenicity, pharmacokinetics, and efficacy. NOTES:

33

Advances in the Development and Characterization of Vaccines This session will highlight recent advances in the development and manufacture of vaccines. Regulatory Agency as well as the industry’s perspective will be presented, discussing current the opportunities and challenges of developing and manufacturing a vaccine for the US and the global market. Life cycle management aspects will be addressed. Additionally, there will be a focus on advances in developing product and process understanding of vaccine candidates, e.g. via Quality by Design approaches, including improvements in the development of control strategies and analytical characterization tools to identifying vaccine quality attributes that may be critical to clinical performance. NOTES:

34

A-Vax—QbD Vaccine Case Study Mark Schenerman MedImmune, Gaithersburg, MD USA Following the publication of the A-Mab Case Study in 2009, applying QbD principles to the development and production of an example monoclonal antibody, there were suggestions to do a vaccine case study. Considering the significant differences in development strategies between a monoclonal antibody and a vaccine, there was a clear rationale for creating a new case study. In early 2010, key industry and regulatory agency thought leaders were consulted to consider the feasibility of such a case study. Based on the feedback from these thought leaders, a consulting group was engaged to further develop the feasibility package and solicit participation from both industry and regulators. Five companies (GSK, MedImmune, Merck, Pfizer, and Sanofi Pasteur) responded to the solicitation and committed to participation in the Vaccine Working Group (VWG). The main objective of the VWG was to work together to see how (and if) QbD could be applied to vaccine development and manufacturing. The presentation will describe the case study that is now published, illustrating application of some key QbD principles for a fictitious carbohydrate-conjugate VLP vaccine. NOTES:

35

Life Cycle Management of Vaccine Products: Adapting Production Processes to Better Meet Tomorrow's Global Needs Colette Ranucci

; Risat Jannat; Tara Tagmyer; Paul Duncan

Merck & Co., Inc., West Point, PA USA Managing the life cycle of a vaccine from a global perspective requires identification of the key process and product characteristics required to expand access to new and emerging markets via sustained research and development targeting process and product improvements, continued post marketing surveillance, as well as post licensure clinical studies. Focus on these key areas with the goal of process redefinition/adaptation, will enable key products to reach those areas with the greatest medical need, thus enabling the full medical impact of the vaccine. Additionally, life cycle management of vaccine products also requires focused efforts on the sustained safety of the vaccine product with regard to emerging adventitious agent concerns. In the area of sustained research and development, Merck's current approach to supporting licensed vaccine products is to assess and address any product risks (deficiencies in the production process, product profile, or cost structure) by more comprehensively exploring the parameter space of the end to end vaccine processes to not only comprehensively map the robust operating space via statistically designed experiments, but also to identify potential opportunities for continued improvements in a manner which can be supported from a regulatory and clinical perspective. Suitability of raw materials for long term manufacture are an additional consideration that can be factored into this design space analysis, thus building a database of suitable alternatives which would provide greater quality assurance. This presentation will describe a holistic approach to the life cycle management process to ensure that regulatory expectations, quality expectations and global customer needs are met throughout the life of a vaccine product. NOTES:

36

Challenges for the Production of a Well-characterized Vaccine Robin Levis CBER, FDA, Rockville, MD USA Vaccine antigens are produced using a wide variety of manufacturing processes. Inherent in these processes is the need to produce a safe, high quality, efficacious product. The demonstration of these features for vaccines is accomplished through careful process development and product characterization, and well designed clinical trials. While some vaccines are comprised of highly purified, well-defined antigens, the ability to “well-characterize” most vaccines is challenged by the complex nature of the process and by the complex nature of the final product. New approaches in the characterization of biological products have been put into practice and have resulted in the development of well controlled manufacturing processes and more highly characterized final products. These approaches are now being applied to the manufacture of vaccine products. This presentation will highlight the regulatory approaches to the application of Quality by Design principles to the development and control of vaccines. NOTES:

37

New & Emerging Technologies: Not Just Because We Can This plenary session will focus on new and emerging technologies that advance analytical evaluation and manufacture of new biological macromolecular entities with improved clinical performance. There are several technical aspects of these biological products where the more we know, the more questions we seem to have. As examples, there are:

• challenges associated with the characterization of post-translational modification variants – how detailed do we need to get?

• new chemically or genetically modified products – can we distinguish relevant changes from “silent” changes?

• the measurement and evaluation of potency as related to clinical biomarkers – can these be tied to efficacy?

• and formulations/delivery systems incorporating nano-particles – do we need to understand the mechanism of action or just show that they are effective?

These and other aspects of characterization of biological products and the processees that make them beg for new tools and analytical approaches to address the new questions posed. These new tools and approaches are the intended focus of this plenary session. NOTES:

38

GlycoPEGylation - A Robust Technology for Modification of Coagulation Factors Are Bogsnes Novo Nordisk A/S, Gentofte, Denmark Several PEGylation chemistries are used for modification of protein drugs with the aim to increase half-life in plasma. Such modifications are most often done via reactive amino acid side groups or terminals. This may lead to a heterogeneous product due to the high number of sites available. A different approach is conjugation of PEG via glycan structures, known as glycoPEGylation. A number of proteins, such as coagulation factors FVII, FVIII and FIX, have inherent N- and/or O-glycan structures suitable for modification. As the number of potential sites available for PEGylation is typically much lower compared to that for reactive amino acids, a more homogenous product may be obtained. GlycoPEGylation is an enzyme mediated reaction, and well defined reactions and specific targeting offer certain possibilities of process control. Aspects of this conjugation approach will be discussed from a manufacturing and regulatory perspective. NOTES:

39

High Throughput Biophysical Approaches to Preformulation Characterization and Comparability Analysis of Protein Drugs and Macromolecular Vaccines David Volkin Department of Pharmaceutical Chemistry, University of Kansas, Lawrence, Kansas USA The ability to better characterize the higher-order structure and conformational stability of protein drug and vaccine candidates remains an important analytical challenge during development. In our laboratories, a combination of biophysical techniques is used to characterize the higher-order structural integrity, hydrodynamic properties, and conformational stability of macromolecules as a function of environmental stress. The data sets from different instruments are combined to produce empirical phase diagrams (EPDs), providing a direct visualization of the overall higher-order structural stability of the sample as a function of solution conditions (e.g., temperature, pH, excipients). By implementing high-throughput approaches, EPDs can be rapidly generated with minimal material. The challenges and opportunities of applying these biophysical analyses to preformulation characterization and comparability assessments will be discussed using several recent case studies, including well-characterized proteins and more complex vaccine antigens such as enveloped virus like-particles. NOTES:

40

Applications of State-of-the-Art Glycan Analysis to Biopharmaceuticals Pauline M Rudd1; Eoin Cosgrave1; Oscar Potter1; Michael Schomberg1; Mark Hilliard1; John O’Rourke1; Giorgio Carta1; Margaret Doherty1; Tharmala Tharmalingen1; Karina Marino1; Jayesh Kattla1; Jonathan Bones2; Weston Struwe1; Catherine Hayes3; Niclas Karlsson3 1NIBRT, Dublin, Ireland; 2The Barnett Institute, Northeastern University, Boston, MA USA; 3University of Gothenburg, Gothenburg, Sweden Biopharmaceuticals can rarely, if ever, be defined with the same precision that applies to small molecule drugs. In contrast to the polypeptide chain, post-translational modifications (PTMs) pose a particular challenge because they are not directly template driven. PTMs reflect the structure of the protein, the cell in which the protein is expressed and the conditions that prevail in the bioprocessor. Most are also subject to ‘flicker’, meaning that a protein is a heterogeneous mixture of post translationally modified variants. It is therefore a significant challenge to set limits for the levels of critical modifications of a biotherapeutic that are important to maintain safety and efficacy. This is particularly true of glycosylation, where there is usually considerable heterogeneity, and the glycans are known to be important in maintaining half life, introducing potential antigenicity and modulating the function of the therapeutic. The glycosylation pathway involves the concerted action of some 600 proteins and stressed, overproducing cells can exhaust the machinery giving rise to mis-glycosylated products. Glycosylation needs to be monitored at every stage of the process, from cell selection to choice of media to batch release, and a suite of technologies are now available. Each technology provides specific information and comes with advantages and constraints so that understanding what is required at any stage in the process is important. This talk will discuss some of the state of the art technologies including HILIC UHPLC, RPHPLC, microfluidics chip cube technology, capillary electrophoresis and mass spectrometry, showing applications to particular products such as erythropoietin and monoclonal antibodies expressed in a variety of cell lines. The development of novel methodologies for 96 well plate robotic platforms for the analysis of N- and O-glycans, milk oligosaccharides and the serum glycome are coupled with the development of data bases and software to enable data interpretation which will be linked to raw data streams from the various separations technologies.

• A systematic approach to protein glycosylation analysis: a path through the maze. Mariño K, Bones J, Kattla JJ, Rudd PM. Nat Chem Biol. 2010 (10):713-23.

• Method for milk oligosaccharide profiling by 2-aminobenzamide labeling and hydrophilic interaction chromatography. Mariño K, Lane JA, Abrahams JL, Struwe WB, Harvey DJ, Marotta M, Hickey RM, Rudd PM. Glycobiology. 2011 21(10):1317-30.

• UniCarb-DB: a database resource for glycomic discovery. Hayes CA, Karlsson NG, Struwe WB, Lisacek F, Rudd PM, Packer NH, Campbell MP. Bioinformatics. 2011 May 1;27(9):1343-4.

NOTES:

41

Assay Lifecycle Management: It’s Not a Mistake, It’s a Design Feature Release assays for biotech products developed and licensed many years ago are often associated with widely variable and dated technologies. In many cases, the original potency assays were animal based and susceptible to wide variation. In other cases, the instrumentation described in the original license application is obsolete and no longer supported by the vendor. There may also be examples where an evolving technology provides a better quantification of results. Furthermore, as older processes are themselves being modernized or revalidated (perhaps because they are being moved to new facilities), appropriate analytical tools with higher levels of precision are expected. Introducing new technology brings with it many challenges – bridging data between methods, understanding offsets, reacting to new data, applying appropriate statistical methods and resetting specifications. In this session, examples of strategies for assay life cycle management as well as examples of implementation of new assays will be presented. NOTES:

42

Management of the Analytical Method Life Cycle for Biotechnology Products: A Regulatory Perspective Rashmi Rawat CBER, FDA, Bethesda, MD USA This presentation will provide a regulatory perspective on analytical method qualification and validation study expectations during product development. Case studies will be presented and discussed to illustrate some of the common hurdles seen during method transfer studies and analytical comparability studies. Regulatory concerns that should be considered when planning a study to support an analytical method change or analytical method transfer or replacement will also be discussed. NOTES:

43

Analytical Lifecycle for Marketed Protein Therapeutics John Stults; Dieter Schmalzing Genentech, a Member of the Roche Group, South San Francisco, CA USA The years following market approval for a protein therapeutic are a period of evolution for product understanding -- a manufacturing history is developed, methods are monitored, and new biological understanding may be discovered. Years, even decades, of experience demonstrate the process is in control and that a safe, consistent product is produced. As time progresses, process improvements and transfers of the manufacturing site may occur. At the same time, the regulatory environment evolves with new regulations and guidances, and with product approvals in multiple countries. Among the many facets of evolution, analytical techniques and strategies continue to improve, and there is an opportunity for updating of the analytical components of the product, including re-examination of protein characterization and of the control system. Among these updates, acceptance criteria are re-examined based on manufacturing history, and assays are evaluated for their ability to measure important quality attributes and understand the stability of the product. New data may provide the justification to modify, add, or remove assays. A risk-based approach is used for prioritization and decision-making. Potential changes to the control system for older products present a number of practical challenges: archival samples from pivotal trials may no longer be available, clinical trial data are often limited, process understanding is less extensive than for new products, old technologies may no longer be available yet state-of-the-art technologies may not be readily available in developing countries. Furthermore, regulatory and supply chain strategies are complex for control system changes in a global manufacturing and supply network. In this presentation, we will examine the analytical lifecycle strategy for marketed products, highlighting both the opportunities and the challenges. NOTES:

44

References/Standards: Best Practices and Emerging Directions Product references/standards are an important component of a biopharmaceutical quality system. They help assure the continuity of the quality of the therapeutic entity throughout product development and commercial manufacture. This plenary session with speakers from the government, industry and standard setting organizations will discuss two critical issues with the selection, qualification and monitoring of product references/standards.

1. How is this done and what are some of the key issues for product references/standards? 2. How does one know that the assays to assess the reference/standard are working as intended?

NOTES:

45

Reference Standards for Monoclonal Antibodies: Key Challenges Addressed Matthew Borer Eli Lilly and Company, Indianapolis, IN USA Reference standards play a critical role during the development of pharmaceutical products and subsequent commercial manufacturing. The reference standard material and associated information serve as the basis for drug substance and product identity and strength, in addition to other critical attributes. As a result, it is important to select the right source material, perform an appropriate qualification, and properly maintain pharmaceutical reference standards. This talk addresses key challenges associated with managing reference standards for monoclonal antibodies. Specifically, solutions for selecting source material, packaging, certification of biological activity, and monitoring stability will be presented. NOTES:

46

Standardization of Optical Particle Counters Dean Ripple National Institute of Standards and Technology (NIST), Gaithersburg, MD USA Particles in protein therapeutics composed of aggregated protein monomers may cause an immune response in patients. Consequently, industry and the FDA desire more accurate methods for counting and characterizing particles. Aggregated proteins have an irregular morphology, low optical contrast, and a wide size variation. Furthermore, the properties of the particles vary depending on what stresses to the protein monomer solution led to particulate formation. This talk will discuss methodologies and standards to support this measurement problem. Typical sources of measurement error will be described for two common optical methods: flow microscopy and light obscuration. The particle sizes reported by light obscuration depend strongly on the particulate refractive index; optical modeling of the scattering of protein particles enables a substantial reduction in light obscuration counting errors. For flow microscopy, tests with existing reference materials are useful for identifying instrument errors and biases. However, rigorous validation of these instruments requires reference materials that mimic the properties of protein particles. At NIST, we have developed a polymer-based protein surrogate based on the partially fluorinated polymer ethylene tetrafluoroethylene (ETFE). Mechanical abrasion of the ETFE followed by size filtration produces a polydisperse suspension of very rugged particles with a morphology quite similar to protein particles. In another approach, we have created artificial polymer particles as small as 3 µm x 4 µm x 40 µm in large quantities using photolithographic methods developed for the semiconductor industry. These reference materials will provide a stringent test of the sensitivity and morphological measurements of optical particle-counting and particle-characterization instruments. NOTES:

47

Novel Standards for Novel Applications Chris Jones Laboratory for Molecular Structure, NIBSC, Hertsfordshire, United Kingdom The rapid pace of change in the development and analysis of biopharmaceutical products demands a continuing reassessment of the types of standards which are required. The approaches by which these standards are produced, calibrated and used also change. The first focus of this talk will the possibilities of new types of standards, including polysaccharide and protein quantification standards calibrated in the Système International d’unités (SI) system and utilising the ISO 34 guideline, developed as a quality standard for producers of reference materials. A change in the paradigm used for these materials has implications for calibration approaches, estimates of uncertainty of measurement and ensuring the standards are accepted by industry and by regulators. The second focus will be on the development of horizontal (or procedural standards) which are suitable for assessing the performance of techniques such as glycan or peptide mapping, or spectroscopic characterisation of conformational integrity. Poor quality and or resolution analytical methods may fail to robustly detect small changes. Horizontal standards, targeted towards establishing the quality of physicochemical methods widely used to define comparability, challenge the performance of the analytical method, rather than being product-specific. Their development and appropriate usage creates challenges for reference material producers. NOTES:

48

NOTES:

49

INSERT CASSS CALENDAR OF EVENTS

50

INSERT AAPS COLOR AD FOR NBC

51

NOTES:

52

Workshop Descriptions

Workshop Session 1

Monday, January 23, 2012 16:45 – 18:00

Specifications for Biotech Products: Still an upward journey? John Dougherty, Eli Lilly and Company; Sarah Kennett, CDER, FDA; Philip Krause, CBER, FDA; Anthony Mire-Sluis, Amgen Inc. Workshop Description Grand Ballroom A The process for developing specifications for biotech products has always been a challenge due to the number of quality attributes they possess, balanced by the understanding of what may or may not be important to measure on lot release and stability. In addition, the number of lots manufactured during clinical development is often limited; therefore do not provide enough data for example, raw material and other process variables to allow for a statistically valid assessment of true product variability. The paradigm of being restricted to ‘clinical qualification’ still remains, although the relevance of specifications based on manufacturing capability in the absence of clinical exposure remains an issue. In addition, there is the apparent requirement to have specifications for some non critical quality attributes for the sake of measuring process consistency, although the definition of a CQA would have relevance here. Therefore, even with the advent of Quality by Design concepts, it seems the list of attributes on specifications only gets longer (e.g. the addition of increasingly smaller sub visible particle sizes) and expectations for more and tighter specifications earlier during the product development lifecycle. This workshop will discuss these issues, attempting to answer the following questions:

1. What level of minor species control is required for an IND filing and how does one set appropriate limits?

2. How can the use of QbD tools (e.g. Product quality and process risk assessments, in vivo clinical sample analysis, prior knowledge etc.) impact the number of quality attributes on the specifications?

3. Do we really need drug substance specifications? Or can thorough process understanding and control allow for reduced DS specifications in association with more standard DP specifications?

4. Would a robust process characterization and control/trending strategy obviate the need for having attributes on specifications to measure consistency?

NOTES:

53

To Boldly Go Where QC Has Never Gone Before...Tales from the Frontier of New Analytical Technology Implementation Lokesh Bhattacharyya, CBER, FDA; Drew Kelner, Amgen Inc.; Jackie Larew, Eli Lilly and Company; Joao Pedras-Vasconcelos, CDER, FDA Workshop Description Grand Ballroom B This workshop will explore multiple dimensions and perspectives for the drivers, barriers, and successful strategies of implementing new technologies in QC laboratories. Participants will be encouraged to share best practices, lessons learned (possibly the hard way!), and unique strategies from their own adventures into the world of new technologies. Considerations ranging from cost implications to CQA measurements, from reduction in assay variability to vendor engagement, and from robustness to regulatory challenges are guaranteed to make for a lively and engaged discussion...don't miss out! NOTES:

54

Antibody Drug Conjugates Kevin Anderson, Seattle Genetics, Inc.; Kathleen Clouse, CDER, FDA; Jonathan Harris, Genentech, a Member of the Roche Group Workshop Description Grand Ballroom C Monoclonal antibodies (MAbs) coupled to cytotoxic agents, referred to as antibody-drug conjugates (ADCs), are rapidly emerging as a platform of products in oncology drug development. The therapeutic approach of ADCs is to utilize the specificity of a monoclonal antibody to deliver a cytotoxic drug to tumor cells. Depending on the chemistry of linkage, sites of attachment, and synthetic route for the small molecule drug, significant chemistry, manufacturing, and control (CMC) issues arise during development and regulatory review. Because conjugation can occur at multiple, but not necessarily all, available sites on the antibody, many species of conjugate molecules are generated for a given ADC. Therefore, since an ADC product is a mixture of conjugated species, appropriate tests are important to measure the heterogeneity of these products and to ensure consistency. These include determining the appropriate assays to characterize and release ADC products as well as identifying potential critical quality attributes (CQAs) for the conjugates, including those for the antibody, the cytotoxic agent, and the linker. The goal of this workshop is to discuss quality attributes and associated analytical approaches for ADCs and their components.

1. What quality attributes are unique to antibody drug conjugates and what analytical tools can be applied to monitor them?

2. How much characterization should be performed on the small molecule and are there different requirements for the linker and cytotoxin?

3. Which attributes are shared by the small molecule and antibody precursors and the antibody drug conjugate?

4. Which aspects of drug load distribution must be captured as part of the certificate of analysis verses characterization and comparability?

5. What happens when the conjugation process interferes with the ability to monitor quality attributes (e.g. charge variant analysis on ADCs with lysine-based conjugation chemistry, whole mass analysis on antibodies with reduced inter-chain disulfides)?

NOTES:

55

Drug Delivery Devices as Combination Products – Understanding the Additional Challenges: Is It Worth All the Added Effort? Julia Ding, PPD, Inc.; Suzanne Kiani, MedImmune; Maria Lugo Gutierrez, CDER, FDA; Deborah Trout, CBER, FDA Creating a combination product involves the technical challenges of developing and manufacturing a device and suitable formulation to deliver a drug/biologic, plus the regulatory challenges for devices (CDRH) and drugs/biologics (CDER/CBER). Innovators typically have a strategy to address the technical aspects of product development, however the regulatory expectations for a particular combination product may be challenging. Why? Because the regulations for devices were not written to address drugs/biologics, and vice versa. In some cases there are gaps between the cGMPs for drugs/biologics and the Quality System (QS) regulations for devices. This workshop will explore the technical and regulatory challenges associated with the development of combination products for devices and drugs/biologics. Questions to be addressed include the following:

1. What have been your successes and frustrations in developing combination products? For example, in regard to areas such as:

• identification of critical quality attributes, • requirements for Human Factors/Usability testing, • development of Labeling and Instructions for Use, • data to support Self Administration and Home Use claims?

2. How are you addressing the gaps between cGMPs for the drugs/biologics and Quality System (QS) regulations for devices?

3. Are you benefitting from a learning curve from one generation of combination product to the next?

4. How would you describe your experiences with combination products, from a risk/reward perspective?

NOTES:

56

Workshop Session 2 Tuesday, January 24, 2012

10:30 – 11:45 Testing Quality Requires Quality Testing: Managing the Introduction of New Technologies for Adventitious Agent Testing Arifa Khan, CBER, FDA; Richard Ledwidge, CDER, FDA; Mark Plavsic, Genzyme, A Sanofi Company; Timothy Schofield, Arlenda, Inc. Workshop Description Grand Ballroom A The goal of this session is to discuss applications of new molecular technologies, such as deep sequencing and microarray based techniques, for detection of broad range of adventitious agents in biological materials. Participants will be encouraged to discuss the benefits and limitations of these and other emerging technologies, their most suitable areas of application, quality requirements related to assay reagents, reference standards, and method validation before routine method implementation. The workshop will focus on questions related to application and validation of the new technologies: Method Application:

• What are the main advantages and disadvantages of the emerging adventitious agent detection technologies?

• What would be the most appropriate applications of these technologies? • Can these techniques be used to fill the current gap in adventitious agents testing methods? • Should these methods be used as complementary to or replacement of the existing methods? • What would be appropriate follow up in case of a positive signal?

Quality & Method Validation:

• What would be appropriate positive controls in these broad ranging detection methodologies? • What reference standards are needed? • How should the critical assay reagents be standardized to ensure high quality and low probability of

cross-contamination via reagents? • Are these methods validate-able, and what are the critical points to be considered for method

validation? NOTES:

57

Achieving and Interpreting Similarity of Biologics: Implications for Product Development Strategies and Product Claims Christopher Joneckis, CBER, FDA; Peter Richardson, European Medicines Agency; Nadine Ritter, Biologics Consulting Group, Inc.; Emily Shacter, CDER, FDA Workshop Description Grand Ballroom B The Biologics Price Competition and Innovation Act (BPCI Act) requires a product to be ‘highly similar’ to its reference product if claimed to be a biosimilar. Currently, a large array of physicochemical and functional assays is available to characterize biological products, which may permit the demonstration of high analytical similarity to the reference product and thus reduced clinical development activities in order to show equivalent safety and efficacy profiles. Remaining challenges cover the development of a manufacturing process that delivers a complex biosimilar product, the justification for interchangeability and the question of how to ensure similarity of biosimilar and reference product over time. The focus of this workshop is on the development and evaluation of biosimilar products. Topics for discussion include the demonstration of high similarity at the quality level and also the conclusions that can be made with regard to the further development program including the final drug product claims. Questions:

1. What are the best approaches to apply characterization methods for the optimization of a biosimilar manufacturing process and do they differ from those used for development of other biotech drugs?

2. What kind of results for similarity at the quality level (physicochemical and biological characterization) can justify the reduction of the preclinical and clinical development program?

3. Do different product classes require different assessments and what are the consequences for the guideline system?

4. How should product development be aligned with the proposed FDA biosimilar product development meetings?

5. How similar is similar enough when it comes to substitution of biologics in clinical practice? 6. What is necessary to justify interchangeability?

NOTES:

58

Sequence Variants: The How, What, When, Why and What Now? Ralph Bernstein, CDER, FDA; John Bishop, CBER, FDA; Reed Harris, Genentech, a Member of the Roche Group; Ziping Wei, MedImmune Workshop Description Grand Ballroom C Two types of sequence variants can be detected during biopharmaceutical development: new variants due to process changes and existing variants due to the application of more sensitive analytical technologies. These newly detected variations of sequence in either the coding DNA or in the amino acid sequence may have potential impacts on patient safety and efficacy, and may need to be controlled and justified for their levels. Changes in sequence can be seen as a rare, but significant event. This workshop will discuss the detection, risk assessment, acceptable ranges and risk mitigation of sequence variants of biopharmaceutical products. Please attend with real world examples to share and contribute to a lively discussion. Questions to be discussed:

1. What is the definition of sequence variants? 2. Which analytical techniques are being employed to detect low levels of sequence variants? 3. What are the approaches for risk assessment of sequence variants to evaluate their potential risk

to patient safety and efficacy? How to establish their acceptable ranges? 4. What are the examples of strategies to mitigate the risks from sequence variants? 5. How are amino acid sequence variants different from other product variants or heterogeneity

(e.g., C-terminal truncation, deamidation, isomerization, oxidation)? NOTES:

59

Linking Product Quality Attributes to Safety and Efficacy of Cell and Gene Therapy Products Pakwai (Patrick) Au, CBER, FDA; Steven Bauer, CBER, FDA; Joann Parker, Pfizer, Inc.; Darin Weber, Mesoblast Workshop Description Union Square Room Cell and gene therapy products are highly complex because of their methods of manufacture, their intricate molecular composition and structure, and the challenge of relating their analytical characteristics to their functional properties. In this workshop, we will discuss tools and approaches that have been or could be used to help to establish the linkage between product quality attributes and safety and efficacy of these complex biologics. The understanding of the linkages is essential for identifying critical quality attributes that may have an impact on product’s safety and efficacy. Discussion will focus on the needs and challenges in the characterization of cell and gene therapy product and the contribution of preclinical studies to product characterization. Questions:

1. How can we establish comparability for cell-based products? • during scale up: small vs. large scale? • comparability of different MCBs? • autologous vs. allogeneic cell-based products?

2. How can we use preclinical (e.g., pharmacology/toxicology) studies to refine product characteristics?

3. Are current approaches to cell product characterization of sufficient predictive value for safety and efficacy?

4. What are the current challenges in gene therapy product characterization? 5. What are the additional challenges for product characterization when using genetically modified

cells? 6. How can we interpret the growing information on genetic variability and how does it impact our

views on cell-product safety? NOTES:

60

Workshop Session 3 Wednesday, January 25, 2012

11:00 – 12:15 Rapid Pharmaceutical Product Development: Getting Off the Critical Path and onto the Right Path Doug Cecchini, Biogen Idec; Roman Drews, CBER, FDA; Ilona Reischl, AGES PharmMed; Barbara Rellahan, CDER, FDA; James Stonecypher, Micromet, Inc. Workshop Description Grand Ballroom A This workshop is a follow up to the full-day CMC Strategy Forum on the topic of Rapid Pharmaceutical Product Development. Both industry and regulators are looking for opportunities to speed up development of life-saving and life-changing drugs without sacrificing their quality or the safety of patients. As such, the identification of issues that can accelerate or inhibit development will help establish best practices that maximize resources and shorten time to market while maintaining regulatory compliance. This workshop will be used to summarize outcomes of the full-day CMC Strategy Forum and to continue the discussion of this topic. Potential areas of discussion include:

• From industry’s perspective, what are the most critical aspects of product development that limit time to approval?

• How the following factors may contribute to rapid product development and successful product licensure:

– Leveraging prior product and process knowledge – Implementation of new manufacturing technologies and analytical methods – Progressing from the R&D stages of early product development to IND phase and

commercial manufacturing (e.g. introduction of Quality System elements) – Use of the QbD approach, including implementation of risk management methods – Earlier evaluation of process scalability and robustness to identify critical control points and

reduce comparability challenges later in development – Introduction of process improvement methodologies (e.g. Six Sigma, Operational

Excellence) – Technology transfer and outsourcing

• From a regulator’s perspective, what are the actions or lack thereof which increase risk and can

potentially cause delays (e.g. clinical hold) NOTES:

61

New and Emerging Technologies – Timing is everything? Lori McNamara, Medtronic; Ashutosh Rao, CDER, FDA; Joseph Siemiatkoski, Percivia, LLC Workshop Description Grand Ballroom B New analytical techniques are rapidly evolving and adapting to these emerging opportunities in analytics is becoming part of product lifecycle management. New technologies present significant scientific opportunities to better understand and control product quality. While the identification of gaps in analytics for each product can be used as a starting point for the adaptation of new technology, other intended or unintended reasons can trigger the need for newer analytics. The development and implementation of new analytical techniques can be challenging, especially for legacy products. This workshop will facilitate a discussion on the needs and challenges for modernizing analytical methods for new and legacy products. The need for a comprehensive risk/benefit analysis that includes scientific, regulatory, and commercial considerations will be discussed with a view to generating ideas on an optimal strategy for adapting to new and emerging technologies.

• Modernizing assays for legacy products, why should you do it? • When do you consider updating your test method with new technologies? What are the

commercial and regulatory considerations? • Examples of new and emerging technologies to tackle gaps in analytics.

NOTES:

62

Quality by Design for Early Phase Development Audrey Jia, CDER, FDA; Elisabeth Kirchisner, F. Hoffmann-La Roche Ltd.; Laurie Norwood, CBER, FDA; Victor Vinci, Eli Lilly and Company Workshop Description Grand Ballroom C Quality by Design is intended to facilitate enhanced product and process understanding in combination with quality risk management with the goal of establishing an appropriate control strategy. Regulatory guidances including those from ICH have described what should be considered in taking such an approach for the lifecycle of a product. For biologics, there have been efforts from industry and worldwide regulatory agencies to provide concepts and suggest approaches – case studies (A-Mab), points to consider (ICH), pilot-programs, etc. These have been primarily directed at the later phases of development and BLA submission. While the benefits of a QbD approach to facilitate product quality and a lifecycle approach may be agreed, the specific tools such as how to do risk assessments, DOEs, and evaluate CQAs are still evolving. In addition, industry is continuing to assess the cost and benefit of how and when to apply QbD to product development. The workshop will discuss how small and large companies might apply efficient use of QbD principles early in development, which tools are appropriate, some examples of how application was done, and regulator thoughts on how the application of QbD principles can help with key questions in early phase development. Some key questions that will be discussed during this workshop:

• When during early development do companies start QbD activities and which principles/ tools are applied?

• How are QbD principles able to guide process and product development during early development?

• What are the expected benefits or limitations in applying QbD principles during early development?

• Are there any specific expectations the reviewers have concerning QbD during early development?

NOTES:

63

Impact of Improved Scientific & Technical Understanding on Vaccine Quality and Supply Collette Ranucci, Merck Manufacturing Division; Steven Rubin, CBER, FDA; Michael Washabaugh, MedImmune Workshop Description Union Square Room The availability & effectiveness of vaccines has been one of the most important contributions to public health in the past century. The development of an increasing number of vaccines is leading to dramatic global health benefits, in both traditional and emerging markets. Vaccines are complex biologic entities whose availability is intimately linked with the manufacturer's level of understanding of the process & the product. In some cases, as much as 50% of the manufactured product is discarded due to manufacturing and control problems, an inefficiency that significantly contributes to product shortages and cost and one that highlights the need to change the paradigm of how vaccines are produced. To a large extent, Quality-by-Design (QbD) is seen as a major step towards addressing these issues; however, due to the complex nature of vaccines, it is challenging to comprehensively define a product’s critical quality attributes to permit the design of a process that yields consistent product quality. Key questions to discuss are:

1. Is it practical to apply to all key steps of the manufacturing process? 2. In what cases is supporting clinical data required, and what regulatory filing requirements will be

required? 3. Are there other characterization strategies to support globalization goals of decreasing product

cost, increasing product volumes and identifying product profiles suitable for emerging markets? 4. What product attributes are desirable for ensuring the supply of the next generation of vaccines?

Freeze-thaw stability? Room temp stability? Dose sparing technologies? NOTES:

64

Workshop Session 4 Wednesday, January 25, 2012

14:00 – 15:15 Assay Lifecycle Management: How, What, When and Why Alfred Del-Grosso, CBER, FDA; David Good, Eli Lilly and Company; Joseph Molon, Biogen Idec; Jennifer Swisher, CDER, FDA Workshop Description Grand Ballroom A There is a fair amount of guidance available on manufacturing process lifecycle management with more coming in ICHQ11, but very little guidance on analytical method lifecycle management. However significant changes and advances in analytical equipment and technologies are occurring at a faster rate than advances in processing technology. Sometimes these changes (i.e. supply of suitable columns) are outside the control of the pharmaceutical analytical laboratory. Coupling these factors with the cGMP expectation to stay current can put considerable pressure to change and improve assays throughout the lifecycle of a product. This in turn leads to numerous questions that will be explored in this workshop. Some examples are:

1. How to handle new impurity information generated post commercialization? 2. What are the best regulatory strategies for implementing new methods? 3. How does one ensure that methods remain in an appropriately validated state throughout the

lifecycle of the method? 4. Participants are encouraged to bring their own questions and concerns for discussion in the

workshop. NOTES:

65

Comparability Evaluations of Well-Characterized Biologics Sid Advant, ImClone Systems Corporation; Karen Lee, Genzyme, A Sanofi Company; Robin Levis, CBER, FDA; Lixin Xu, CDER, FDA Workshop Description Grand Ballroom B Over the course of a product lifecycle, manufacturers of well characterized biologics, including therapeutic vaccines, may elect to make changes to approved production processes. Implementation of significant process changes such as process scale-up or transfer of processes to new sites requires that the manufacturer demonstrate that the safety, identity, purity and potency of the pre- and post-change products are comparable. In this workshop, we will discuss how the assays used to monitor CQA’s during a comparability exercise are selected and applied. The development of comparability criteria and how these criteria are applied during the comparability exercise will also be a topic of discussion. Related topics such as the use of historical lots and statistics in setting comparability criteria as well as the role that degradation pathways studies and pre-clinical evaluations play in comparability studies will be discussed. A. What is a “significant process change”? B. For criteria:

1. What are the essential considerations for acceptance criteria of comparability study and how much are considered enough?

2. What role does QbD play in setting criteria for or performing comparability evaluations? C. Regarding CQAs:

1. How to address the prospect of increasing heterogeneity (including higher order structure)? 2. As analytic methods increase in sensitivity and the regulatory authorities demand the use of the

newest technologies, how will this impact upon comparability studies in the future? For example: • Is technological progression complicating matters? • How much variability is acceptable? • Does the commercial product represent the clinically tested version (historical data/trend)? • What if historical lots aren’t available for analysis? • How are criteria set for characterization assays, e.g. those used to probe higher order

structure?

D. Identify key considerations for comparability studies at all stages of development? 1. An argument for doing QbD, early?

E. Comparability Studies and Protocols

1. Are comparability protocols the most common regulatory vehicle for establishing comparability of approved products?

2. What is the role that degradation pathway studies play in comparability evaluations? How are these studies typically performed and interpreted?

3. How important is the availability of pre-clinical data in a comparability evaluation? Only important when the analytical results are inconclusive? What if an excellent animal model isn’t available?

NOTES:

66

Spotlight on Reference Standards - What is Required to Become “The Standard”? Markus Blümel, Novartis Pharma AG; Barry Cherney, CDER, FDA; Rajesh Gupta, CBER, FDA; Dieter Schmalzing, Genentech, a Member of the Roche Group Workshop Description Grand Ballroom C This workshop deals with the preparation, control and life-cycle concept for manufacturer’s in house reference standards (for the active pharmaceutical ingredient, API). The applicability of these concepts to other reference materials required for analytical purposes may also be discussed (only if time allows). Reference standards are used in quality control testing to monitor various attributes of the API. These standards are typically prepared from lots representative of clinical or marketed material and (if applicable) may also be calibrated against an (inter)national reference standard. Reference standards are a key element in each control strategy and help to assure continuity of assessment throughout product development and commercial manufacturing. Reference materials are used to quantify product- or process-related impurities or to establish system suitability. They are typically defined for a specific test. Examples are process-specific host cell protein mixtures, isolated product-related variants or standards for sub-visible particles. During this workshop you are invited to contribute your thoughts on topics of concern to both sponsors and regulatory agencies, such as: • Criteria for selection and preparation of a reference standard: Which representative lot to select? Manufactured under cGMP? What are necessary properties or critical quality attributes of a reference standard? Does the standard need to have the same formulation as the final product? What is the most suitable container for a reference standard?

• Which strategy for characterization and stability monitoring of reference standards are applied in

your company? How is the potency assigned and controlled? How do you show that your reference is stable (or still suitable for purpose)? How do you select appropriate tests and how do you justify requirements? How do you assess historic data obtained for the reference standard?

• Which life-cycle concept for reference standards is applied in your company – and why? Concept A: Declaration of consecutive standards (meet pre-defined specifications, establish equivalency to previous standard) Concept B: Definition of a primary standard, which is mainly used to qualify a secondary standard

intended for routine analytics. NOTES:

67

Sorting Out the Best Methods and Techniques for Visible Appearance Testing Pat Cash, MedImmune; Erwin Freund, Amgen Inc.; Laurie Graham, CDER, FDA; Deborah Trout, CBER, FDA Workshop Description Union Square Room While parental products are required to undergo a 100% inspection for particulate matter and primary packaging defects, this required testing may not be sufficient, for example, when proteinaceous visible particles are a pathway of degradation and potential CQA. In this case, the control strategy may need to include a semi-quantitative or quantitative visible particulate assay. This workshop will discuss overall strategies for monitoring and control of particulate matter, various aspects of the 100% visual inspection process, and the challenges associated with the development and validation of semi quantitative/quantitative visible particle assays. Some of the topics for discussion include:

• The definition of “visible” and “essentially free”; • re-inspection of culled containers and AQL acceptance criteria; • defining critical and major defects; • limitations and validation challenges of various techniques; • life-cycle approaches to visible particle analysis and control, including determining when a

quantitative/semi-quantitative assay is needed; • challenges with the detection of proteinaceous visible particles; • and the development of appropriate reference standards.

NOTES:

68

Technical Seminar Abstracts

Monday, January 23, 2012

13:00 – 14:00 In Grand Ballroom A Sponsored by Waters Corporation Simplifying Glycoprotein Characterization with Released Glycan UPLC Analysis and a Customized Biotherapeutic Glycan Database Elizabeth Gildea1; Ying Qing Yu1; Mark Hilliard2 1Waters Corporation, Milford, MA USA; 2National Institute for Bioprocessing Research and Training (NIBRT), Dublin, Ireland

Protein glycosylation has a critical impact on the safety and efficacy of most biothereaputic proteins, necessitating strong analytical and process controls during their production. To support such efforts, Waters has developed enabling technology for routine high performance glycan analysis by combining the separations power of ACQUITY UPLC with the unique glycan selectivity obtained with our proprietary HILIC mode Glycan Separation Technology (GST) chemistry. Excellent peak resolution, peak shape and run to run reproducibility using UPLC have enabled qualitative and quantitative released glycan analyses unmatched by other technologies and vendors. Working with our partners at NIBRT, we have now simplified and automated the task of data interpretation to give more robust and efficient glycan analysis workflows.

Waters and NIBRT, led by Prof Pauline Rudd, have co-developed GlycoBase 3+, a web-enabled proprietary database enabling glycan identification using calibrated and normalized peak retention data. GlycoBase 3+ contains retention data (expressed as Glucose Units or GU values) for more than 600 2AB-labeled N-linked Glycan structures derived from a diverse set of glycoproteins of biotherapeutic, and diagnostic interest. These values were obtained by strict and systematic analysis of 2AB labeled released glycans, analyzed using Waters leading HPLC/UPLC technologies as part of the NIBRT glycan analytical platform. Hydrophilic interaction liquid chromatography combined with fluorescence detection (HILIC-fluorescence), supplemented by exoglycosidase sequencing and mass spectrometric confirmation, was used to generated this high confidence labeled glycan library.

The resulting database has been made accessible through a customized Web-application containing a simple and intuitive interface for assignment and confirmation of glycan structures. The combined efficiencies of high resolution glycan analysis and simplified data processing will enable the biopharmaceutical industry to perform released glycan analysis with greater confidence and speed than has not been possible using existing analysis workflows.

This workshop, conducted jointly by Waters and NIBRT, will detail our platform for comprehensive glycan characterization, focusing on the benefits of UPLC technology and the novel implementation of GlycoBase 3+. Several case studies will document how the rigorous characterization of biopharmaceutical protein products enables biopharmaceutical companies to achieve the high standards required by today’s more aggressive regulatory environment. NOTES:

69

Monday, January 23, 2012 13:00 – 14:00

In Grand Ballroom C Sponsored by ProteinSimple Protein Particles and Purity – Now Simple, Now Automated Susan Darling1; David Thomas2 1ProteinSimple, Santa Clara, CA USA; 2ProteinSimple, Ottawa, ON Canada Proteins are complex, no way around it. They behave differently, come in a variety of shapes and sizes and some are linked to other molecules. Getting to “Well Characterized” is more of a challenge than ever and you need tools to help you get there, faster. iCE and MFI technology are routinely used to address protein purity and aggregation analysis and are now better than ever. Please join us for lunch and see how iCE and MFI help scientists simplify and accelerate complex protein analytics. NOTES:

70

Tuesday, January 24, 2012 12:15 – 13:15

In Grand Ballroom A Sponsored by Agilent Technologies Accelerate mAb Characterization Using Automated Sample Prep David Knorr and Ning Tang Agilent Technologies, Santa Clara, CA USA Scientists working in early mAb discovery and development often require small-scale, high-recovery sample preparation for their analytical characterization methods. The fully automated Agilent AssayMAP Bravo platform provides the highest antibody recovery possible and parallel processing of more samples in less time, with the reproducibility and purity needed to build a workflow. Discover how the AssayMAP Bravo automation platform provides true walk-away automation with exceptional reproducibility for mAb screening, capture, and sample preparation. Learn about available sample preparation workflows for mAbs and other recombinant proteins. Explore how automated sample preparation enables rapid characterization of mAbs by LC/MS. NOTES:

71

Tuesday, January 24, 2012 12:15 – 13:15

In Grand Ballroom C Sponsored by Bruker Daltonics, Inc. So You Think You Know Me?...Using Mass Spectrometry to Understand Your Biotherapeutic Laura Main Bruker UK Limited, Coventry, United Kingdom With ever more stringent demands from regulators, it is vital to stay ahead of the game to ensure you understand your biotherapeutics, so that they reach the market as fast as possible. Characterizing and understanding your biotherapeutic is essential at every stage during its lifecycle. From early discovery through the identification of potential “hot spots” to clone selection, scale-up, production and the determination of critical quality attributes to QC, mass spectrometric tools can help provide you with the necessary characterization knowledge quickly and confidently. Bruker offers a number of powerful and efficient mass spectrometry solutions for both routine and in-depth biopharmaceutical characterization, which combine industry-leading mass spectrometry performance with innovative software, ensuring accurate, sensitive and precise data from every analysis - delivering certainty first time, every time for your biotherapeutics. Here we present an overview of Bruker’s mass spectrometric solutions, with examples from pharmaceutical companies illustrating:

• Routine native intact, subunit and peptide mapping workflows • Screening and quantitation of Post-Translational Modifications and impurities • Automated characterization and profiling of glycopeptides and glycans • Rapid N and C terminal analysis • Fast detection of truncations • Automated Data Acquisition, Processing and Reporting

NOTES:

72

Tuesday, January 24, 2012 17:15 – 18:15

In Grand Ballroom A Sponsored by Beckman Coulter Aggregate and Particle Analysis in Biologics Production: Assessing Objects of All Sizes with Physical Techniques Matthew Rhyner Beckman Coulter, Brea, CA USA Protein scientists have long been interested in aggregates and particulates of all sizes: at molecular lengths, scientists are interested in complexes a protein may form with other molecules or binding ligands; at very large sizes, the main concerns are contamination and possible blood vessel occlusion for injectable therapeutics. The tools used to probe these biologic and chemical particulates vary from analytical ultracentrifugation to microscopy, depending on the size of the object. Particulate analytical techniques are often rooted in pure physical sciences, which were typically designed for distinctly non-biologic applications. This can sometimes cause confusion as concepts from the field of physics are translated into the field of protein science. Fortunately, there are some well-known pioneers who have effectively straddled the physical and biochemical fields and produced a deep body of work spanning several decades and utilizing many physical principles. This workshop will present a brief historical review of these works, and a sampling of the latest high-impact studies using analytical ultracentrifugation, dynamic light scattering, and the Coulter Principle. NOTES:

73

Tuesday, January 24, 2012 17:15 – 18:15

In Grand Ballroom B Sponsored by Catalent Pharma Solutions Virological Safety of Biopharmaceuticals: Regulatory Considerations and Pragmatic Approaches Hazel Aranha Catalent Pharma Solutions, Morrisville, NC USA Clinical acceptability of biologicals must, of necessity, be guided by a risk-based approach. While product virus safety requirements for marketing authorization applications (MAAs) are clearly promulgated in ICH Q5A (R1) and other guidance documents, specific guidance related to clinical products in development was not available until recently. Guidance pertaining to investigational medicinal products (IMPs) that came into effect in February 2009 is firmly grounded within the principles of risk assessment and management and allows for a more pragmatic approach. This presentation will review virus safety requirements applicable to MAAs as well as IMPs and their applicability from a process standpoint NOTES:

74

Tuesday, January 24, 2012 17:15 – 18:15

In Grand Ballroom C Sponsored by Wyatt Technology Corporation Absolute Macromolecular Characterization of Proteins Using Light Scattering and Related Techniques John Champagne Wyatt Technology Corporation, Santa Barbara, CA USA Light scattering (LS), including classical and dynamic, has been widely employed to characterize biotechnology-derived pharmaceuticals, such as proteins, DNA, polysaccharides, and viruses. Classical LS, especially multi-angle light scattering (MALS), determines the absolute molecular weight of pharmaceuticals in solution. Dynamic light scattering (DLS) directly measures diffusion coefficient and thus derives the hydrodynamic radius. Both MALS and DLS can be used as stand-alone (batch) instruments or online coupled with a separation system, such as liquid chromatography and field flow fractionation (FFF). When used in conjunction with a separation system, light scattering detection provides the absolute molecular weight, root-mean square radius, and hydrodynamic radius of individual peaks of interest. In addition, it can be used for the assessment of size heterogeneity (aggregation and fragmentation), molecular weight distribution, stoichiometry, and preliminary conformational analysis of proteins. Electrophoretic mobility (zeta potential) has been used as a means of measuring electrostatic interactions of proteins and colloidal particles and thus assessing their formulation stability. However, reliable measurements of molecular charge of different protein formulations are challenging due to the low detection sensitivity of the existing instrumentation. This challenge has been overcome by the newly-developed Massively Phase Analysis Light Scattering (MP-PALS) technology. In this technical seminar, we will discuss the latest advancements in MALS, DLS and mobility measurements and their applications for measuring protein oligomerization state and aggregation as well as protein-protein interaction, pegylated proteins, and membrane proteins. We will also present how, for the first time, electrophoretic mobility of protein formulations can be measured consistently and non-destructively at a reasonable sample concentration and volume. NOTES:

75

NOTES:

76

Poster Abstracts

Poster Session One

Monday, January 23, 2012 15:45 – 16:45

Grand Ballroom Foyer, Third Floor Pacific Terrace Foyer, Fourth Floor

P-101-M A Portal to the Lymphatic System: Intradermal Drug Delivery by 3M Microstructured Transdermal Systems Peter Johnson 3M Drug Delivery Systems Division, St. Paul, MN USA Blood plasma is continually leaking from arterial capillaries. Lymphatic capillaries collect interstitial fluid which is not reabsorbed by veinous capillaries, and return it to the circulatory system. The freely permeable structure of lymphatic capillaries enables macromolecules and cells to enter the lymphatic system much more readily than the blood capillaries. From a drug delivery perspective, the lymphatic system offers 1) an opportunity for uptake of biotherapeutics for delivery to the circulatory system, and 2) the uptake of vaccines for delivery to the lymph nodes. Access to the lymphatic system, via the lymphatic capillaries, is extremely challenging for delivery devices, because the lymphatic capillaries are just 0.4 mm below the surface of the skin. A great deal of skill is required to insert the needle of a syringe just 0.4 mm below the skin, and the volume of liquid formulation may be limited to just 0.1 mL. Consequently, there has not been much effort to develop delivery capabilities to utilize the lymphatic system, nor is there much data to demonstrate the therapeutic opportunities for the delivery of drugs and vaccines for uptake by the lymphatic system. This poster will review the basic physiology of the lymphatic system, and summarize studies that use the 3M Microstructured Transdermal Systems (MTS) to demonstrate the practicality of intradermal delivery, as well as the therapeutic effectiveness for uptake of biotherapeutica and vaccines by the lymphatic system. The MTS hollow microstructured arrays can deliver up to 2mL of solution formulation. For drugs, the t(max) value is faster than for subcutaneous delivery, and the C(max) value is greater than for subcutaneous delivery. For vaccines, the intradermal space is rich in dendritic cells, and the transport of antigens through the lymphatic system and to the lymph nodes, enables a faster and more potent immune response. P-102-M Trisulfides Identified as a Source of Variability in ADC Conjugations Katherine Cumnock Genentech, a Member of the Roche Group, South San Francisco, CA USA

77

Antibody-drug conjugates (ADCs) are therapeutic agents where cytotoxins or chemotherapeutic molecules are chemically linked to monoclonal antibodies using cysteine, lysine, or other chemistries. A critical quality attribute of ADCs is the average drug-to-antibody ratio (DAR). During the development of multiple ADC products utilizing cysteine linker chemistry, which targets interchain disulfides, variability was observed in the amount of reducing reagent required to achieve a target DAR. Variability was observed both between different lots of a given antibody as well as between different antibodies. This variability in required reductant seen during conjugation is a challenge for process development and has the potential to increase the variability of DARs during manufacturing. Interchain trisulfides (RS-S-SR) were identified as a major factor affecting the conjugation reduction stoichiometry. Using mass spectrometric analysis of peptide digests, trisulfide levels ranging from 0-13.6% were measured in preparations of unconjugated antibodies used as intermediates in the ADC process. Each trisulfide bond requires two moles of reductant to generate two free thiols for conjugation whereas a disulfide bond only requires one mole of reductant to generate the same two free thiols. The excess reductant required to reduce interchain trisulfide bonds is responsible for the deviation from the theoretical reduction stoichiometry and the variability observed during ADC manufacture. NOTES:

78

P-103-M Analytical Characterization of Antibody Drug Conjugates Samadhi Vitharana; Tamara Gelzleichter; Tian-Yi Cui; Kamal Kannan; Mike Buckley Takeda California, South San Francisco, CA USA Monoclonal Antibodies linked to cytotoxic small molecules known as Antibody Drug Conjugates (ADC), is an emerging technology for evaluation in clinical oncology. ADCs utilize the specificity of a monoclonal Antibody to deliver a cytotoxic drug to targeted tumor cells. The chemistry of the linker, the cytotoxic drug and the sites of attachment could result in significantly complex and heterogeneous preparations. This could lead to issues during development in the chemistry, manufacturing and control as well as regulatory review of ADCs. Therefore it is critical to have the appropriate analytical tools to characterize and assess the heterogeneity of ADCs for quality assurance. Analytical methods have been developed to characterize and evaluate different ADC architectures. The panel of assays include, methods such as Size Exclusion Chromatography (SEC) to determine the size distribution, CE-SDS and SDS-PAGE to explore the banding patterns while Hydrophobic Interaction along with Reverse Phase Chromatography to determine the drug-to-antibody molar ratio (DAR), as part of the core characterization techniques. For additional characterization tools such as UV, LC/MS and capillary IEF as well as assays to evaluate potency and stability of ADCs are being established. The analytical methods have demonstrated the ability to characterize, evaluate and compare different ADC technologies as well as understand the biochemical nature of ADCs successfully. P-104-M Analysis of Oxidation and PEGylation of Anti Staphylococcal Therapeutic Protein from Bacteriophage using HPLC Chip coupled to Advanced QTOF Ravindra Gudihal1, 3; Suresh Babu CV1, 3; Sundaram Palaniswamy2; Umamaheshwari S2; Suneel Basingi2; Ning Tang1, 3

1Agilent Technologies, Bangalore, India; 2Gangagen Biotechnologies Pvt Ltd, Bangalore, India; 3Agilent Technologies, Santa Clara, CA USA The biopharmaceutical industry requires detailed characterization of the protein drugs as mandated by regulatory agencies. Among many important protein characterizations, stability towards oxidation and PEGylation of the protein which is used to enhance the stability of the protein and increase the half life are very important. LC/MS technology is a powerful and sensitive technique for characterization of therapeutic protein drugs. In this presentation we show characterization of an anti-staphylococcal protein (P128) derived from bacteriophage for its stability towards oxidation and the PEGylation stability using Liquid chromatography/Mass spectrometry (LC/MS) technique. Forced oxidation studies using t-butyl hydroperoxide (t-BHP) was performed in order to investigate the possible sites that are susceptible to oxidation and the effect on the activity due to these oxidations. Samples were analyzed at low levels using a peptide mapping procedure on an HPLC-Chip coupled with an Accurate-Mass Q-TOF LC/MS.

79

The oxidized samples were compared to a non oxidized control sample. Statistical software was used to demonstrate the differentiation between levels of oxidation and was also correlated with the activity of the protein. For studying the PEGylation of protein a modified version of ESI-MS based “charge stripping” approach has been employed. Here the charge stripping agent (triethylamine) was used postelectrospray to aid in simpler mass spectrum and hence better interpretation of the results. The results show good improvement in the mass spectrum quality for studied PEGylated protein. NOTES:

80

P-105-M Host Cell Protein Characterization for Bioprocess and Product Quality Improvement Rong-Rong Zhu EMD Millipore, Bedford, MA USA Host cell proteins (HCPs) are among the major biomanufacturing process-related impurities and are considered as CQA (critical quality attributes) by regulatory agencies. HCPs must be closely monitored for downstream purification steps and final drug product to ensure process consistence and product safety. Although ELISA has been the most common method for HCP quantification, one of the major drawbacks of HCP ELISA is assay accuracy since no single available ELISA assay kit has demonstrated to contain antibodies capable of detecting all the HCPs in CHO based biologics production process. In addition, ELISA assays can not reveal HCP identities and characteristics such as molecular weight and pI, which are critical for upstream online monitoring and downstream purification process development and optimization. In this presentation, orthogonal methods such as 2D-silver/western, 2D-DIGE and LCMSMS were used to achieve comparative study of identity of HCP standard used in Cygnus HPC ELISA (Cygnus 3G) with HCPs from three in-house null CHO cell lines. The overlaps between Cyguns HCP standard with in house HCP are 33% to 78% by western blotting and 50% to 60% by LCMSMS, which provides higher confidence on using Cygnus commercial ELISA kit for HCP quantification of all in-house developed cell lines. In addition, LCMSMS identified a number of HCP as product quality related biomarkers and those biomarkers can be used in multiplex assays for bioprocess monitoring and improvement. P-106-M Development and NDA-Level Validation of a Real-Time PCR Procedure for Detection and Quantification of Residual E.coli DNA in Biopharmaceutical Products Dan Papa; Pedro Morales; Michael Sadick Aptuit, Kansas City, MO USA Escherichia coli (E. coli) has been commonly used for the production of biopharmaceuticals. Among the impurities that must be monitored in biopharmaceuticals is residual host-cell DNA (HCD). This study describes the development and subsequent NDA-level validation of a real time PCR procedure developed in response to a client’s need to improve the sensitivity of detection and quantification of residual E. coli HCD in their drug product, previously determined via total DNA threshold assay. This approach involved the design of a primer set capable of priming and amplifying a very short sequence (122 bp) inside the 16S rRNA – Glu-tRNA region of rrnB operon of E. coli, using TaqMan chemistry. Based on preliminary end-point PCR tests, the visual detection limit was 0.001 pg genomic DNA (gDNA). The procedure was successfully developed, optimized and validated for use in commercial drug product release. The limit of detection (LOD) was calculated to be 0.050 pg gDNA, whereas the limit of quantification was calculated to be 0.500 pg gDNA. The assay showed accuracy and precision (both repeatability and intermediate precision) with DNA content of even 10 pg and below. Thus, the real-time PCR procedure has proven to be a very sensitive and powerful tool for cGMP quantitative detection of E. coli HCD.

81

NOTES:

82

P-107-M A Generic Assay for Host-Cell Proteins in Therapeutic Proteins using LC/MS Methods Catalin E. Doneanu1; Alex Xenopoulos2; Keith Fadge1; Jim Murphy1; Martha Stapels1; St John Skilton1; Weibin Chen1 1Waters Corporation, Milford, MA USA; 2EMD Millipore Corporation, Bedford, MA USA Assays for identification and quantification of host-cell proteins (HCPs) in biotherapeutic proteins over 5 orders of magnitude in concentration are presented. The HCP assays consist of two types: HCP identification using comprehensive online two-dimensional liquid chromatography coupled with high-resolution MS (2D-LC/MS), and high-throughput HCP quantification by LC-MRM method. The former is described as a ‘discovery’ assay, the latter as a ‘monitoring’ assay. In the discovery assay, biotherapeutic protein (e.g. mAbs) digests were fractionated using reversed-phase (RP) chromatography at high pH (pH 10) by a step gradient in the first dimension, followed by a high-resolution separation at low pH (pH 2.5) in the second dimension. A quadrupole time-of-flight mass spectrometer was used to simultaneously detect the eluted peptides and their fragments by alternating the collision cell energy between a low and elevated energy state (MSE methodology). The MS data was used to identify the HCP proteins in the mixture. In the monitoring assay, the same dataset was mined to develop target peptides and transitions for monitoring the concentration of HCPs on a triple quadrupole mass spectrometer in a high-throughput manner (20 min LC-MRM analysis). The analytical methodology was applied to the identification and quantification of low-abundance HCPs in six samples of a recombinant mAb (PTG1). Thirty three HCPs were identified from six PTG1 samples among which 21 HCP isoforms were selected for MRM monitoring. The absolute quantification of three selected HCPs was undertaken on two different LC-MRM platforms after spiking isotopically labeled peptides in the samples. The results show that the assays provide detailed valuable information to understand the relative contributions of purification schemes to the nature and concentrations of HCP impurities in biopharmaceutical samples, and the assays can be used as generic methods for HCP analysis in the biopharmaceutical industry. P-108-M Size Separations of Proteins: What We Need to Achieve It Keeley Mapp2; Linda Lloyd2; James Martosella1; Phu Duong1 1Agilent Technologies, Wilmington, DE USA; 2Agilent Technologies UK, Shropshire, United Kingdom The requirement for precise and accurate information relating to protein molecular weight, size, aggregation and degradation varies according to the application. In drug discovery and process monitoring there is a requirement for speed but in QA/QC of final product then the accuracy of the quantitation will be paramount. It is essential when looking at the choice of column and the method parameters that the objective of the separation is clearly defined and the parameters that control the speed and resolution of the separation are fully understood. In the work presented here we use a set of protein standards that span the typical range of characteristics of recombinant biopharmaceuticals – from antibodies to small globular proteins, to determine the influence on resolution of the selected parameters.

83

Parameters investigated include: 1. Choice of SEC media, pore size and pore size distribution, particle size and surface chemistry. 2. Columns dimensions. 3. Salt content of eluent. 4.Linear velocity. The results of our experimental investigation shows the preferred options for SEC separations based on the size and characteristics of the protein and enables recommendation for the method parameters. NOTES:

84

P-109-M Charge Variant Analysis of Monoclonal Antibodies by pH Gradient Separation on Cation Exchange Columns Andrew Coffey2; Linda Lloyd2; James Martosella1; Phu Duong1 1Agilent Technologies, Wilmington, DE USA; 2Agilent Technologies UK, Shropshire, United Kingdom Monoclonal antibodies are complex recombinant proteins with remarkable therapeutic potential. IgG has a molecular weight of around 150-160kDa, equating to around 1330 individual amino acids (approximately 450 amino acids per heavy chain and 215 amino acid per light chain). Charge variants can arise from a variety of side reactions and may include loss of C-terminal lysine, deamidation and sialylation (making the molecule more acidic), or succinimide formation and amidation of COOH side chains (making the molecule more basic). Separation of these charge variants is often difficult to achieve with conventional salt gradient elution on cation exchange columns. Here we describe methods to generate pH gradient elution profiles that can lead to improved resolution for monoclonal antibody characterization and furthermore can be applied to high speed (< 5 minute) separations on short 5cm columns packed with 3µm or sub-2µm particles. P-110-M Recent Developments in Alternative Systems for Amino Acid Analysis Thomas Wheat; Margaret Maziarz; Patricia McConville Waters Corporation, Milford, MA USA Accurate, precise, and robust amino acid analysis is required as fundamental information in several kinds of investigational projects as well as in routine QC and process monitoring. In biochemical structural analysis, the amino acid composition of hydrolyzed proteins is used to confirm sequence and to determine extinction coefficients. In contrast to measurements of protein bound amino acids, free amino acids are important indicators of metabolic status and genetic origin. Such analyses are required for diverse applications such as optimizing cell culture conditions in the production of biopharmaceuticals as well as process monitoring in industrial scale fermentations. The demand of these applications led to the invention of amino acid analyzers and the continuous improvement of the technology over several decades. A turnkey, total application solution was developed. The technique based on a well-characterized derivatization protocol followed by a high resolution chromatographic separation. The amino acids are derivatized with 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate (AQC) under largely aqueous conditions. The derivatives are separated using UltraPerformance Liquid Chromatography® (UPLC®). The enhanced resolution of UPLC ensures a robust chromatographic method that reliably gives accurate and precise measurements. This technique has been widely adopted over the last five years. This solution has now been extended to the most modern instrument configurations and detection protocols. The testing of this system for linearity, accuracy, and precision will be discussed. The use of the application solution will be illustrated in analyzing protein structure and in monitoring cell culture nutritional status. The continuous enhancement of this application solution ensures the most convenient and efficient generation of reliable amino acid analyses.

85

NOTES:

86

P-111-M Quantitative Method for Sub-visible Particles using UV Spectroscopy Rajiv Nayar and Mitra Mosharraf HTD Biosystems Inc., Pleasanton, CA USA Purpose In recent years, a great deal of attention has been given to the characterization of sub-visible particles in biopharmaceuticals because of product quality concerns and immunogenicity potential of protein aggregates. A number of techniques have been adopted in evaluating the identity, size, nature of the particles, each with their limitations. It has been a challenge to develop a quantitative method for measuring the relative amounts of sub-visible particles that can be utilized in a QC environment. We have evaluated a straight-forward spectroscopic method to characterize the presence and relative amounts of sub-visible protein aggregates using various proteins. The purpose was to show the utility of such measurements to characterize changes in the protein solution by having an Aggregation Index value for a drug product. Methods Different protein solutions were stressed thermally to induce various levels of sub-visible aggregates. A UV diode-array spectrometer was employed to analyzed protein solutions with various levels of aggregates. The amount of aggregates is calculated using a simple equation that assigns an Aggregation Index (A.I) value to a protein formulation.

A.I. = 100 X (As,280/( A280- As,280)). Where, A280 is the absorbance at 280 nm and As,280 is the apparent absorbance from light scattering at 280 nm extrapotated from the long wavelenghts prior to the onset of true electronic absorption using the following equation.

As,280 = 10 (2.5log A320

– 1.5log A350

) Dynamic light scattering technique was used for determination of the average hydrodynamic radii of the aggregates and 90 oC light scattering intensity of the samples. Results The results showed that there was a direct correlation between the A.I values, the average hydrodynamic radii of the protein aggregates and 90 oC light scattering intensity. Non-aggregated solutions gave A.I values around 1-2. Conclusion These studies suggest that by utilizing a simple UV measurement, aggregation in protein formulations can be measured in conveniently and reproducibly using standard UV/Vis spectroscopy. This technique is highly sensitive and able to detect sub-visible particles. Finally, this method has the potential to be used in a QC environment if validated for a specific protein drug product.

87

NOTES:

88

P-112-M Use of a Strong Cation-Exchange Stationary Phase for the Analysis of Charged Isoforms for a Commercially Available Antibody Jeffrey Kakaley1; Takashi Sato2; Ernest Sobkow1; Jeff Wilson1 1YMC America, Inc., Allentown, PA USA; 2YMC Co., Ltd., Kyoto, Japan Ion exchange chromatography (IEC) is an important tool for evaluating the extent of heterogeneity due to charged isoforms in large molecule pharmaceutical and antibody drugs. Weak cation-exchange (WCX) columns using 10 micron particles have been considered the gold standard for these types of analyses. This work details an analytical comparison between a strong cation-exchange (SCX) resin versus the conventional WCX stationary phase for the analysis of a commercially available antibody drug and discusses additional benefits associated with employing the SCX resin. P-113-M MAbPac SCX 3 and 5 μm Particle Phases for Monoclonal Antibody (MAb) Variant Analysis Srinivasa Rao; Yuanxue Hou; Hongmin Zhang; Yury Agroskin; Chris Pohl Thermo Fisher Scientific, Sunnyvale, CA USA Monoclonal antibodies (MAbs) undergo several post-translational modifications including oxidation, deamidation, and truncation. Manufacturing of MAbs and subsequent stability testing procedures involve routine analysis and monitoring of the impurities resulting from these modifications. Earlier, we introduced Thermo Fisher Scientific MAbPac™ strong cation-exchange phase (MAbPac SCX) based on 10 μm particles for the characterization of heterogeneity of MAbs. It is a complimentary addition to the existing Thermo Scientific ProPac™ WCX-10 column that provides orthogonal selectivity for MAb charge variant analysis. Both ProPac WCX (Carboxylic acid functionality) and MAbPac SCX (Sulfonic acid functionality) stationary phases are based on nonporous, highly cross-linked styrenic e polymeric media. They differ in proprietary hydrophilic coatings as well as functional groups and grafting processes. Now, we are introducing the new MAbPac SCX columns with 5 μm and 3 μm particle phases for MAb variant analysis. The main objective for this development was to produce high-resolution columns comparable to the MAbPac SCX-10, 10 μm columns (4 × 250 mm), but with a much faster analysis time. The length advantage of a 250 mm column for producing high resolution is compensated by a smaller particle size in a 4 × 50 mm column. Since the column is shorter, analysis time can be reduced and throughput can be increased. Using these small particle columns several fast applications including acidic, basic variant analysis, and lysine truncation variant characterization of a MAb were performed. Both salt and pH gradients were used in these experiments. In addition, a longer length version (4x250 mm) of the 5 μm phase columns are introduced for very high resolution purposes that takes advantage of both length and smaller particle size for superior resolution as compared to the 10 μm phase columns. Ruggedness of both the 3 and 5 μm phases is also demonstrated.

89

NOTES:

90

P-114-M Advantages of Using 125Å Pore Size, Sub-2µm, Size-Exclusion Particles for the Analysis of Peptides and Small Proteins Paula Hong; Stephan Koza; Kenneth Fountain Waters Corporation, Milford, MA USA Size exclusion chromatography (SEC) is typically used to measure the product of interest, aggregates, and potentially other size variants present in biopharmaceuticals. Unlike reversed-phase and ion-exchange chromatographic separations the elution volume and resolution of a true size-exclusion separation, where the mobile phase composition has been appropriately optimized to minimize protein-column interaction, is predominantly dependent on the packing materials (pore size, pore volume, and particle size), geometry of the column, and performance of the chromatographic system used. Size-based HPLC separations have traditionally been performed at low pressure conditions with 5µm and larger silica-based packing materials. However, new advances in a 125 Å pore size, sub-2µm hybrid silica packing material and Ultra Performance Liquid Chromatography (UPLC) instrumentation have allowed faster, higher sensitivity, and higher resolving separations of peptides and small proteins to be achieved while at the same time greatly reducing waste-stream volumes. The advantages of using a packing-material with the best suited physical characteristics of pore and particle size for the size-exclusion chromatography separation of peptides and small proteins, including the pharmacopoeial insulin method, will be demonstrated. Additionally, the investigation of column geometry, as well as other physical variables including flow rate and chromatographic system performance will also be presented P-115-M Ultra-Fast Reversed-phase UHPLC Separations for the High Resolution Analysis of Intact and Reduced Monoclonal Antibodies. James Martosella1; Phu Duong1; Brian Bidlingmeyer1; Linda Lloyd2 1Agilent Technologies, Wilmington , DE USA; 2Agilent Technologies UK, Shropshire, United Kingdom Monoclonal antibodies (MAbs) are critically important to drug therapies. Today, they represent the largest class of therapeutic drugs made by the biotechnology industry and will continue to play a significant role in the future of all pharmacological interventions of disease. Separation and purification by reversed-phase chromatography is gaining in popularity for monitoring purity and stability during mAb manufacturing and formulation. The work described herein highlights a new sub 2um UHPLC Zorbax family of wide-pore reversed-phase columns that demonstrate utility for ultra-fast analysis of mAb’s. Specifically, we have optimized chromatographic conditions and methods for the repeated high resolution analysis of intact and reduced monoclonal antibodies achieved during very fast and efficient column run times. Alternate phase choices of C18, C8, C3 and Diphenyl provided unique selectivity options for mAb’s differing in hydrophobicity, and thus provided a wide range of separation capabilities. Mobile phase, flow rate and elevated temperature investigations for each phase were examined and optimized for resolution, speed and post run recovery. Additionally, column lifetime stability at elevated operating pressures (1000 bar) and low pH will be demonstrated.

91

NOTES:

92

P-116-M UHPLC Optimizations for the Rapid Separation of Reduced Monoclonal Antibodies using Sub 2um C3, C8 and Diphenyl Selectivities. James Martosella1; Phu Duong1; Linda Lloyd2 1Agilent Technologies, Wilmington , DE USA; 2Agilent Technologies UK, Shropshire, United Kingdom Monoclonal antibody (mAb) is a glycoprotein comprised of two identical copies of heavy chains (50kDa) and two identical copies of light chains (25kDa) attached through disulfide bridges. Fast and efficient characterization of these chains and chain isoforms are becoming increasingly desired, and critical, for mAb characterization and screening methods. In this work, we have evaluated ZORBAX Rapid Resolution High Definition (RRHD) 300A sub 2um columns under carefully optimized chromatographic conditions to demonstrate utility for ultra-fast analysis of reduced monoclonal antibodies. With use of ZORBAX StableBond C3, C8 and diphenyl phase chemistries, elevated operating temperatures and optimized gradient conditions we have achieved ultra high resolution of the light and heavy chain antibodies in an extremely short elution time, and demonstrated unique selectivity capabilities. The separations were performed under mobile phase compositions comprising of various ion pairing additives to provide flexibility for LC/MS user preference. The columns displayed exceptional tolerance to high back pressure increases (>1000 bar) and provided reproducible column operation under acidic conditions and high temperatures. Additionally, column recovery was evaluated during and after repeated injection sequences. P-117-M Peptide Separations Using Size Exclusion Chromatography Haiying Chen Sepax Technologies, Inc., Newark, DE USA Size exclusion chromatography (SEC) has been applied successfully to separate different sizes of proteins under native conditions. Different pore sizes have been developed to accommodate different ranges of molecular weight of biological samples. In order to apply the size exclusion chromatography to peptides under 10,000 Da, a few limitations have to be overcome. Even very small peptides can exist in different conformations and exhibit secondary structures. Therefore, peptides tend to adsorb to column matrices by ionic and hydrophobic interactions. High salt concentrations, denaturing agents, and organic additives will minimize such interactions, thus enabling the separation of peptides according to their molecular weights. Sepax Zenix SEC columns are based on uniform, hydrophilic, and neutral nanometer thick films chemically bonded on high purity and mechanically stabilized silica. Zenix SEC-80 is specifically designed for small protein and peptide separations. Its phase is similar to the other Zenix SEC with the same particle size 3µm, but different pore size at 80 Å. Here we present separation of four peptides bradykinin (1,060 Da), angiotensin I (1,297 Da), glucagon (3,483 Da) and insulin (5,778 Da) under different separation conditions. Separation of E. coli tryptic digest on Zenix-80 is also investigated with different mobile phase conditions.

93

NOTES:

94

P-118-M Exploration of pH-Gradient Ion-Exchange Chromatography High-Resolution Protein Separations in Biotechnology and Proteomics Gurmil Gendeh Thermo Fisher Scientific, Sunnyvale, CA USA Ion-exchange chromatography (IEC) is a versatile separation technique for profiling the charge heterogeneity of biotherapeutic proteins, including monoclonal antibodies. Despite good resolving power and robustness, ionic-strength-based ion-exchange separations are product specific and time consuming to develop. Although salt gradients are more commonly applied, the utilization of pH gradients can provide significant advantages such as: 1) improved separation resolution; 2) lower salt concentration in collected fractions; and 3) the possibility to correlate the protein isoelectric point (pI) data with elution profiles. Recently, the application of pH-gradient IEC has been described for the separation of standard proteins and monoclonal antibodies. The work shown here describes the application of pH-gradient IEC as compared to salt-gradient IEC for the separation of proteins from various sources. Highresolution separations of a monoclonal antibody and its isoforms were achieved using a new, nonporous, strong cation-exchange resin. Results were compared to those obtained with salt-gradient IEC. Complex protein mixtures typically found in proteomics were separated with pH-gradient IEC. Developed methodology was validated for pH profile shape and precision, retentiontime precision, peak capacity, and robustness towards sample solvent composition. P-119-M Two-dimensional SEC/RP capillary LC for Top-down Proteomics Analysis Gurmil Gendeh Thermo Fisher Scientific, Sunnyvale, CA USA Top-down proteomic liquid chromatography (LC) combined with mass spectrometry (MS) is used to investigate protein structure and post-translational modifications (PTM) through liquid-phase separation and MS analysis of intact proteins. The advantage of top-down over bottom-up approaches is the absence of a proteolytic cleavage step. This keeps all protein information within one molecule and does not multiply sample complexity. However, working with intact proteins introduces challenges for the analytical technology used. Proteins are more difficult to separate than peptides using standard LC methods and more difficult to study by MS. Recently, different electron-based dissociation techniques, e.g., electron-capture detection (ECD) and electron-transfer dissociation (ETD), have been successfully applied to intact proteins to obtain sequence and PTM information. The successful application of these techniques in tandem with MS for protein mixtures requires a separation step. But powerful LC methods for proteins are published infrequently, and need to be developed and optimized for direct interfacing with MS.

95

This study describes a two-dimensional (2D) capillaryscale LC method for the separation of intact proteins. Proteins are separated and fractionated on a capillary size-exclusion chromatography (SEC) column. Next, proteins contained in the size fractions are separated on reversed-phase (RP) capillary monolithic columns prior to UV and MS detection. The method has been optimized with respect to SEC fractionation and RP gradient conditions for standard proteins and complex mixtures. NOTES:

96

P-120-M LC/MS Analysis of Antibodies using a Novel High Resolution Quadrupole Time-of-Flight Mass Spectrometry Platform Asish Chakraborty1; StJohn Skilton1; Martin Palmer2; Keith Richardson2; Jason Wildgoose2; Kevin Giles2; Martin Green2; Weibin Chen1

1Waters Corporation, Milford, MA USA; 2Waters MS Technologies, Manchester, United Kingdom The pipeline for biotherapeutics is growing rapidly as pharmaceutical organizations shift their focus from small molecule drugs to biotherapeutic drugs. The efficient characterization of antibody drugs is increasingly important to both regulatory agencies and to pharmaceutical companies to ensure the safety and efficacy of biotherapeutic products. Researchers want intact protein characterization by mass spectrometry to open new doors in the characterization of biotherapeutics. Isotopic resolution in combination with exact mass measurement allows direct detection of modifications like oxidations, deamidations or reduction of disulfides at the intact protein level. In this presentation, a high-sensitivity quadrupole/time-of-flight system (SYNAPT G2-S) with novel StepWave ion optics was used to resolve the isotopes of monoclonal antibody subunits. Sample delivery to the mass spectrometer employed two drastically different chromatographic conditions: a generic, high-throughput denaturing size-exclusion chromatography (dSEC) method that can be applied to profile monoclonal antibodies and a high-resolution reversed-phase (RP) separation for the detection of modifications at the intact or subunit levels. The use of dSEC directly coupled to a quadruple time-of-flight mass spectrometer for the mass analyses of several antibody drugs (Trastuzumab, Rituximab, Natalizumab, Infliximab) provided greater detail and productivity. BEH SEC column with sub-2-micron packing materials showed significant improvements in size-based separation. Unlike RP columns, BEH SEC columns showed no memory effect. Baseline separation of the light chain and heavy chain was achieved for all antibody samples. The dSEC, using 30% acetronitrile with acidic modifiers as mobile phase, directly coupled to ESI-MS provided a routine analysis method for antibody glycoprofiling, sequence confirmation (isotope resolution), and molecular weight measurement. High-resolution MS detection was able to istopically resolve proteins up to 30 kDa. Identities of protein and mAb-subunits were confirmed by deconvoluting the mass spectra using both maximum entropy and BayesSpray deconvolution algorithms. The LC-TOF MS approach presented here facilitates rapid intact protein characterization. P-121-M Optimizing MS Workflows for Complete Characterization of Therapeutic Monoclonal Antibodies Eric Johansen; Jenny Albanese; Christie Hunter AB SCIEX, Foster City, CA USA Monoclonal antibodies (mAb) are becoming major target-oriented biotherapeutics to treat an array of human diseases. Current therapeutic monoclonal antibodies are immunoglobulin G (IgG)1 derivatives. These antibodies have typically been produced by mammalian cell culture using Chinese hamster ovary (CHO) cells, mouse NSO, or SP2/0 plasma cell lines. In recent years, mass spectrometry has emerged as

97

a superior method for the characterization of the heterogeneity of these proteins due to post-translational modifications (PTMs), sequence variations generated from proteolysis or transcriptional / translational errors, and degradation products which are formed during processing or final product storage. Here we describe analytical methods for complete characterization using LC/MS on the TripleTOF™ 5600 system. The high mass accuracy and resolution of the TOF MS data combined with the sophisticated protein reconstruction tools provides high quality protein molecular weight information with accurate mass assignment, even for highly heterogeneous antibodies with overlapping peaks. In addition, high resolution accurate mass MS/MS can be acquired at high speed to enable comprehensive peptide mapping experiments on these IgG proteins. NOTES:

98

P-122-M Quantitation of Mutated Protein X in Human Population Using Triple Quadrupole (QQQ) Zhi Zeng; Shiaw-Lin Wu; Barry L. Karger Barnett Institute, Northeastern University, Boston, MA USA LC-MS/MS based methods using triple quadrupole (QQQ) were developed to characterize and quantitate protein mutants, which are only one amino acid different from the wild type protein X. RNAi therapeutics was developed to target the disease-causing genes and hence reduced the expression of mutated X circulating in blood. The efficacy and lifetime of the RNAi can be evaluated by a pharmacodynamics (PD) study. As one of the most stable proteins, protein X exists as a homotetramer, showing more resistant to denaturation and digestion by enzymes when compared to other serum proteins. Conditions such as temperature and time were optimized to increase the yield of monomers and subsequently the enzyme digestion efficiency. The first method combining SDS-PAGE, in-gel digestion and LC-MS/MS (QQQ) analysis was able to separate protein X from other serum protein and greatly improve the detection limits. The second method includes in-solution digestion and LC-MS/MS (QQQ) analysis was less sensitive but with higher through put. Multi-enzyme digestion approaches (Lys-C, trypsin or Glu-C) were used to achieve the characterization of the signature peptide derived from each mutant, and limited misscleavages. The absolute quantitations of the mutants were based on the external standard curve method using 15N labeled mutants as internal standards (ISs). As a result, these two LC-MS/MS based methods with different merits could be applied in the further PD study of the RNAi therapeutics. This is difficult to be achieved by ELISA due to the challenge in the development of antibodies that could differentiate mutants with only one amino acid difference from their WT. P-123-M Stability of Trisulfide Modification in an IgG1 Monoclonal Antibody (mAb) Biopharmaceutical L. Zang1; K. Lucas2; J. Fernandez3; A. Weiskopf1 1Biogen Idec, Cambridge, MA USA; 2Biotech Primer Inc., San Diego, CA USA Trisulfide modification was found to be one of the common modifications to natural and recombinant monoclonal antibodies of all immunoglobulin G types1. This modification occurs most predominantly to the heavy chain-light chain (HC-LC) disulfide linkage. Exposure of monoclonal antibody biopharmaceuticals to H2S during protein production or purification can potentially lead to increased level of trisulfide content in the final product. An IgG1 type mAb biopharmaceutical was produced with approximately 2.9% trisulfide in its HC-LC disulfide linkage and the trisulfide content was monitored over 9 months in a stability study conducted at three different temperatures, 2-8°C, 25°C and 40°C. The mAb was found to have consistent trisulfide content at the HC-LC disulfide when stored at 2-8°C, while at elevated temperatures, especially at 40°C, the relative percentage of trisulfide at HC-LC linkage slightly increased over time. The increase was suspected to be due to a small amount of H2S present in the mAb solution, or produced as a byproduct of β-elimination reactions that form thioether bonds between HC and LC during the storage at elevated temperatures2. In order to verify the correlation between trisulfide increase and free

99

H2S content in mAb solution during storage at elevated temperature, a 3 month stability study of the mAb at 40°C was performed with formulation of the mAb in two different buffers with various pHs and salt concentrations. The results indicated that sodium phosphate buffer, which reportedly scavenges free H2S3 more effectively, was capable of slowing down the trisulfide increase and when added at a concentration of 100 mM, essentially stopping the trisulfide increase over time. In comparison, the N2 gas sparge, also reported to reduce the presence of the trisulfide species4, did not appear to affect the trisulfide content increase over time. References: (1) Gu, S.; Wen, D.; Weinreb, P. H.; Sun, Y.; Zhang, L.; Foley, S. F.; Kshirsagar, R.; Evans, D.; Mi,

S.; Meier, W.; Pepinsky, R. B. Anal Biochem 2010, 400, 89-98. (2) Cohen, S. L.; Price, C.; Vlasak, J. J Am Chem Soc 2007, 129, 6976-6977. (3) Hemmendorff, B.; Castan, A.; Persson, A. US2007/0232792 A1 2007. (4) Becker, P.; Christensen, T. US2010/0160236 A1 2010. NOTES:

100

P-124-M Monoclonal Antibody Fragment Separation and Characterization Using Size Exclusion Chromatography Coupled with Mass Spectrometry Haiying Chen; Katherine Mclaughlin Sepax Technologies, inc., Newark, DE USA Monoclonal antibodies have increasingly become a major part of protein therapeutics. Monoclonal antibody fragments (such as Fab and F(ab’) 2) offer advantages over using intact MAbs such as reducing nonspecific antigen binding from Fc. Size exclusion chromatography has been widely used in protein analysis. Aggregates, monomers and degradation products of monoclonal antibodies can be separated on size exclusion columns based on their molecular weights under native conditions. In general, protein native buffer conditions such as salts at neutral pH are not mass spectrometry friendly. In this study, we investigated antibody fragments such as heavy and light chains, Fab/Fc, and F(ab’) 2 using SEC separation. Heavy and light chains were also analyzed by SEC coupled with mass spectrometry with volatile mobile phases. The effect of different percentages of TFA, formic acid and acetonitrile in mobile phases on the antibody fragments separation was also explored. P-125-M Affinity Purification and Mass Spectrometric Analysis of Monoclonal Antibodies Using Online Two Dimensional UPLC Sean McCarthy; Thomas E. Wheat; Ying Qing Yu; Jeffery R. Mazzeo Waters Corporation, Milford, MA USA Identification, characterization, and quantitation of monoclonal antibodies (mAbs) is required at many stages of biopharmaceutical research and development. The analytical tools to address these issues are liquid chromatography coupled with UV or mass spectrometry (MS). Both techniques can be compromised by interferences arising from the sample matrix, including high salt concentrations, other proteins, or the components of cell culture media. Affinity chromatography with immobilized Protein A can be used to isolate the antibody from a complex matrix while reversed phase LC is useful for introducing a salt free, concentrated sample into an MS ion-source. Affinity chromatography is often used to determine protein titer by UV absorbance using peak area due to its high specificity, linearity, and reproducibility. Mass spectrometric analysis allows the practitioner to determine the mass and glycoprofile of the analyte. While each of these methods can be performed very rapidly, they are generally accomplished by separate chromatographic systems since the eluents used for affinity chromatography are incompatible with mass spectrometry, and vice versa. In addition, each analysis requires different sample preparation protocols which can increase the time necessary for analysis. In this presentation we describe the use of an ACQUITY UPLC system with 2D technology. The 2D system allows for simultaneous purification and quantitation of monoclonal antibodies by Protein A affinity chromatography and determination of mass profile by MS analysis after desalting on a short reversed phase column. Our data demonstrates the linearity and reproducibility of response for mAbs from complex matricies. We also discuss carryover and chromatographic reproducibility observed when performing protein A affinity chromatography with the 2D system. 2D UPLC system optimization for peak collection, desalting, and MS detection of antibody samples is also discussed.

101

NOTES:

102

P-126-M Characterization of Unexpected By-products in a Recombinant TAU Protein by Capillary LC Coupled to MALDI-Topdown-Sequencing (LC-MALDI-TDS) Arndt Asperger2; Catherine Evans1; Branislav Kovacech3; Andrej Kovac3 1Bruker Daltonics GmbH, Faellanden, Switzerland; 2Bruker Daltonik GmbH, Bremen, Germany; 3Ins. of Neuroimmunology of Slovak Ac. of Sciences, Bratislava, Slovakia Compared to Edman sequencing, MALDI-TOFMS based top-down sequencing of proteins (MALDI-TDS) is much faster, cheaper, can deliver significantly longer sequence readout from both, protein N- and C-terminus, and has no limitations in case of terminally modifed proteins. These features make MALDI-TDS an extremely appealing method for the QC analysis of recombinant protein products, for instance biopharmaceuticals Direct MALDI-TDS analysis requires purified, homogeneous protein samples. However, coupling MALDI-TDS with upfront LC separation allows the analysis of mixtures of proteins. Since recombinant protein products often contain unexpected heterogeneities, LC-MALDI-TDS represents a promising method to characterize such undesired by-products. We describe the characterization by LC-MALDI-TDS of unexpected by-products occuring in a recombinant TAU protein. Initial MALDI-TOF analysis revealed at least two minor components being different from the expected protein mass. These sample constituents were separated using capillary LC. The LC fractions were deposited on a MALDI target pre-spotted with sDHB matrix. Subsequently, MALDI-ISD spectra of the LC separated compounds were acquired using a Ultraflex III MALDI-TOF/TOF. Data analysis was performed using BioTools. Using this LC-MALDI-TDS approach, one of the by-products, detected at 16.6kDa, was determined to be a truncated version of the expected TAU protein, containing the sequence range [1-161]. C-terminal ISD fragments yielded unambiguous evidence for the correct assignment of the truncation site. Another unexpected by-product, being approx. 130Da lower in mass than the expected protein product, was lacking the N-terminal methionine residue. This was concluded from the assignment of the respective N-terminal fragments in the ISD spectrum. We have demonstrated here that compared to alternative approaches e.g. Edman sequencing or typical bottom-up mass spectrometric approaches, the described LC-MALDI- TDS method was significantly faster and produced a minimum amount of data to generate the requested information. P-127-M Fast QC of Intact Monoclonal Antibodies Using MALDI-Top-Down Sequencing Anja Resemann1; Catherine Evans2; Rainer Paape1; Lars Vorwerg1; Detlev Suckau1 1Bruker Daltonik GmbH, Bremen, Germany; 2Bruker Daltonics GmbH, Faellanden, Switzerland Antibodies are playing an important role in drug development in the last years. The function of therapeutic monoclonal antibodies (mAbs) is directly depending on their structure including terminal modifications like C-terminal lysine excision and N-terminal pyroglutamylation of the heavy chain.

103

Here, we describe a method for very fast and simple validation of the terminal sequences of heavy and light chain of intact monoclonal antibodies using MALDI-Top-Down-Sequencing (MALDITDS) using in-source decay (ISD). Intact mAb (IgG) in a concentration of 1 mg to 5 mg/ml was directly mixed with matrix solution without prior reduction and alkylation or separation of the different chains. We used 1,5-diaminonaphtalene (DAN) as matrix because of its reductive and ISD enhancing properties. The samples were mixed with matrix solution and dried at ambient air on the MALDI sample plate. In a second experiment, the same samples were prepared using sDHB (super DHB), which is known to be a good ISD matrix but without reduction capacity The samples in sDHB didn’t generate any ISD fragmentation due to the intact disulfide crosslinks in IgG. In DAN, the same samples generated rich ISD spectra containing fragments from both the heavy and the light chain of the antibodies. The ISD spectra permitted reading through the cysteine residues in the sequences implicating the disulfide bridges were reduced by DAN. Up to 60 residues of the different chains of various mAbs were matched and validated, respectively. A dedicated software tool allowed a very fast interpretation of the spectra. Overall the method takes only a few minutes from sample preparation, spectrum acquisition and processing to the result in form of sequence annotations in the ISD spectra. NOTES:

104

P-128-M A High-performance Benchtop Quadrupole-Orbitrap LC-MS/MS Workflow Solution for Intact Monoclonal Antibody Characterization Zhiqi Hao; Yi Zhang; David Horn; Seema Sharma; Andreas Huhmer Thermo Fisher Scientific, San Jose, CA USA Monoclonal antibodies (mAbs) are increasingly developed and utilized for the diagnostic and therapeutic of diseases including cancer. Due to the heterogeneity of mAb products, thorough characterization is necessary for their reproducible as well as safe production. Among the analytical tools used for the analysis of therapeutic mAb, mass spectrometry has become more and more important in providing valuable information on various protein properties, such as intact mass, amino acid sequence, post-translational modification including glycosylation form distribution, minor impurities due to sample processing and handling and high order structure, etc. In this study, a high resolution LC-MS based workflow solution was developed and evaluated for robust, accurate and comprehensive intact mAb characterization using the Q Exactive, a routine bench-top orbitrap mass spectrometer. Full MS spectra were analyzed using Protein Deconvolution software for intact mass determination and the results indicated that the high mass accuracy, stability and high resolution of the Q Exactive produced accurate and reproducible intact mAb mass measurement with a mass error of less than 10 ppm. To obtain sequence information of the intact mAb, a unique, high throughput HCD capability of the Q Exactive was used to generate high resolution, top-down spectra on intact light or heavy chain on the LC time scale. Analysis of the top-down MS/MS spectra using ProSightPC software presented extensive sequence coverage with a mass error of less than 5 ppm for fragment ions. Results from this study indicated that both precise mass measurement and extensive, high confident sequence information can be obtained for mAb using the same workflow solution. The fast chromatography, the superior resolution, mass accuracy and the high throughput MS/MS combined with the easy, accurate data analysis of this workflow solution provide a high-confident screening tool to accelerate biopharmaceutical product development cycles. P-129-M Quantification of a Model Protein Lysozyme and Peptide MUC5AC-13 in Serum Matrix Using LC-QTOF Vadiraja B. Bhat; Ning Tang Agilent Technologies, Santa Clara, CA USA Drug metabolism and phomacokinetics of biologics has been an increasingly interesting topic. LC-MS techniques can potentially provide more accurate understanding of the biotransformation of therapeutic proteins and complimentary information to ELISA methodologies. Although, using LC-MS/MS as the technology for small molecule DMPK has been established in the industry, using LC-MS as a qualititative and quantitative tool for large molecules DMPK have not been widely accepted. In this abstract, a model protein, Lysozyme and model peptide MUC5AC-13 were spiked into human serum. An analog of MUC5AC-13 was spiked in as an internal calibration standard. A high resolution Q-TOF mass spectrometry was used to acquire data. For the peptide, LOQ of 50pg on 2.1mm column and 4pg on 75um HPLC-Chip were achieved. Average RSD was 4.2% with three orders of magnitude linearity.

105

At the same time, MS/MS provided qualitative information for the validation the peptide sequence. For model protein lysozyme, linearity was achieved from 1 – 1000ug/mL in TOF scan mode. These results have shown excellent potential of Q-TOF as a Qual/Quant technique for biologic samples. NOTES:

106

P-130-M Heart-cutting Two Dimensional LC with QTOF Detection for the Determination of Peak Purity of Forced Degraded Insulin Dawn Stickle; Bob Giuffre Agilent Technologies, Wilmington , DE USA One critical aspect of drug safety involves determining suitable storage conditions for biological products by conducting stability studies under accelerated storage conditions. Prior to beginning a stability study one must develop stability indicating analytical methods or methods capable of detecting potential changes that occur during the study. In order to determine if one’s analytical method is stability indicating, the drug substance and drug product is subjected to forced degradation such as by acid, base, oxidation and light and the main drug peak is evaluated for peak purity. Typically the analytical method that indicates the impurity profile of a drug and that can be proven to be stability indicating is a HPLC method using UV detection. Often times the peak purity is determined by a comparison of UV spectra or by examining the peak using mass spectrometry. Although these techniques are effective at determining peak purity, they are not 100% reliable as low level impurities still may be not be detected under the main peak for various reasons such as many impurities can have similar UV spectra to the main peak or ion suppression in mass spectrometry. Another approach to determine peak purity would be to use a heart-cutting two dimensional LC technique, in which the main peak would be sent to a second HPLC column and subjected to further separation. If impurities are present under the main peak, the second dimension LC column can likely separate these impurities from the main species. Historically two dimensional LC techniques are not in routine use as they can be difficult to construct and troubleshoot. In this work, we will present a robust and easy to use heart-cutting two dimensional LC attached to a QTOF mass spectrometer detector for the peak purity determination of base degraded insulin. P-131-M Principle to Practice: Adapting MALDI-TOF-PMF for Protein Product Ddentity Testing in a Regulated Environment Edward Hoff; Lin Gan; Mary Zhu; Min Zhu; Viswanatham Katta Genentech, a Member of the Roche Group, South San Francisco, CA USA Peptide mass fingerprinting (PMF) is a powerful research tool for checking if an ‘unknown’ protein is already known in a sequence database. We have adapted this method for distinguishing closely related therapeutic recombinant antibodies (WCBP 2008, WCBP 2009). Here, we describe the practice of this identity method, to release pre-clinical and clinical protein proteins in a regulated environment. To date, the method has been validated for over 40 of pre-clinical and clinical proteins covering both antibody and non-antibody products.

107

Many aspects of the PMF method were simplified – digestion (denature, buffer exchange, digest with trypsin and then reduce the disulfides), sample spotting (with DHB matrix, no desalting), mass spectrometer (adequate resolution and accuracy across the mass range of interest), submission of spectrum for search (Mascot daemon), use of a proprietary protein database. Interpretation of search results was simplified using criteria such as minimum Mascot score for the top hit, score differential from the next hit, sequence coverage, etc. for pass/fail for specific identification. In addition, instrument suitability (MALDI-TOF resolution and mass accuracy), valid method criteria for the Reference protein were designed to make the method robust and simple in the hands of analysts with limited mass spec expertize. Building and maintaining the proprietary database containing ‘representative sequences’ of Genentech’s clinical proteins created some additional challenges, but the process ensured that backward compatibility could be easily established when a new protein enters the clinical pipeline. Eight analyst have been trained to perform the method. Our initial concerns about analysts, not previously trained in mass spectrometry, performing a mass spectrometry method were quickly relieved by the testing laboratory experience. In the first three years our MALDI-PMF based identity method was successfully practiced with over a 98% success rate. Generally the discrepancies related to sample preparation and documentation and not the technology. NOTES:

108

P-132-M Comprehensive Characterization of Therapeutic Monoclonal Antibodies by LC−MS Based Approaches Yi Wang Northeastern University, Boston, MA USA Comprehensive characterization of therapeutic monoclonal antibodies (mAbs) is critical to the biotech industry. Particularly, detections of functionality-sensitive modifications in recombinant mAbs, such as glycosylation and disulfides, have significant quality implications. In this report, a range of LC-MS based methods have been developed for different purposes: detections of primary sequence and modifications by on-line LC-MS analysis of reduced and digested monoclonal antibodies; use of an online LC-MS platform with ETD for the detailed characterization of disulfide linkages and scrambling; profiling glycans by a new chip-based LC-MS approach, mAb-Glyco-chip-Q-TOF. In the mAb-Glyco-chip, immobilized PNGase F and graphitized carbon are integrated together in the chip to on-line release the glycans for mass spectrometric analysis. The glycan distribution result generated from this new approach is compared to PGC chip (off-line release of glycans). Finally, innovator and biosimilar therapeutic mAbs are compared by these LC-MS analysis platforms. P-133-M Rapid Pre-Formulation Development Using iFormulate™ Plates Rajiv Nayar and Mitra Mosharraf HTD Biosystems Inc., Pleasanton, CA USA Purpose Protein formulation is still performed empirically in many laboratories, although efforts are made in using DOE approaches. We have designed a formulation development plate that can be used for pre-formulation studies. Here, we show the use of the iFormulate™ plate in performing pre-formulations studies of proteins using appropriate stability-indicating assays. Methods The iFormulate™ plate is a plate pre-loaded with formulations to investigate key formulation variables such as pH, ionic strength, buffer concentration, and stabilizer concentration. The formulations were generated using multivariate DOE methodology the used a minimum number of trials necessary for a valid statistical output. Variables evaluated were pH (5.0 – 8.0), NaCl concentration (0-200 mM), buffer concentration (10-50 mM), and sucrose concentration (0-10 wt%). Using a response-surface quadratic design, the iFormulate™ plate has 20 unique formulations with 5 replicates in a 96-well plate. Ovalbumin was utilized as a model protein at 1 mg/ml for formulating a protein with low aggregation potential. The protein formulations were stressed at 70 oC for 30 minutes to induce aggregation. The amount of aggregates is calculated using a simple equation that assigns an Aggregation Index (A.I) value to a protein formulation. A.I. = 100 X (Abs 340 nm/(Abs 280-Abs 340)). The results were analyzed using a DOE software (ECHIP) for an optimal formulation.

109

Results The results from the the iFormulate™ plate, show the critical formulation variable using the Pareto analysis, any interactions between the 4 variables investigated, and identified the optimal formulation as per the Product Target Profile. Using the iFormulate™ plate, an optimal formulation of ovalbumin with low aggregation potential was identified using only 5 mg of total protein. The results from the the iFormulate™ plate demonstrated that pH was the most critical variable . Conclusion Using a statistically designed multivariate formulation design, protein formulations can be rapidly developed using small quantities of protein. The formulation can also be rationalized using this approach by QbD principles. This technique has utility for quickly and rationally provide information for designing stable protein formulations. NOTES:

110

P-135-M Determination of Liposome Concentration by Light Scattering Intensities from DLS Mitra Mosharraf; Aryo Sorayya; Rajiv Nayar HTD Biosystems Inc., Pleasanton, CA USA Purpose Total lipid content in liposomes is typically determined using a phosphorus assay or by HPLC lipid analysis. The phosphorus assay employs toxic chemicals and requires harsh conditions to release the phosphorus molecule from the lipids before quantitation using a chemical reaction. The HPLC assay used for lipid analysis requires complex sample preparation and use of detectors that are not very accurate. We investigated the use of dynamic light scattering intensity to quantitate liposome concentration. Methods Liposomes of different lipid compositions, sizes, and charge were prepared. Liposomes were sized by extrusion through 100 nm, 200 nm, and 400 nm polycarbonate filters to generate liposomes of different sizes. The liposome preparations were diluted to various concentrations in PBS and analyzed using a DLS instrument. Both particle size and 90 oC light intensity were determined and correlations between liposome concentration and scattering light intensity were evaluated. Results Excellent correlations were observed between liposomal lipid concentration and 90 oC right angle light scattering intensity. In liposomes having a net neutral charge containing saturated lipid components, such as, DPPC, slightly lower correlations were observed due to liposome-liposome interactions. Conclusion Light scattering intensity measurements provide an excellent non-destructive method for obtaining the total liposome concentration. The method is highly reproducible, accurate, and has a wide dynamic range. The method is also insensitive to formulation components that may interfere in other lipid analytical methods. P-136-M Biopharmaceutical System Solution with UNIFI Assesses the Molecular Similarity among Candidate Biosimilars and an Innovator mAb Vera Ivleva; Ying Qing Yu; Hongwei Xie; Weibin Chen Waters Corporation, Milford, MA USA Today biopharmaceutical companies are challenged with the design of an efficient analytical strategy for detailed assessment of the structural comparability of the biosimilar and innovator products. Extensive characterization assures that the biosmiliar product is safe and meets regulatory compliance criteria. Here we demonstrate how the newly developed Biopharmaceutical System Solution with UNIFI addresses the challenges by integrating data acquisition, data processing, and analysis result reporting

111

into a seamless workflow for in-depth LC/MS characterization of therapeutic protein drugs. Fully integrated UPLC/TOF MS approach with UNIFI Scientific Information System allows routing analysis of a biosimilar product under development and enables quick and efficient data management. Comparability analysis between a candidate biosimilar and an innovator Rituximab monoclonal antibody (mAb) drug was performed at intact protein and peptide mapping levels. Mass discrepancy between the innovator drug and one of the biosimilar candidates was revealed by LC/MS intact protein mass measurement, and the mutation site was identified by analyzing the chymotryptic digest of the proteins using high-resolution UPLC separation and a data independent acquisition approach (MSE). Comparison of amino acid modification level of the two biosimilars was also performed, and the results were reported under the same method and data processing platform. The comprehensive biosimilar mAb drug characterization and sophisticated data management capability were proven to be a dependable integrated analytical platform for the strict control of a biopharmaceutical regulatory process. NOTES:

112

P-137-M Considerations in the Establishment of an Analytical Program for Biosimilar Development Clara R. Rangel; Greg W. Adams; George A. Koch Fijifilm Diosynth Biotechnologies, Cary, NC USA Numerous challenges exist for the analytical technical program manager of a biosimilar development program, all of which may have a major impact on the success of the program as a while. Our experience with a mAb biosimilar candidate has proven that a high level of organization up front is required to ensure that the project flows smoothly over the years of development to come. Establishment of a successful analytical biosimilar program will require thorough information gathering, cross-departmental risk assessment, solid establishment of assays and test panels, plans for testing a wide range of originator lots, extended characterization, comparative stability and degradation pathway determinations, and excellent control over sample management and data tracking. Data will be shown which demonstrate the importance of these elements of managing the analytical aspects of a successful biosimilar candidate. Historical knowledge of the originator manufacturing variability is key in understanding bandwidth and setting specifications for final product. The statistical requirements for the number of lots of originator must be balanced with broad historical data collection and on-going monitoring of variability. By thoroughly understanding this variability, more flexibility may be built into the development of a competitive manufacturing process. P-138-M Comparability Study of the Structure of Filgrastim and Mutants using the NMR Fingerprint Assay Derek J. Hodgson1; William Thach1; Geneviève Gingras1; Adrian Bristow2; Yves Aubin1 1Health Canada, Ottawa, Canada; 2NIBSC, Potters Bar, United Kingdom Filgrastim is the generic name for recombinant methionyl human granulocyte colony-stimulating factor (r-metHuG-CSF). It is produced in Escherichia coli (E. coli) in a non-glycosylated form. Filgrastim is marketed under the brand name Neupogen® by Amgen. Since this product has lost patent protection, many subsequent entry versions have been approved or are in the process of filing for market authorization throughout the world, including Canada. In order to be authorized as a subsequent entry product, the sponsors must demonstrate similarity with an approved product in Canada via an appropriate comparability exercise. Here we show that the NMR fingerprint assay can be used to assess the three-dimensional structure of the active ingredient in the formulated product from two difference sources as well as a comparator, the approved product Neupogen®. Recombinant metHuG-CSF was prepared in E. coli and isotopically enriched with 13C and 15N isotopes. Samples were analysed by NMR to study the effects of varying the pH, the concentration of excipients (sorbitol and polysorbate-80), the ionic strengths with several salts, and co-solutes. Spectra of mutants have been recorded to assess the sensitivity of the method to small structural changes. Finally, NMR spectra were recorded for Neupogen®, purchased at a local pharmacy, and a chemical reference standard from the European Directorate for Quality Medicine (EDQM).

113

The NMR fingerprint assay applied to Filgrastim provided residue specific information of the structure of the active ingredient of a product. In addition to current methods, the ability to assess the conformation with a high degree of resolution can greatly facilitate the comparability exercise. NOTES:

114

P-139-M Improved System to System Reproducibility with the iCE3 Jiaqi Wu ProteinSimple, Toronto, Canada During capillary isoelectric focusing proteins are focused at their isoelectric point. When performing IEF in a capillary (cIEF), two forces within the capillary must be controlled: electro-osmotic flow (EOF) and hydrodynamic flow. For reproducible cIEF both of these forces must be eliminated. EOF is suppressed by neutral coatings on the inner wall of the capillary. Elimination of hydrodynamic flow is dependent on the system design. Traditional capillary electrophoresis instruments minimize hydrodynamic flow by leveling the inlet and outlet vials. In imaged cIEF, specifically the iCE280, elimination of hydrodynamic flow depends on the autosampler. In any system small differences in hydrodynamic flow can result in differences in reported pI values from system to system, creating issues during method transfer between labs. In this poster we describe how the new iCE3 minimizes system to system variability and simplifies method transfer. P-140-M Carrier Ampholytes, Which One Do You Use? Jiaqi Wu ProteinSimple, Toronto, Canada Carrier ampholytes are the single most important component of a cIEF separation. Unfortunately all carrier ampholytes are not created equal. While carrier ampholytes are available from a number of manufacturers they have significant differences. Understanding those differences accelerates method development. Today three brands of carrier ampholytes are commercially available: Pharmalytes by GE, Servalyts by Serva and Biolytes by Bio-Rad. While pH gradients from all three are quite linear, they do differ in peak resolution and baseline noise. For example, Servalyts have higher resolutions for fusion proteins and proteins with heavy sialylation. In this poster we will discuss the effects of the carrier ampholytes on peak pattern, baseline noise and resolution. P-141-M Hands-Free cIEF Jiaqi Wu ProteinSimple, Toronto, Canada The well characterized biopharmaceutical requires an assessment of charge heterogeneity. Techniques like IEF gels, ion exchange chromatography, and traditional capillary IEF all have benefits but each one has its own set of challenges. Imaged cIEF combines the best of these three worlds by providing rapid

115

analysis, platform methods, and simple method development. For cIEF analysis, samples are pre- mixed with carrier ampholytes, pI markers, and other additives. Although imaged cIEF offers rapid analysis and high throughput, proteins can still experience degradation when exposed to these conditions for extended periods. On-Board Sample Preparation with the new iCE3 solves this problem. The system prepares the sample immediately prior to injection limiting sample exposure to cIEF buffers and preventing degradation. An additional benefit, automated sample preparation eliminates tedious pipetting. Simply load your samples and go. This poster presents the application of On-Board Sample Preparation to the analysis of proteins. NOTES:

116

Poster Abstracts

Poster Session Two

Tuesday, January 24, 2012 14:45 - 15:45

Grand Ballroom Foyer, Third Floor Pacific Terrace Foyer, Fourth Floor

P-201-T A Novel Automated System for Protein Aggregation and Particle Analysis Susan Darling ProteinSimple, Santa Clara, CA USA Characterization and quantification of sub-visible particle populations in protein formulations is a regulatory requirement during formulation development. Many protein formulations are fragile and highly sensitive to sample preparation and handling techniques. An automated sample introduction system is critical to provide consistent sample handling to minimize run-to-run and instrument-to-instrument variability. This poster presents the qualification of a novel auto-sampler solution for the Micro-Flow Imaging (MFI) technology. The system was evaluated for accuracy and repeatability over 48 sample analysis operations, and compared with results obtained using manual MFI operation. Results show robust and consistent system cleanliness levels between each analysis, and consistent run-to-run accuracy and repeatability for polystyrene bead and protein surrogate Sensitivity Reference Suspension (SRS) samples. P-202-T Characterization of Protein and Protein Aggregates using Temperature Controlled Hollow Fiber Flow Field-Flow Fractionation Soheyl Tadjiki; Trevor Havard Postnova Analytics Inc., Salt Lake City, UT USA Asymmetrical Flow Field-Flow Fraction (AF4) is a powerful separation and characterization technique for biological macromolecules, lipids, viruses and cell particulates. The separation in AF4 takes place in an open channel and is based on diffusion coefficient or hydrodynamic diameter of sample components. The AF4 channel is made of two parallel walls with a thickness of 0.0190-0.0500 cm. Sample components are pushed toward the semi-permeable lower channel wall by the separation field (cross flow) and transported along the channel at different flow velocities by a secondary flow (channel flow). The AF4 channel can also be made of a semipermeable ultrafiltration hollow fiber where the cross flow exits radially through the fiber wall and the channel flow is flowing axially along the fiber length. Resolution, reproducibility and sample recovery of the hollow fiber cartridge was studied using a

117

Bovine Serum Albumin (BSA) standard for more than 100 consecutive injections. The average retention time RSD of the monomer and dimer peaks was found to be below 1%. The hollow fiber flow FFF system was hyphenated with a dynamic light scattering (DLS) detector to characterize protein aggregates. BSA and gamma globulin standards were stressed by heat at a low pH buffer condition to induce aggregation. The aggregated samples were then analyzed by the hollow fiber flow FFF-DLS system. The results showed the separation of the large protein aggregates from a few tens to hundreds of nanometers. NOTES:

118

P-203-T Characterization of Aggregates in a Model Fc-Fusion Protein James Strand1; Jin Xu2; Chi-Ting Huang1 1Acceleron Pharma, Cambridge, MA USA; 2University of Massachusetts - Lowell, Lowell, MA USA Aggregation is one of the more common impurities observed in recombinantly produced biotherapeutics and has implications on productivity and purity. The mechanisms behind the aggregation of recombinant IgG protein therapeutics from mammalian expression systems have been extensively studied; however, less information is available on the mechanism of aggregation in recombinant Fc-fusion proteins. In an effort to better understand the mechanisms behind aggregation of glycosylated fusion proteins produced in CHO cells, we have examined the high molecular weight (HMW) forms of a model Fc-fusion protein present during production in cell culture media using a combination of analytical and biophysical techniques. We observed two different classes of HMW aggregates from ProA purified ECD-Fc fusion protein, one disulfide-linked, and the other mediated by hydrophobic interactions, each with different folds. Both populations display less processed glycan structures. P-204-T Behavior of Monoclonal Antibodies: Relation Between Second Virial Coefficient at Low Concentrations and Aggregation Propensity and Viscosity at High Concentrations Shuntaro Saito Daiichi Sankyo Co. Ltd., Hiratsuka-Shi, Japan In order to investigate relationship between second virial coefficient (B2) and viscosity and aggregation propensity of highly concentrated monoclonal antibody (MAbs) solutions, the intermolecular interactions of 3 MAbs solutions with varying pH were characterized according to B2 estimated by analytical ultracentrifugation sedimentation equilibrium with initial loading concentrations <10 mg/mL. Viscosity measurements and stability assessments of MAb solutions at concentrations higher than 100 mg/mL were conducted. B2 of all MAb solutions depended on solution pH and have qualitative correlation with viscosity and aggregation propensity. The more negative the B2 values, the more viscous the solution, acquiring increased propensity to aggregate. Solutions with B2 values of ~2 × 10−5 mL·mol/g2 acquire similar viscosity and aggregation propensity regardless of amino acid sequences; for solutions with negative B2 values, viscosity and aggregation propensity differed depending on sequences. Results suggest attractive intermolecular interactions represented by negative B2 values are influenced by surface properties of individual MAbs. In conclusions, B2 can be used, within certain limitations, as an effective indicator of viscosity and aggregation propensity of highly concentrated MAb solutions.

119

NOTES:

120

P-205-T Aggregates and Their Analysis---Some Perspectives and Lessons from Working with Over 250 Proteins John Philo Alliance Protein Laboratories, Thousand Oaks, CA USA Over the last 13 years our laboratory has worked on aggregation issues for hundreds of clients and with over 250 proteins. The purpose of this poster is (1) to try to share some perspectives derived from that experience, (2) to illustrate a range of aggregation and analytical method issues and how those sometimes relate to fundamental aggregation mechanisms, and (3) to point out some things that our clients often fail to do themselves or misunderstand. A particular focus will be examples from SV-AUC or SEC-MALS experiments. P-206-T The Effect of Thresholding on Particle Size and Count When Using Dynamic Fluid Imaging to Characterize Biologicals Lew Brown Fluid Imaging Technologies, Yarmouth, ME USA Dynamic Fluid Imaging (DFI) is being used increasingly as a method for characterizing sub-visible particulates in biologics. It overcomes the limitations of volumetric-based particle counters by being able to go beyond characterizing simple particle size and count based upon the assumption of spherical shape. DFI adds in the ability to make particle shape measurements, enabling differentiation of particle types in a heterogeneous mixture. In biologics, this is a critical ability, since it can separate intrinsic particles such as silicone droplets or air bubbles from protein aggregates, counting each separately. Additionally, DFI is much more discriminatory of transparent particles than volumetric techniques that rely on principles such as light obscuration. This discrimination and the ability to measure shape parameters beyond Equivalent Spherical Diameter (ESD) yields much more accurate sizing and concentration information. However, the measurements made in DFI systems are based upon a "thresholded" (binary) image, and the process used to determine particle boundaries in the image will greatly affect both the sizing and concentration calculations. This poster will show how the characterization of biologics using DFI is affected by the thresholding parameters used to create the binary images for measurements. Biologics samples will be analyzed with DFI using different thresholding parameters on the same data sets to closely examine both the sizing and concentration calculations. The resulting tabular data will clearly show the differences in results, while the accompanying images will show the same information in an easy to see visual format. This will help those using DFI to better understand the results they see, and how to insure that the results are optimized to best reflect reality.

121

NOTES:

122

P-207-T Effect of Particle Size, Pore Size and Column Configuration for the Analysis of Antibody Aggregates by SEC Jeffrey Kakaley1; Takashi Sato2; Ernest Sobkow1; Jeff Wilson1

1YMC America, Inc., Allentown, PA USA; 2YMC Co., Ltd., Kyoto, Japan Size Exclusion Chromatography (SEC) is a pervasive technique for monitoring aggregate formation for antibody and other protein solutions in the pharmaceutical and biotechnology industries. The increased use of various HPLC and uHPLC systems designed for narrow dispersion and rapid throughput has led to increased interest in faster SEC analyses using smaller stationary phase particle sizes. This investigation looks at the effects on resolution, speed of separation and integrity of results for various combinations of particle physical attributes coupled with variations in column geometries (length and inner diameter). P-208-T Separation and Characterization of Sub-visible Protein Particles using Centrifugal Field-flow Fractionation Coupled with Online Optical Microscopy Soheyl Tadjiki; Trevor Havard; Evelin Moldenhauer; Thorsten Klein Postnova Analytics Inc., Salt Lake City, UT USA The size range and quantity of sub-visible protein particles in therapeutic products are key product quality aspects. In this poster we demonstrate the feasibility of interfacing an online optical microscope with a centrifugal field-flow fraction (CFFF) system to separate, detect and image sub-visible protein particles (1 µm and above) in aggregated protein suspensions. CFFF is a mass based separation and characterization technique for nanoparticles and particles. In CFFF, separation is a result of the interaction between a centrifugal field and the sample species. In CFFF the channel encircles the centrifuge axis like a belt. The spinning of the channel generates differential acceleration forces at a right angle to the flow. Equilibrium will be achieved when the field-induced and diffusion-induced migrations of sample species are balanced. Smaller sized sample species located closer to channel center will be swept out faster than the larger ones. Protein nanoparticles and particles (28 nm to 30 µm, density of 1.4 mg/mL) can be separated by CFFF with a resolution 3 times higher than any other field-flow fraction techniques. A CFFF separation methodology was developed to separate protein particles in heat stressed Bovine Serum Albumin and Gamma Globulin solutions at low pH. An online optical microscope was interfaced with the CFFF system to image the elution of the protein particles. The images were then processed to obtain number based particle distributions. The method was validated using NIST narrow Latex bead standards. The detection limit was determined to be 104 particles per milliliter.

123

NOTES:

124

P-209-T Exploring Archimedes’ Principle to Distinguish Subvisible Protein Particles from Silicone Oil in the 0.5µm to 5µm Range Ankit R. Patel; Doris Lau; Jun Liu Genentech, a Member of the Roche Group, South San Francisco, CA USA While a number of characterization methods are available for analyzing subvisible particle content in protein pharmaceuticals, counting and characterizing particles within the entire subvisible size range remains a challenge due to the properties of the particles themselves and to limitations of the instrumentation. Methods relying on differences in refractive indices of the particle and solution (e.g. light obscuration, flow microscopy, etc.) are limited for small particle sizes (< 2µm) while methods for characterizing smaller aggregates (e.g. SEC, DLS, FFF, AUC, etc.) are not able to count individual particles. Additionally, silicone oil droplets contribute to subvisible particle counts for samples stored in prefilled syringes, which are increasingly being used for biotherapeutic products. Here, we evaluate a method that relies on differences in particle buoyant mass for characterization of particles in the range of ca 0.5 – 5µm. A model particle system was specifically designed to evaluate the suspended microchannel resonator (SMR)’s ability to distinguish between buoyant particles (e.g. silicone oil) and dense particles (e.g. protein particles). In addition, this emerging technique was applied to high concentration monoclonal antibody solutions stored in prefilled syringes in stressed stability studies. It is shown that the SMR system can potentially distinguish between silicone oil droplets and protein particles in a size range that is challenging for many subvisible particle characterization methods. P-210-T A Rapid Approach to Developing Reporter Cell Lines for Bioassays Aaron Miller; Brett Marshall; Max Tejada; Helene Gazzano-Santoro Genentech, a Member of the Roche Group, South San Francisco, CA USA Typical in vitro methods used to determine the activity of therapeutic antibodies rely on measuring functional endpoints such as cell proliferation or cell death using cell viability indicators such as alamarBlue™. These types of assays perform well but are often time-consuming and many of the cell viability indicators used are influenced by the metabolic status of the cells. In contrast, the upstream signaling events preceding the functional endpoints take place on the order of hours not days. Therefore, activity can be measured much more quickly using reporter cell lines. We present a case study describing a rapid way to generate reporter cell lines and the development and initial characterization of one such reporter cell line, which has been incorporated into multiple methods.

125

NOTES:

126

P-211-T Binding Activity of a FAb using Octet RED Sydney Zaremba; Kenji Furuya Boehringer Ingelheim, Fremont, CA USA A surrogate bioassay to determine the binding activity of a FAb has been developed using ForteBio’s Octet RED instrument. The assay is accurate, rapid, and robust, with inter and intra precision of less than 10%. The small molecule ligand of the FAb is immobilized onto Streptavidin biosensors and the binding rate is measured for a series of FAb concentrations. Samples of known concentration generate a standard curve which is then used to determine the concentration of unknown samples through interpolation. The activity of the sample is calculated by dividing the Octet measured Fab concentration by the A280 total protein concentration. The activity assay has faster turnaround times than either ELISA or Biacore, generating results in under one hour. The assay has been qualified for use in process development to monitor the activity of the FAb molecule during purification and in QC for release and stability sample testing. P-212-T Characterization of MAb1 Charge Patch Mutants Using Bioassays Guoying Jiang Genentech, a Member of the Roche Group, South San Francisco, CA USA A positive charge patch composed of 3 Lys and 1 Arg was identified on the surface of MAb1. In order to assess the impacts of the charge patch on PK, six charge patch mutants were constructed by making either one or double mutations of the 4 amino acids and tested in three bioassays. Our data indicate that charge patch removal impacts binding with target antigen and other anionic molecules that are abundant in vivo including phospholipids and hyaluronic acid. Two of the mutants (M3 and M4) were found to have comparable antigen binding with WT, but with much reduced binding to phospholipids and hyoluronic acid, making them suitable candidates for further in vivo PK study. P-213-T Development and Validation of a Cell-based Viral Neutralization Assay Peter Day Genentech, a Member of the Roche Group, South San Francisco, CA USA In this presentation, we describe the development and validation of a viral neutralization assay to support clinical lot release and stability testing. This bioassay measures the ability of Antibody A to neutralize viral entry into a target cell, which is reflective of the mechanism of action, and which requires the use of both cells and live virus. Despite the variability associated with both cells and virus, the assay was shown to be robust when we varied cell number, incubation times and critical reagent/virus lots. The assay was also shown to be accurate, with an overall recovery of 104%, and precise, with an overall RSD of 12%. The assay was implemented as the potency assay in the Phase I control system.

127

NOTES:

128

P-214-T Automation of Complex Workflows Using the AssayMAP Technology: Automating Sample Preparation and Peptide Mapping Steve Murphy; Michael Bovee; Zach Van Den Heuvel; Jennifer Reich; Scott Fulton BioSystem Development, Madison, WI USA Recent dramatic improvements in cell culture expression levels have had an impact on post-translational modification (PTM) patterns of the product. Peptide mapping (using HPLC or LC/MS for analysis) is the core method for monitoring changes in PTM, and demand for this type of product characterization is increasing. The complex sample preparation process for peptide mapping includes purification of the target protein, denaturation, reduction and alkylation, enzymatic digestion and peptide cleanup, and has become a major bottleneck and source of variability. The AssayMAP® high throughput microchromatography platform enables full automation of the complex, multi-step, peptide mapping workflow. The system utilizes 5 µL packed bed cartridges run on a 96-channel liquid handler (Agilent AssayMAP Bravo), and provides precise positive displacement flow control through a special probe syringe head. Cartridge types include affinity purification (using protein A or biotinylated antibodies/antigens as affinity ligands), immobilized trypsin (or other enzymes) and peptide cleanup (by reversed-phase, ion exchange or other resins). Binding, elution and digestion reactions are highly quantitative and reproducible due to high-precision flow-rate control Here we demonstrate the use of the AssayMAP Bravo platform to automate complex workflows using peptide mapping work flow of monoclonal antibodies as an example. The components of the peptide mapping workflow include affinity purification using protein A or Streptavidin cartridges, tryptic digestion using immobilized trypsin cartridges, and tryptic digest clean up using reverse phase cartridges. Characterization of each of these workflow modules will be presented along with the results of a peptide mapping workflow using these modules in series. P-215-T Microchromatography Platform Enables High Throughput Purification of Antibodies from Cell Culture Fluids David Knorr; Marc Beban; Ning Tang; Zachary Van Den Heuvel Agilent Technologies, Santa Clara, CA USA To address a growing desire for robust, high-throughput chromatography capability, we have developed a liquid-handling platform capable of processing up to 96 samples simultaneously. The Agilent AssayMAP Bravo is designed to utilize interchangeable media-filled AssayMAP cartridges. The platform features ultra-low dead-volume syringes capable of driving fluids at rates from below 1 µL/min. to 375 µL/second for true chromatography performance. This work describes purification of poly- and monoclonal antibodies from different media backgrounds constructed to mimic those encountered during antibody screening, production, and scale-up in commercial settings. A well-characterized monoclonal antibody was added at different concentrations to phosphate buffered saline or supernatant media harvested from Chinese hamster ovary cell culture. Automated purifications were

129

performed on the AssayMAP platform using specially formulated microchromatography cartridges containing immobilized protein A. Yields from each antibody dilution were not altered by the background media, or the presence of up to 10% fetal bovine serum in the media. Examination via LC-MS of molecular weight, peptide maps, and glycosylation patterns of the AssayMAP-purified material were identical to those of the original molecule. Additional media cartridges are being developed to widen the possible uses for this platform. By increasing sample throughput, the AssayMAP Bravo platform will shorten timelines for development of new antibodies for research and therapeutics. NOTES:

130

P-216-T A Fully Automated Solution for Quality Control of Biological Therapeutics Christian Albers1; Laura Main2; Catherine Evans3; Carsten Baessmann1; Wolfgang Jabs1 1Bruker Daltonik GmbH, Bremen, Germany; 2Bruker Daltonics Ltd., Coventry, United Kingdom; 3Bruker Daltonics GmbH, Faellanden, Switzerland Subtle changes in the manufacturing process of biological therapeutics can introduce unexpected and unwanted modifications to the product. Therefore, a sophisticated quality control (QC) solution is required for fast, reliable, and quantitative characterization of biopharmaceuticals in high throughput. The developed characterization workflow for quality control consists of intact protein analysis and peptide mapping of protein digests. To demonstrate the intact protein analysis we used 20 samples of human IgG1. Samples are injected, LC-MS data acquired, deconvoluted, and the resulting peak list qualitatively and quantitatively compared to a “gold” reference standard. To show the peptide mapping analysis BSA and Transferrin were reduced, alkylated and digested with trypsin. 96 samples were injected, LC-MS/MS data acquired and the data is matched to the theoretical digests of the proteins with possible modifications considered. Pass / fail criteria used for quality assessment can be adapted for different requirements. In this study mass accuracy was used as QC criterion. As soon as batch acquisition is started, a QC result table and a 96 tray view are generated. For each QC criterion a color code used in these displays can be defined. In our experiments we assigned a mass deviation less than 3 ppm to green, a mass deviation between 3 and 5 ppm to yellow and a mass deviation larger than 5 ppm to red, indicating that samples with a mass deviation larger than 5 ppm has failed quality testing. Reports containing chromatograms, list of compounds, and comparison with a “gold” standard are presented for the preformed batch measurements. The reports also include peptide maps for the protein digests, visualizing which peptides (MS) and fragments of the peptide (MS/MS) have been identified. For Transferrin sequence coverage of 94% is achieved from a single tryptic digest. P-217-T Automated MAb 2-Dimensional Workflow: from Harvest Cell Culture to Variant Analysis Shanhua Lin; Srinivasa Rao; Jim Thayer; Yury Agroskin; Chris Pohl Thermo Fisher Scientific, Sunnyvale, CA USA Early in the development of recombinant monoclonal antibodies (MAbs), a large of number of harvest cell culture (HCC) samples must be screened for IgG titer, aggregations, and charge variants. Affinity chromatography is often used first to purify MAbs, with typical yields of more than 95%. Size-exclusion chromatography is followed to identify and quantify MAb aggregations. Ion-exchange chromatography is then used to characterize charge variants. For the final biopharmaceutical product approval and subsequent manufacturing processes, a comprehensive characterization of MAb purity, aggregate forms, and charge variants is required by the regulatory agency.

131

In the present study, we automate 2-Dimensional HPLC workflow using a single HPLC system. This system consists of a dual-gradient pump, a UV/VIS detector, a column oven, and an autosampler capable of both sample injection and fraction collection. First, the HCC is injected onto the protein A column and IgG fractions are collected by the autosampler. Subsequently, the IgG fractions are injected onto MAbPacTm SEC-1 size-exclusion, or MAbPacTm SCX-10, 3 µm strong cation ion-exchange column for further analysis. We have recently introduced a 3 µm MAbPac SCX-10, 3 µm column in 4x50 mm format for high throughput MAb variant analysis. This column delivers high resolution separation with shorter run time using either salt or pH gradient. Incorporating this column into the workflow, we complete affinity purification, size-exclusion analysis and charge variant analysis in less than one hour. The successful automation of this 2D workflow system enables robust and high throughput MAb analysis. NOTES:

132

P-218-T Do You Really Know What Your Robot is Doing? The Importance of Paying Attention to Liquid Handling Details Kevin Khovananth; John Thomas Bradshaw; Keith Albert Artel, Westbrook, ME USA The introduction of automation into biology and chemistry labs has arguably lead to significant advances in testing capabilities over the past 20+ years. Automation has certainly led to increased numbers of experiments, as compared to manual testing, particularly for pipetting operations. Because of this advantage, liquid handling robots have become commonplace even in small laboratories. However, in spite of all the advantages that something like a liquid handling robot brings to a laboratory, it also brings a different set of commonly overlooked challenges. It may be argued that the largest challenge presented by using a liquid handling robot is the potentially incorrect assumption that the robot is doing what it is “supposed” to be doing. The robot may in fact be doing exactly what the user told it to do, but is that really what the user wanted? One may say that the real question is, do you really know how your robot is behaving, and particularly, do you really know how your robot is performing your assays? Herein, we present examples of the importance of monitoring various commonly employed tasks, which are likely considered mundane and often assumed to have little bearing on overall robot performance. Specific examples of best practices in programming automated liquid handling parameters that will be presented include; 1) pre-wetting tips, 2) using a blowout volume after pre-wet, 3) protocol transfer from one robot to a sibling, 4) protocol differences between high volume and low volume dispenses, 5) etc. The examples presented herein will help users to think more about the specific tasks they are asking their robots to perform, and hopefully uncover certain steps that, if observed and controlled, will result in better performance. P-219-T Characterization of Currently Marketed Heparin Products: Composition Analysis by 2D-NMR David Keire; Lucinda Buhse; Ali Al-Hakim US-FDA, St Louis, MO USA In this study, we evaluated 2D-1H-13C-heteronuclear single quantum coherence (HSQC) NMR methods to characterize the normal variation of heparin sodium and LMWH molecular composition in lots available on the US market in the summer of 2009. We tested 20 heparin sodium samples from 6 manufacturers and the USP heparin sodium identification standard. In addition, we tested 10 LMWH samples from three manufactures of different types of LMWH and the USP enoxaparin identification standard. One major goal of collecting these data was to establish the normal range of variability for each of the heparin types in this assay. Monosaccharide composition values and standard deviations are derived from multiple lots of heparin sodium, dalteparin, tinzaparin and enoxaparin and show mean %RSDs of 3 to 9% across the lots tested. Disaccharide composition values and standard deviations are derived from multiple lots of dalteparin, tinzaparin and enoxaparin and show mean %RSDs of 4 to 6%

133

across the lots tested. The monosaccharide values give a complete picture of the composition while the disaccharide values are a selected set which are simpler to measure and calculate. Each type of heparin has a unique pattern of signals in the NMR data because of the process specific chemistry used to manufacture them that can be used for assessing the quality of the drug. NOTES:

134

P-220-T Stability of Trisulfide Modification in an IgG1 Monoclonal Antibody (mAb) Biopharmaceutical Li Zang; Jason Fernandez; Andrew Weiskopf Biogen Idec, Cambridge, MA USA Trisulfide modification was found to be one of the common modifications to natural and recombinant monoclonal antibodies of all immunoglobulin G types. This modification occurs most predominantly to the heavy chain-light chain (HC-LC) disulfide linkage. Exposure of monoclonal antibody biopharmaceuticals to H2S during protein production or purification can potentially lead to increased level of trisulfide content in the final product. An IgG1 type mAb biopharmaceutical was produced with approximately 2.9% trisulfide in its HC-LC disulfide linkage and the trisulfide content was monitored over 9 months in a stability study conducted at three different temperatures, 2-8°C, 25°C and 40°C. The mAb was found to have consistent trisulfide content at the HC-LC disulfide when stored at 2-8°C, while at elevated temperatures, especially at 40°C, the relative percentage of trisulfide at HC-LC linkage slightly increased over time. The increase was suspected to be due to a small amount of H2S present in the mAb solution, or produced as a byproduct of β-elimination reactions that form thioether bonds between HC and LC during the storage at elevated temperatures. In order to verify the correlation between trisulfide increase and free H2S content in mAb solution during storage at elevated temperature, a 3 month stability study of the mAb at 40°C was performed with formulation of the mAb in two different buffers with various pHs and salt concentrations. The results indicated that sodium phosphate buffer, which reportedly scavenges free H2S more effectively, was capable of slowing down the trisulfide increase and when added at a concentration of 100 mM, essentially stopping the trisulfide increase over time. In comparison, the N2 gas sparge, also reported to reduce the presence of the trisulfide species, did not appear to affect the trisulfide content increase over time. P-221-T Localized Conformation Analyses of Mutated Human IgG1 by Hydrogen Deuterium Exchange Mass Spectrometry and Differential Scanning Calorimetry Joomi Ahn1; Xiaojun Lu2; Colette Quinn3; St John Skilton1; Ying Qing Yu1; Jihong Wang2 1Waters Corporation, Milford, MA USA; 2MedImmune, Gaithersburg, MD USA; 3TA Instruments, New Castle, DE USA Hydrogen/deuterium exchange mass spectrometry (HDX MS) technology constitutes an indispensible method for the study of local changes to protein conformation. Recent improvements in LC-MS systems have made HDX MS a useful biophysical tool for the development and discovery of protein drugs. Amino acid substitution on biopharmaceuticals is often explored to improve pharmacokinetics properties. The higher order of protein structure is closely related to its function. Therefore the localized conformation due to the mutation needs to be carefully monitored. Typically the three-dimensional structure is determined by X-ray crystallography. However the localized conformational changes at the substitution sites may not be easily detected in a crystal structure comparison. The HDX LC-MS workflow and its system were deployed in this study to detect differences in deuterium uptakes at

135

peptide level. This information could be related to the subtle and distinct changes of conformation due to local effect. Time consuming HDX data processing was improved by an innovative HDX software tool. The software automatically calculated the deuterium uptakes and the HDX results were displayed in convenient comparative views: uptake curves, a butterfly chart, and a difference plot. We report the efficient comparability studies using a well-established HDX LC-MS method including an informatics package. In addition, differential scanning calorimetry (DSC) was used as a complementary method to confirm the conformation at global level. NOTES:

136

P-222-T Protein Concentration Determination in Crude E Coli Cell Paste Extract Arugadoss Devakumar; Alice Yee; Anu Bansal Genentech, a Member of the Roche Group, South San Francisco, CA USA Escherichia coli are commonly used host cells for the large-scale production of therapeutic proteins. The ability to detect, identify, and accurately quantify the therapeutic protein concentration in samples containing high levels of process related impurities is an ongoing challenge due to matrix impact on the overall process yield and final product quality. In particular, minimizing or removal of the interfering process related impurities (such as host DNA and host cell proteins) is a key step for accurate protein determination. We have developed an efficient method for protein analysis by HPLC for recombinant therapeutic protein in crude cell paste. This method has resulted in high yield extractions of the therapeutic protein from E coli inclusion bodies for accurate protein concentration determination in downstream processing. Development work for this method will be presented. P-223-T Improving Standard N-glycan Sample Preparation with Manual Automation Using Microchromatography to Improve Efficiency, Accuracy, and Reproducibility Toni Hofhine1; Craig Nishida2; Jo Wegstein2 1Gilson, Inc, Middleton, WI USA; 2ProZyme, Inc, Hayward, CA USA The search for more efficient tools that streamline the sample preparation process for characterization of N-glycans continues to be at the forefront in the field of glycomics. The goal is to more effectively streamline the sample preparation and data analysis to allow for a deeper understanding of the multiple structures and functions that result from glycosylation. Adding sugars to proteins in the process of forming glycoproteins can be complex. Abnormal glycosylation, congenital disorders of glycosylation (CDG), can occur and be linked to a number of diseases. Better data and documentation of this structure and function relationship can assist with understanding how these diseases occur and thus further the development of disease biomarkers. Monitoring bioprocessing of monoclonal antibodies (mAb) is another area where product glycosylation is important. Eliminating manufacturing variability from batch-to-batch is critical, but the traditional laboratory sample preparation approach can often take too long. Traditionally, sample preparation for N-glycans takes more than three days and often involves specific laboratory technique, not including the time for sample analysis. The transition to manual automation tools combined with microchromatography reduces sample preparation to less than 2.5 hours, while providing the added confidence that N-glycan sample preparation and the sample data are reliable. Human Immunoglobulin G (hIgG) was mixed with bovine fetuin prior to microchromatography purification and subsequent sample preparation, labeling and analysis using HPLC with fluorescence detection. Results indicate that manual automation tools show improved accuracy and reproducibility between samples while allowing for more effective and more efficient characterization of N-glycans to be performed.

137

NOTES:

138

P-224-T Automatic MS/MS Characterization of N-linked Glycopeptides Anja Resemann1; Laura Main2; Ulrike Schweiger-Hufnagel1; Rainer Paape1; Arndt Asperger1; Detlev Suckau1 1Bruker Daltonik GmbH, Bremen, Germany; 2Bruker Daltonics Ltd, Coventry, United Kingdom Glycosylation is the most abundant posttranslational protein modification and is involved in many important biological processes and pathways. However, acquisition and interpretation of the complex fragmentation patterns of glycopeptides can be challenging. GlycoQuestTM is a recently developed search engine which provides glycan identification after an automatic classification step. All glycoproteins were reduced, alkylated and digested (trypsin or Glu C) in solution. The digest solutions were separated by nano-LC and either online analyzed by ESI-ion trap or the fractions were automatically spotted to a sample holder and subsequently analyzed by MALDI-TOF/TOF-MS. All MS/MS spectra were processed as it is routine for proteomics. Database searches were performed using Swissprot and NCBInr for protein and GlycomeDB for glycan identification. MALDI-TOF/TOF spectra of N-linked glycopeptides show a characteristic fragmentation, which can be used to classify the compound as a glycopeptide and to determine the peptide moiety mass. The spectra contained glycan fragments as well as peptide fragments. Using the peptide mass as modification of the glycan reducing end in the data system, a search in glycan databases allows identification of the glycan moiety structures with matching molecular weight. In a second step, Mascot protein database searches were performed using the resulting peptide mass as precursor. By contrast, ESI-CID-ion trap spectra neither show peptide fragmentation nor a common characteristic peak pattern. Here, our algorithm looked for peak distances matching monosaccharide masses in order to localize glycan fragment series and deduce the most likely aglycone mass. Experiments with tryptic and Glu C digests of a human monoclonal antibody revealed 28 potential glycopeptides, 24 of them contained the identical peptide sequence and 15 different glycan compositions. Further experiments were performed on more complex samples such as bovine alpha-1-acid-glycoprotein. P-225-T Rapid Profiling of N-linked Glycans from Monoclonal Antibodies Using Microfluidic-Based Chip NanoLC-MS Niclas Tan; May Zhu Millennium Pharmaceuticals, Cambridge, MA USA Characterization of glycan post-translational modifications of therapeutic antibodies is of utmost importance due to their impact on efficacy, half-life, stability and mechanism of action. Glycosylation is affected by many factors including production cell line as well as production parameters such as culture medium, pH and temperature. Thus, the ability to rapidly and precisely analyze these glycans will provide crucial feedback to product quality evaluation during the cell line development phase. Traditional assays encompass the separation and subsequent fluorescence detection of Peptide-N-glycosidase F (PNGase-F)-released and fluorophore-labeled glycans. This multi-step approach is tedious

139

and typically ranges from one to four days to complete. Additional labeled glycan standards may be required for identification. With the growing number of new drug development candidates in the pipeline, a fast and sensitive characterization approach is highly desired. Here we present the results from the optimization of a novel integrated microfluidic-based LC-MS chip for rapid online cleavage, purification, separation, identification and quantitation of label-free N-linked glycans from monoclonal antibodies (mAbs) based on accurate mass. The entire run time is significantly reduced to 16 minutes per sample with identification of afucosylated, sialylated and isomeric forms of major and minor glycans species. The N-linked glycan characterization results for mAbs derived from different production cell lines and commercial therapeutic antibodies are also presented to examine similarities and differences. This automated workflow is amenable to high-throughput analysis of large sample sets during clone selection, process development and quality control of different manufacturing lots. NOTES:

140

P-226-T Feasibility of GlykoPrep Sample Preparation for Glycoanalysis on the Bravo Liquid Handling System Using the 96AM Syringe Head Ted Haxo1; Michael Kimzey1; Sybil Lockhart1; Steven Mast1; Abdel Minalla1; Craig Nishida1; Bopha Sun1; Shiva Pourkaveh1; Jo Wegstein1; Randy Bolger2; Scott Fulton2; David Knorr2; Steve Murphy2; Zach Van Den Heuvel2 1ProZyme, Hayward, CA USA; 2Agilent Technologies, Santa Clara, CA USA Until recently, sample preparation for glycoanalysis has been a manual, complex and multi‐day process that usually limits analysis to only the most critical decision points. To address the growing need for increased throughput of N‐glycan profiling, here we present feasibility studies of the GlykoPrep(tm) Sample Preparation for glycoanalysis using the Agilent Bravo liquid handling system with the 96AM Syringe Head, and compare throughput and data quality for a number of analytical methods. P-227-T Development of an N-linked Glycan Database for Waters Ultra Performance Liquid Chromatography (UPLC) systems Mark Hilliard; Weston Struwe; Barbara Adamczyk; Giorgio Carta; John O Rourke; Pauline Rudd National Institute for Bioprocessing Research (NIBRT), Dublin, Ireland Glycosylation is the most complex post-translational protein modification and influences protein structure and function. Over half of all proteins are glycosylated and in order to elucidate their function we also need to understand their glycan composition and structure. The two major classes of glycans are N-and O-linked glycans. N-glycans derive their name from the nitrogen in the side chain of asparagine in the sequence Asn-Xaa-Ser/Thr, whereas O-glycans are formed by a linkage between an amino acid hydroxyl side chain from serine or threonine (Ser or S, Thr or T) residues, most commonly to GalNAc. Manufacturing of protein life sciences products can be very difficult and a number of factors can have a major effect upon their glycosylation, such as dissolved oxygen, pH, carbon source and temperature. Such fluctuations in the process can put product integrity at risk and thus the ability to correctly monitor glycosylation is required. In collaboration with Waters, NIBRT has developed an N- glycan database for the ACQUITY /H-class UPLC. GlycoBase 3+ is based upon reproducible separation technology (HILIC-UPLC) with tight standard deviation, coupled with a bioinformatics/web-enabled database solution. Glycobase 3+ is currently populated with over 300 2-AB-labeled N-linked Glycan structures common to the biopharmaceutical industry. This novel database will assist in the identification of N-linked glycans based on a variety of data attributes and will facilitate the rigorous characterization of biopharmaceutical protein products.

141

NOTES:

142

P-228-T Rapid Glycoprotein Sialic Acid Determination by High-Performance Anion-Exchange Chromatography with Pulsed Amperometric Detection Deanna Hurum; Jeffrey Rohrer Thermo Fisher Scientific, Sunnyvale, CA USA Sialic acids are often determined in protein therapeutic products due to their critical role in controlling glycoprotein bioavailability, function, stability, and metabolism. Two sialic acids, N-acetylneuraminic acid (Neu5Ac) and N-glycolylneuraminic acid (Neu5Gc), are commonly determined in glycoproteins by a two step process. The sialic acids are first released by either acid hydrolysis or enzymatic digestion, and then the analytes are quantified by a chromatographic method determining both the individual amounts of Neu5Ac and Neu5Gc and total sialic acid content. Glycoprotein expression optimization experiments have the potential to generate numerous samples requiring this analysis to monitor reactor performance. There are two classes of common chromatographic determination methods; direct detection methods, such as High-Performance Anion-Exchange Chromatography with Pulsed Amperometric Detection (HPAE-PAD), and those that require sample derivatization for analyte detection, such as fluorescent labeling followed by UHPLC. This work describes the development of a 5 min HPAE-PAD assay for glycoprotein sialic acids after glycoprotein hydrolysis in acetic acid. Calibration ranges were chosen spanning the expected Neu5Ac and Neu5Gc amounts in five representative glycoproteins with a range of percent glycosylation by weight. Response is linear in the investigated ranges, with determination coefficients &ge;0.9995. The method LODs of 0.11 pmol and 0.058 pmol for Neu5Ac and Neu5Gc, respectively, allow sialic acids detection in diluted acid hydrolyzates without lyophilization prior to sample analysis. This reduces sample preparation time. Retention time precisions (RSDs of <0.8) and peak area precisions (RSDs of < 1.98) for standards are excellent. Recoveries from the five glycoprotein hydrolyzate samples ranged from 81-96% for Neu5Ac and 82-106% for Neu5Gc, suggesting method accuracy. This 5 min assay allows the rapid, direct, and accurate quantification of glycoprotein sialic acids, providing a convenient method for quick expression optimization experiment screening without the costly and time consuming derivatization steps required for existing UHPLC fluorescence detection methods. P-229-T On the Path to Biobetter Therapeutic Glycoproteins: Simple and Rapid Domain-Specific Screening to Target and Control Optimal Glycan Profiles Steven Mast1; Bopha Sun1; Michael Kimzey1; Shiva Pourkaveh1; Abdel Minalla1; Ted Haxo1; Craig Nishida1; Jo Wegstein1; Scott Fulton2 1ProZyme, Inc, Hayward, CA USA; 2BioSystem Development, LLC, Madison, WI USA A strategy for “Target-directed Product Development” has been described, where multiple glycosylated forms of the protein were generated and evaluated for Critical Quality Attributes (Toll, 2011). In the cited case, a follow-on biological candidate was selected for further development, whose N-glycan profile demonstrated improved efficacy compared to the innovator product. Subsequently this optimal profile was controlled within tight specifications throughout clone selection, cell-culture optimization and formulation in order to speed regulatory approval. Yet significant analytical challenges were

143

reported due to the tedious, manual methods employed and the variability of the results, requiring a significant investment of resources to accomplish. This poster presents methods for simple and rapid screening of N-glycan profiles suitable for every aspect of drug development and manufacturing, including biocomparability studies and bioreactor monitoring and control (PAT). Use of the GlykoPrep™ platform to dramatically streamline glycoprotein sample preparation, coupled with rapid analysis using a Waters® UPLC®, allows the generation of up to 96 high-quality results overnight. For some glycoproteins, the information presented in the total N-glycan profile may be insufficiently detailed for Target-directed Product Development. Here we show a domain-specific analysis of the N-glycans of a commercially available Fc-fusion protein. Analysis of the individual glycosylation profiles may be a better indicator to investigate the role glycosylation plays in the efficacy of the protein. NOTES:

144

P-230-T A High-Throughput Platform for Preparation of APTS-Labeled N-Glycans: Improving the Accuracy, Reproducibility and Time-to-Results of N-Glycan Profiling Michael Kimzey1; Shiva Pourkaveh1; Steven Mast1; Bopha Sun1; Abdel Minalla1; Ted Haxo1; Craig Nishida1; Jo Wegstein1; Andras Guttman2, 3; Akos Szekrenyes3; Zoltan Szabo2; Barry Karger2; Jeff Chapman4; Hans Dewald4 1ProZyme, Inc, Hayward, CA USA; 2Barnett Institute, Northeastern University, Boston, MA USA; 3Horvath Laboratory of Bioseparation Sciences, Debrecen, Hungary; 4Beckman Coulter, Inc, Brea, CA USA Analysis of APTS-labeled glycan conjugates using capillary electrophoresis (CE) is a sensitive, robust and fast approach to characterize N-linked sugar structures. However, this method has not been widely implemented in screening large numbers of samples, due in part to the lack of commercially available tools for automated sample preparation and analysis. Figure 1 presents an overview of a glycoanalysis platform comprising rapid sample preparation coupled with CE-based glycan analysis, which includes: 1) an automatable format with an optional purification module to allow direct screening of monoclonal antibody (MAb) cell-culture samples; 2) quantitative deglycosylation and separation of N-glycans; 3) complete glycan labeling for laser-induced fluorescence (LIF) detection; 4) efficient glycan sample cleanup and desalting to reduce excess reagent peaks; and 5) N-glycan profiling and glucose unit (GU) value-based structural prediction with a high degree of accuracy. The platform enables the generation of up to 96 high-quality results overnight. The GlykoPrep® N-Glycan Sample Preparation protocol features optional purification of MAbs, enzymatic deglycosylation, APTS labeling, cleanup and standardization of N-glycans optimized for downstream CE-LIF analysis. Using proprietary reagents, the optimized labeling reagents provide complete derivatization with only one-hour incubation time, without degradation of important labile groups, such as sialic acid and core or outer-arm fucose. After labeling, the remaining free APTS is efficiently (>99.9%) removed and the sample eluted in water, enabling the potential for mass spectrometry confirmation of atypical peaks. Spike-in, Lower and Upper Internal Mobility Standards are used to normalize glycan migration times further reducing variability between runs; GU assignment based on normalized glycan migration times showed significant improvement in precision compared to GU assignment without use of the Internal Mobility Standards. It is envisioned that this unique glycoanalysis solution will be important for the standardization and expansion of N-glycan profiling in such applications as clone selection and cell-culture optimization. P-231-T Rapid Sample Preparation of Biologics to Support High-throughput and High-resolution Glycan Analysis by Capillary Electrophoresis Andras Guttman1; Zoltan Szabo1; Barry Karger1; Jo Wegstein2; Craig Nishida2; Jeff Chapman3; Hans Dewald3; Stefan Mittermayr4; Jonathan Bones4; Margaret Doherty4; Pauline Rudd4 1Barnett Institute, Northeastern University, Boston, MA USA; 2ProZyme, Inc, Hayward, CA USA; 3Beckman Coulter, Inc, Brea, CA USA; 4National Institute for Bioprocessing Research and Training (NIBRT), Dublin, Ireland

145

Structural characterization of the glycan moieties of biologics, especially recombinant monoclonal antibody therapeutics (rMAb), is critical during clone selection, cell-culture optimization and product characterization. High-throughput screening methods are required in order to return results promptly to allow multiple iterations for selection of optimal candidates. This poster gives an overview of a state-of-the-art screening protocol with rapid sample preparation coupled with capillary electrophoresis (CE)-based glycan analysis that includes: automatable sample preparation with optional purification modules to allow direct screening of cell-culture samples; glycan labeling for laser-induced fluorescent (LIF) detection; clean-up to reduce excess reagent peaks and desalting followed by oligosaccharide CE profiling and/or carbohydrate sequencing and glucose unit (GU) value-based structural prediction. Glycans released from subnanomolar amounts of rMAbs are quickly and accurately prepared and analyzed, e.g., rapid (<7 min) CE separation of twelve key IgG glycans can be obtained along with 96-well format operation for convenient overnight processing. Particular attention is paid to full separation of core-fucosylated and afucosylated forms, as the presence or absence of this modification is of high importance in regulating the effector function of rMAbs. The excellent relative migration time reproducibility of the optimized CE separation method (RSD <0.09%) facilitates high-fidelity peak assignments for the individual components in the glycan pools, and consequently allows high-precision, structural predictions using GU value databases. Exoglycosidase array-based sequence verification of the predicted glycans is also presented. NOTES:

146

P-232-T Rapid, Site-Specific Glycoprotein Analysis by Nano-LC/MS/MS with Automated Spectral Interpretation and Differentiation of Structural Isomers Serenus Hua1; Charles Nwosu1; John Strum1; Ning Tang2; Keith Waddell2; Hyun Joo An3; J. Bruce German1; Carlito B. Lebrilla1 1UC Davis, Davis, CA USA; 2Agilent Technologies, Santa Clara, CA USA; 3Chungnam National University, Daejeon, Korea Protein glycosylation is increasingly recognized as crucial to understanding protein structure and function. The type and extent of glycosyl modifications on protein therapeutics mediates their mechanism of action, their rate of clearance from the body, and ultimately, their effectiveness. However, glycan microheterogeneity significantly complicates glycoprotein analysis from an analytical perspective. We have developed a rapid method to simultaneously and comprehensively profile the glycosylation of a protein or protein mixture within the context of its location in the protein structure. Our method may be applied to both N- and O-linked glycosylation and provides comprehensive analysis of glycan structure, glycosylation site, and relative quantities of all glycoforms in the system. Most importantly, it is exceptionally quick, taking just a few hours from the start of sample digestion to completed analysis. Our method has been optimized for rapid yet sensitive profiling of glycosylated biosimilars. Glycoproteins are quickly digested in a microwave reactor using controlled non-specific proteolysis in order to produce informative glycopeptides. High-resolution, isomer-sensitive chromatographic separation of the glycopeptides is achieved using chip-based nano-LC with porous graphitized carbon as the stationary phase. Integrated accurate-mass MS and MS/MS not only confirms glycopeptide composition but also differentiates glycan and peptide isomers and yields structural information on both the glycan and peptide moieties. Automated MS and MS/MS spectral interpretation enables rapid identification of each analyzed glycopeptide based on known fragmentation patterns and provides a comprehensive picture of all glycoforms associated with all glycoproteins in the sample. P-233-T Analysis of N-Linked Glycans using a 1.7 µm Amide-Bonded Hybrid Stationary Phase Stephan Koza; Paula Hong; Thomas Wheat; Kim Van Tran; Joomi Ahn; Jonathan turner; Diane Diehl; Kenneth Fountain Waters Corporation, Milford, MA USA Glycosylation is a post-translational modification of proteins. The oligosaccharides on glycoproteins can mediate biological activity, and are therefore associated with the safety and efficacy attributes of many biopharmaceuticals. The relative amounts of the individual glycan structures may need to be monitored at all stages of research and development.

147

In this poster, N-linked glycans are released from glycoproteins and are labeled with 2-aminobenzamide (2-AB). The derivatized oligosaccharides are then separated using an ACQUITY UPLC® BEH Glycan Separation Technology column containing 1.7 micron particles. The use of a fluorescence detector with a low volume flow cell (< 2 µL) also enhances the sensitivity of the method. The Waters ACQUITY UPLC instrument provides the operating characteristics to realize the resolution, sensitivity and speed benefits of a sub-2 micron particle packing material. NOTES:

148

P-234-T Development of a Glycan Library for Rapid Identification of Protein Glycosylation Danielle Aldredge1; Hyun Joo An2; Ning Tang3; Keith Waddell3; Carlito Lebrilla1 1UC Davis, Davis , CA USA; 2Chungnam National University, Daejeon, Korea; 3Agilent Technologies, Santa Clara, CA USA Glycosylation is one of the most common post-translational modifications in proteins, with 70% of human serum proteins being glycosylated. Recently, an increasing significance has been placed on the role of glycan structure in relation to protein structure and function. Glycans serve important structural roles by participating in protein folding, enhancing protein solubility, attaching proteins to cell membranes, protecting proteins from degradation and mediating many important biological functions. Changes in glycosylation have been linked to various pathological states including cancer, affirming their significance as an important target for analysis. Alteration of glycan structures can take on several forms including different monosaccharide compositions, changes in connectivity (sequence), and most interestingly changes in the types of linkages between monosaccharides. Slight variations in glycosylation can lead to significant implications in protein structure and function, especially in the case of protein therapeutics. The ability to monitor these changes quantitatively and specifically can make monitoring of glycosylation a valuable tool by providing greater sensitivity and specificity when characterizing protein glycoforms. To address this need, we have developed a robust and sensitive method to accurately identify glycan structures from unknown protein sources based on unique retention time and isomer-specific tandem MS. We employ a detailed structure library, previously established with exoglycosidase and tandem MS, with reproducible nano-LC separation of positional and linkage isomers. In addition, for each glycan isomer, unique isomer-specific tandem MS is obtained, allowing for identification of glycan structures using spectral matching based on differences in fragmentation patterns as well as diagnostic peaks unique to one isomer. These tools allow for rapid and accurate glycan structural identification based on accurate mass, retention time, and isomer-specific tandem MS from any protein source. P-235-T Development of a Novel Protein-specific Method for Characterizing Glycosylation in Protein Mixtures Termed INPEG (In-gel Nonspecific Proteolysis for Elucidating Glycoproteins) Charles Nwosu; Jincui Huang; Danielle Aldredge; John Strum; Serenus Hua; Carlito Lebrilla University of California, Davis, Davis, CA USA For many recombinant protein drugs, glycosylation is of great importance as changes in glycan profiles can lead to dramatic differences in glycoprotein pharmaceutical efficacy. Traditionally, there are five major steps in profiling the N-glycans of monoclonal antibodies (mAbs): (i) purification of the mAb from a complex cell culture medium via affinity chromatography, (ii) glycan release from the purified mAb via PNGase F digestion, (iii) purification of the glycans via solid phase extraction (SPE) or ethanol precipitation, (iv) derivatization of the glycans and (v) detection, identification and quantification of the release N-glycans via HPLC-, CE- or MS-based approaches. These traditional steps require microgram

149

amounts of protein, are tedious and time-consuming routinely requiring up to 4 days of analysis time from protein purification to the actual glycan identification. Additionally, such analyses are not protein-specific as there is no verification that the entire N-glycan pool released from the purified protein batch is solely associated with the target mAb. As an alternative, we present a unique approach combining gel separation of a crude protein mixture, in-gel nonspecific proteolysis and mass spectrometric analysis. We have termed this novel approach ‘I-N-P-E-G’ (In-gel Nonspecific Proteolysis for Elucidating Glycoproteins) and have successfully shown its application to model glycoproteins such as commercial immunoglobulin G (IgG) as well as IgG in complex mixtures such as milk and serum. The entire process involved in INPEG from gel analysis to analyte identification was achieved within 1 day. Here, small glycopeptides are generated via in-gel nonspecific proteolysis with Pronase thereby affording (a) glycan, (b) glycosite, (c) protein and (d) site-heterogeneity information for the target protein spot investigated. Short peptide tags (typically containing 3-5 amino acids) attached to all the different glycans on a protein allows for comprehensive coverage of the protein’s glycosylation via nano-LC MS analysis using a porous graphitized carbon chip. NOTES:

150

P-236-T Identification and Quantitation of Glycans Using LC/MS Approach and MS/MS Library Ning Tang; Shuai Wu; Keith Waddell Agilent Technologies, Santa Clara, CA USA Biologics, such as therapeutic monoclonal antibodies are sensitive to manufacturing process changes. The glycans are important for therapeutic monoclonal antibodies to maintain their drug efficacy. Small changes in manufacturing process may affect the composition of the glycans, therefore change the drug efficacy and even have serious health implications. Monitoring the IgG glycosylation is essential for quality control of therapeutic monoclonal antibodies. A robust, fast and reproducible method has become an analytical challenge. A microfluidic chip that integrated an enzyme reactor packed with PNGase F, glycan capture trap and glycan analysis column have been developed. The N-glycans were cleaved and analyzed within 12 minutes. The glycans were identified by searching an accurate mass N-glycan database. Relative quantitation of the glycan profile was calculated based on integrated ion chromatogram. In this study, MS/MS experiments were performed and a MS/MS spectra library was compiled. Tandem MS spectra provided glycan structure information and helped to validate the identification of the glycans. P-237-T Continued Process Verification for Biological Processes Julia O'Neill Merck, West Point, PA USA The January 2011 FDA Guidance on Process Validation requires manufacturers to implement continued process verification. In this third stage of validation, ongoing assurance is gained during routine production that the process remains in a state of control. Biological processes typically exhibit high relative variation, and sequentially dependent results (autocorrelation). These properties may prevent standard statistical control tools from working reliably. Challenges and recommended practices will be presented. Topics will include: Evaluating stability and estimating variation for autocorrelated processes; Detecting trends and shifts in the presence of high inherent variation; and assessing process capability relative to specifications and business requirements. P-238-T Introduction of New Technologies in IMP QC John De Los Santos; Kimia Rahimi; Sarah Du Genentech, a Member of the Roche Group, South San Francisco, CA USA In recent years, new technologies and capabilities have been implemented in IMP QC at Genentech. This poster presents an overview of the application of MALDI-TOF peptide mass fingerprinting method

151

as an identity test, the implementation of an innovative ELSD mixed-mode chromatography system for polysorbate quantification, and the qualification of the new Beckman PA800 Plus systems in IMP QC for improvement of system robustness for CE test methods. The MALDI-TOF Peptide Mass Fingerprinting for protein identification has the advantages of high specificity, robustness, minimal sample handling and short data analysis time. Since it was first introduced to IMP QC in February 2008 (V. Katta and E. Hoff et al), this method has been successfully validated and used in clinical lot release testing of more than 40 products. The HPLC - ELSD method retains polysorbate or other surfactants on a mixed mode column, elutes, then nebulizes the chromatographic effluent, evaporates the mobile phase and measures the light scattered by particles of analytes (T. Zhang and D. Hewitt et al). IMP QC successfully implemented HPLC - ELSD systems in September 2009. The method has been proven robust and accurate and has been validated for multiple products. Multiple CE methodologies have routinely been used for characterization, QC lot release and stability monitoring of therapeutic proteins in development. The new Beckman PA800 Plus systems have been introduced (D. Michels and M. Parker et al), qualified and implemented in IMP QC, and it has demonstrated overall improvements in operational efficiency and system robustness. NOTES:

152

P-239-T Generation and Uses of a Critical Quality Attribute Risk Assessment: The Importance of Collaboration between Functional Areas Taro Fujimori Abbott, Worcester, MA USA The Critical Quality Attributes Risk Assessment is a key link between product understanding and process understanding. Performing the CQA Risk Assessment, with broad representation from the Development Team is important in ensuring that the potential impact of quality attributes on safety and efficacy are considered by all stakeholders (Discovery, Toxicology, Clinical, etc). The CQA Risk Assessment provides guidance to various functional areas regarding potential follow-up activities, such as biomarker studies, in vitro and/or in vivo mechanism of action studies and process development studies. A CQA Risk Assessment generated in such a way aids in the prioritization of development activities, which should lead to an improved understanding of product structure-function relationships and a robust control strategy. NOTES: