Ovarian Cancers:Evolving Paradigms in Research and Care

Report ReleaseWednesday, March 2, 2016

• Congressionally mandated report

• Sponsored by the Centers for Disease Control and Prevention

• IOM charged with completing the work

• Multidisciplinary committee assembled

• Four in-person meetings • Invited comment by

stakeholders and presentations by experts

Background

Statement of TaskAn ad hoc committee under the auspices of the Institute of Medicine will review the state of the science in ovarian cancer and formulate recommendations for action to advance the field. The committee will:• Summarize and examine the state of the

science in ovarian cancer research, • Identify key gaps in the evidence base and

the challenges to addressing those gaps, continued

Statement of Task (continued)

• Consider opportunities for advancing ovarian cancer research, and

• Examine avenues for translation and dissemination of new findings and communication of new information to patients and others.

The committee will make recommendations for public- and private-sector efforts that could facilitate progress in reducing the incidence of and morbidity and mortality from ovarian cancer.

Biology

Innovative Research Designs

Supportive Care Research & Practice

Prevention & Early

Detection

Diagnosis & Treatment

Secondary Prevention & Monitoring for Recurrence Management

of Recurrent Disease

End-of-Life Care

Long-Term Survivorship

Methods to Reduce Practice-Related Disparities

Intervention Development

Previvorship Survivorship

Conceptual Model

1. Prioritize study of high-grade serous carcinoma

2. More subtype-specific research is needed to define various subtype characteristics

3. Collaboration is essentiala. Pooling and sharing of data and

biospecimensb. Use of consortia

4. Dissemination and implementation are final steps for knowledge translation

Overarching Concepts

Ovarian Cancers:Evolving Paradigms inResearch and Care

#Ovarian Cancerswww.nas.edu/OvarianCancers

High-grade serous car-cinomaCarcinosarcoma

70% - 74%

7% - 24%

10% - 26%

2% - 6% 0.6% - 7.1%3% - 5%

1% - 7%

Ovarian Carcinomas – Not one disease

Ovarian Carcinomas – Origins

The fallopian tube is a likely site of most HGSC and genetic models are expanding

Recommendation

• Strategies to increase genetic counseling and testing for all women with ovarian cancer

• Wider offering of cascade testing• Determine analytic performance and

clinical utility of testing for germline mutations beyond BRCA1 and BRCA2 and mismatch repair genes associated with Lynch Syndrome.

Supportive Care Along the Survivorship

Trajectory

• Most research focuses on treatment rather than on how to improve the management of the acute and long-term physical and psychosocial effects of diagnosis and treatment across the trajectory of survivorship.

• Most research on survivorship aggregates patients of all cancer types

• Survivorship research on ovarian cancer rarely distinguishes different subgroups – age, racial and ethnic groups, stage, histology, etc.

Key Findings in IOM Report

Dissemination & Implementation of

Knowledge“an active approach of spreading evidence-based

interventions to the target audience via determined channels using planned strategies”

Biology

Innovative Research Designs

Supportive Care Research & Practice

Prevention & Early

Detection

Diagnosis & Treatment

Secondary Prevention & Monitoring for Recurrence Management

of Recurrent Disease

End-of-Life Care

Long-Term Survivorship

Intervention Development

Previvorship Survivorship

Prevention & Early

Detection

Diagnosis &

Treatment

Secondary Prevention

& Monitoring

for Recurrence

Previvorship Survivorship

Management of

Recurrent Disease

End-of-Life Care

Long-Term Survivorship

Methods to Reduce Practice-Related

Disparities

Methods to Reduce Practice-Related

DisparitiesSupportive Care Research & Practice

Final steps for knowledge translation into practice for all stakeholders

• Current methods for early detection in the general or high-risk population do not have substantial impact on mortality.

• Proven preventive strategies exist.• All women with invasive ovarian cancer should

receive germline genetic testing.• Genetic counseling and testing for the first-

degree relatives of women with a hereditary cancer syndrome or germline mutation.

• Uniform implementation of the standard of care and the inclusion of supportive care across the survivorship trajectory.

Some key messages

Survivors’ acceptance of treatment side effects evolves as goals of care change over the

cancer continuum

Melissa K. FreyNew York University Langone Medical Center

Verbal Disclosure

• Nothing to disclose.

• Ovarian cancer disease course– Long overall survival– Multiple treatment regimens

• What are meaningful clinical trial endpoints?– Overall survival– Progression-free survival– Patient reported outcomes– Health-related quality of life

• FDA workshop on alternative clinical trial endpoints (September 3, 2015)

– Co-sponsored by:• Society of Gynecologic Oncology (SGO)• American Society of Clinical Oncology (ASCO)• American Association for Cancer Research (AACR)

Herzog TJ et al. Gynecol Oncol. 2014.

Background

• Exploring patient preferences– OCNA Clinical Trial Endpoints: What do our patients consider

important (2013)

– NYU/SHARE: A qualitative study of ovarian cancer survivors' perceptions of endpoints and goals of care (2014)

• Shared decision-making– Physician awareness of patient goals– Incorporating goals when selecting treatment– Maximizing treatment efficacy AND quality of life

Minion LE et al. Gynecol Oncol. 2016.Frey MK et al. Gynecol Oncol. 2014.

Background

To determine whether survivors’ acceptance of treatment side effects changes over the disease continuum.

Objective

• Ovarian Cancer Survivorship Questionnaire– Developed by Annie Ellis– Combination of OCNA Clinical Trial Endpoints and NYU/SHARE– 30 questions, Likert-type scale and multiple choice

• Questionnaire available online (8/1/2015-8/12/2015)– Survivor networks– Social media

• Completed online by self-identified ovarian cancer survivors

• Consent for participation provided electronically

• Exempt status from NYU Langone Medical Center Institutional Review Board

Annie EllisPatient Advocate

Methods

Age (years) (N = 328)

18-35 7 (2%)

36-50 63 (19%)

51-60 142 (43%)

61-70 91 (28%)

>70 25 (8%)

Race / Ethnicity

Non-Hispanic white 303 (92%)

Hispanic 10 (3%)

Non-Hispanic black 5 (2%)

Asian/Pacific Islander 3 (1%)

Other or Unknown 7 (2%)

Disease site

Ovary 267 (81%)

Fallopian tube 21 (6%)

Primary peritoneal 37 (11%)

Unknown 3 (1%)

Disease stage

I 55 (17%)

II 33 (10%)

III 194 (59%)

IV 37 (11%)

Unknown 9 (3%)

Results - Demographics

Time since cancer diagnosis (N = 328)

<12 months 44 (13%)

1-4 years 184 (56%)

5-9 years 59 (18%)

10-14 years 23 (7%)

15-19 years 10 (3%)

> 20 years 7 (2%)

Recurrent disease

Yes 142 (43%)

No 180 (55%)

No answer 6 (2%)

Undergoing treatment at time of questionnaire completion

Yes 119 (36%)

No 205 (63%)

No answer 4 (1%)

Results – Demographics

0 1 2 3 4 > 50

102030405060708090

100

# Pa

rtic

ipan

ts

# Prior treatment regimens received

Prior treatment regimens

What is your treatment goal?

Overall survival45%

Progression-free survival12%

Quality of life41%

No answer2%

What is most meaningful to you?

Overall survival39%

Minimizing treatment side effects11%Minimizing disease symptoms

6%

Ability to engage in daily activities

35%

Attend a life event0%

Time off treatment6%

Other2%

When selecting a treatment, what is your expectation?

Cure35%

Remission 49%

Stable disease16%

All participants

When selecting a treatment, what is your expectation?

P < 0.001

Cure50%Remission

45%

Stable disease5%

Participants without recurrence (N = 180)

Cure16%

Remission54%

Stable disease31%

Participants with at least one re-currence (N = 142)

Which of these treatment side effects would you tolerate?

Goal of treatment:Cure

Remission Stable disease

Cure

Remission

Stable disease

Bowel obstructionShortness of breath

InfectionMucositis

OtotoxicityHand-foot syndrome

PainMemory loss

Nausea/vomitingHeadache

Sexual side effectsFlu-like symptoms

ArthralgiaSkin changesNeuropathyConstipation

DiarrheaFatigue

Alopecia

Treatment side effects (Goal = Cure)

% Participants willing to accept the side effect

Bowel obstructionShortness of breath

InfectionMucositis

OtotoxicityHand-foot syndrome

PainMemory loss

Nausea/vomitingHeadache

Sexual side effectsFlu-like symptoms

ArthralgiaSkin changesNeuropathyConstipation

DiarrheaFatigue

Alopecia

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Percentage of survivors who would accept treatment side effects based on the goal of treatment (Goal = Cure)

Alopecia

Fatigue

Diarrhea

Constipati

on

Skin ch

anges

Neuropath

y

Arthralgia

Flu-lik

e symptoms

Sexu

al side eff

ects

Headache

Memory loss

Nausea/vomitingPain

Hand-foot s

ynd...

Mucositi

s

Ototoxicity

Infection

Shortn

ess of b

reath

Bowel obstr

uction

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Percentage of survivors who would accept treatment side effects based on the goal of treatment (Cure vs. Remission)

Alopecia

Fatigue

Diarrhea

Constipati

on

Skin ch

anges

Neuropath

y

Arthralgia

Flu-lik

e symptoms

Sexu

al side eff

ects

Headache

Memory loss

Nausea/vomitingPain

Hand-foot s

ynd...

Mucositi

s

Ototoxicity

Infection

Shortn

ess of b

reath

Bowel obstr

uction

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Percentage of survivors who would accept treatment side effects based on the goal of treatment (Cure vs. Remission vs. Stable disease)

Alopecia

Fatigue

Diarrhea

Constipati

on

Skin ch

anges

Neuropath

y

Arthralgia

Flu-lik

e symptoms

Sexu

al side eff

ects

Headache

Memory loss

Nausea/vomitingPain

Hand-foot s

ynd...

Mucositi

s

Ototoxicity

Infection

Shortn

ess of b

reath

Bowel obstr

uction

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Expectation of cure among survivors with recurrent disease

Cure16%

Remission54%

Stable disease31%

Participants with recurrent disease(N=142, 43%)

Expectation of cure among survivors with recurrent disease

Participants with recurrent disease(N=142, 43%)

Stable disease

Cure (22 participants)

Remission

Bowel obstructionShortness of breath

InfectionMucositis

OtotoxicityHand-foot syndrome

PainMemory loss

Nausea/vomitingHeadache

Sexual side effectsFlu-like symptoms

ArthralgiaSkin changesNeuropathy

ConstipationDiarrheaAlopecia

Fatigue

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Cure Remission Stable disease

I would accept any side effects for a CURE!!

I’d take anything that has less than 20% chance

of killing me immediately.

I would tolerate these side effects if they were temporary.

I'd rather not tolerate any side effects.

None!

Depends on severity and quality of life tradeoffs.

This is a much bigger issue if ‘stable disease’is the best you can get.

Facing this now, I have learned that my tolerance for

side effects is pretty low under these conditions.

Survivor comments by goal of treatment

• Acceptance of treatment side effects declines with changing goals– Cure remission stable disease

• Goal of cure drives willing acceptance of treatment toxicity

• When selecting treatment balance treatment toxicities and quality of life measures

• Survivors’ decision tool for selecting treatment therapies for recurrent disease in development

Conclusions

Co-authors• Annie E. Ellis, Patient Advocate• Laura Koontz, PhD• Savannah Shyne, MPH• Jing-Yi Chern, MD• Jessica Lee, MD• Stephanie V. Blank, MD

Acknowledgements

Higher rates of clinically actionable multigene panel results in Ashkenazi Jewish patients

Melissa FreyNew York University Langone Medical Center

Verbal disclosure

• Nothing to disclose.

• Next-generation sequencing– Rapid– Cost-effective – Virtually unlimited number of genes

– Ovarian cancer genes – BRCA1, BRCA2, BARD1, BRIP1, MLH1, MSH2, MSH6, PALB2, PMS2, RAD51C, RAD51D

• Clinical actionability

– Actionable mutations • Mutations with known clinical manifestations and well-outlined cancer screening

guidelines

– Non-actionable mutations• Mutations in low- to moderate-risk genes for which consensus management

guidelines have not been establishedNorquist BM et al. JAMA Oncol. 2015

Multigene panel testing for cancer syndromes

To determine if there are specific patient populations in which multigene panels would be more likely to affect clinical management.

Objective

• Review of multigene panel testing results at a single institution • Study period: June 2013 - January 2015• All genetic testing performed after consultation by a certified genetic counselor• Mutations were characterized as actionable or non-actionable

– National Comprehensive Cancer Network (NCCN) guidelines – Genes not addressed by NCCN guidelines

– Consensus statements – Expert opinion

• Statistical methods: T-test, Chi-square, Fisher’s exact test

Methods

Results - Patient Demographics

Gender (N = 454)

Female 435 (96%)

Male 19 (4%)

Race/ethnicity

Non-Hispanic white

362 (80%)

Non-Hispanic black

31 (7%)

Hispanic 29 (6%)

Asian/Pacific Islander

32 (7%)

Ashkenazi Jewish ancestry

Ashkenazi Jewish 138 (30%)

Non-Ashkenazi Jewish

316 (70%)

Personal history of cancer

354 (78%)

Breast 251 (55%)Ovarian 49 (11%)Uterine 26 (6%)Colorectal 20 (4%)

Family history of cancer

417 (92%)

Results - Multigene panels454 patients

Pathogenic mutations62 mutations identified

56 patients (12%)19 genes

VUS291 VUS identified

196 patients (43%)38 genes

Actionable mutationsAPC, ATM, BARD1, BRCA1, BRCA2,

BRIP1, CHEK2, MEN1, MLH1, MSH6, PALB2, PMS2, PTEN,

RAD51D

Non-actionable mutationsCDKN2A, FAM175A, FANCC,

FLCN, MUTYH

Actionable 79%

Non-ac-tionable

21%• Actionable mutations

– 10% of all patients (47/454) – 79% of all mutations (49/62)

• Non-actionable mutations– 2% of all patients (9/454) – 21% of all mutations (13/62)

• Age, gender, race, ethnicity, personal history of cancer and family history of cancer were NOT associated with finding an actionable mutation

Results – Clinical Actionability

Mutations (N = 62)

• No differences when comparing Ashkenazi to non-Ashkenazi patients:– Gender, personal history of cancer, family history of cancer, #

mutations, # VUS

• Mutations in Ashkenazi patients were significantly more likely to be clinically actionable

Results - Ashkenazi Jewish ancestry

Ashkenazi Jewish(N=20)

Non-Ashkenazi Jewish(N=42)

P value

Actionable mutations

(N = 49)19 (95%) 30 (71%)

Non-actionable mutations

(N = 13)

1 (5%) 12 (29%) 0.04

Ashkenazi Jewish (N=138)

Actionable mutationNon-actionable mutation

Pathogenic mutations in Ashkenazi and non-Ashkenazi Patients

Non-Ashkenazi Jewish (N=316)

FANCC; 1

APC; 6

BARD1; 1

BRCA2; 1BRCA1; 1BRIP1; 1

CHEK2; 7

MLH1; 1MSH6; 1 CDKN2A; 1 FAM175A; 1

FANCC; 2FLCN; 1

MUTYH; 7

ATM; 3

BRCA2; 4

BRCA1; 2BRIP1; 5

CHEK2; 6

MEN1; 1

MLH1; 2

PALB2; 2PMS2; 1

PTEN; 3 RAD51D; 1

Ashkenazi Jewish (N=138)

Actionable mutationNon-actionable mutation

Pathogenic mutations in Ashkenazi and non-Ashkenazi Patients

Non-Ashkenazi Jewish (N=316)

FANCC; 1

APC; 6

BARD1; 1

BRCA2; 1BRCA1; 1BRIP1; 1

CHEK2; 7

MLH1; 1MSH6; 1 CDKN2A; 1 FAM175A; 1

FANCC; 2FLCN; 1

MUTYH; 7

ATM; 3

BRCA2; 4

BRCA1; 2BRIP1; 5

CHEK2; 6

MEN1; 1

MLH1; 2

PALB2; 2PMS2; 1

PTEN; 3 RAD51D; 1

2 BRCA1/2 mutations BRCA1 (1)BRCA2 (1)

6 BRCA1/2 mutations BRCA1 (2)BRCA2 (4)

FANCC; 1

APC; 6

BARD1; 1

BRCA2; 4

BRCA1; 4BRIP1; 1

CHEK2; 7

MLH1; 1 MSH6; 1

Ashkenazi Jewish (N=183)

Actionable mutationNon-actionable mutation

Non-Ashkenazi Jewish (N=374)

CDKN2A; 1 FAM175A; 1FANCC; 2

FLCN; 1

MUTYH; 7

ATM; 3

BRCA2; 4BRCA1; 2BRIP1; 5

CHEK2; 6

MEN1; 1

MLH1; 2

PALB2; 2PMS2; 1

PTEN; 3 RAD51D; 1

8 BRCA1/2 mutations BRCA1 (4)BRCA2 (4)

6 BRCA1/2 mutations BRCA1 (2)BRCA2 (4)

BRCA1/2 mutations from multigene panels and targeted single gene BRCA1/2 screening (N=557)

Ashkenazi Jewish (N=138)

Actionable mutationNon-actionable mutation

Pathogenic mutations in Ashkenazi and non-Ashkenazi Patients

Non-Ashkenazi Jewish (N=316)

FANCC; 1

APC; 6

BARD1; 1

BRCA2; 1BRCA1; 1BRIP1; 1

CHEK2; 7

MLH1; 1MSH6; 1 CDKN2A; 1 FAM175A; 1

FANCC; 2FLCN; 1

MUTYH; 7

ATM; 3

BRCA2; 4

BRCA1; 2BRIP1; 5

CHEK2; 6

MEN1; 1

MLH1; 2

PALB2; 2PMS2; 1

PTEN; 3 RAD51D; 1

Evolving understanding of actionable vs. non-actionable

• Multigene panel testing discovered: • Mutations in 56 patients (12%) • VUS in 196 patients (43%)

• Multigene panel testing should be considered in Ashkenazi Jewish patients• Majority of mutations are actionable (95%)• Mutations occur in multiple genes and are not limited

to BRCA1/2

Conclusions

Co-authors• Gabriella Sandler, BS• Rachel Sobolev, BS• Sarah Kim, MD• Rachelle Chambers, MS, CGC• Jessica Martineau, MS, CGC• Rebecca Y. Bassett, MS, CGC• Stephanie V. Blank, MD

Acknowledgments

Gene Recommendations SourceAPC Actionable Colorectal cancer - Colonoscopy every 5 years starting at age 40 Expert opinion

ATM Actionable Breast cancer - Recommend breast MRI NCCN Guidelines

BARD1 Emerging evidence suggesting actionability

Ovarian cancer - Consider risk-reducing salpingo-oophorectomy Norquist et al. JAMA Oncol . 2015.

Ovarian cancer - Recommend/consider risk-reducing salpingo-oophorectomy

Breast cancer - Recommend breast MRI, discuss option of risk-reducing mastectomy

Ovarian cancer - Recommend/consider risk-reducing salpingo-oophorectomy

Breast cancer - Recommend breast MRI, discuss option of risk-reducing mastectomy

BRIP1 Actionable Ovarian cancer - Risk-reducing salpingo-oophorectomy NCCN Guidelines

CHEK2 Actionable Breast cancer - Recommend breast MRI NCCN Guidelines

Parathyroid glands, anterior pituitary, enteropancreatic endocrine cell tumorsBiochemical tests – calcium, PTH, gastrin, gastric acid output, secretin-stimulated gastrin, fasting glucose, insulin, PRL, IGF-1Imaging tests – MRI/CT scan

Colorectal cancer - Colonoscopy

Endometrial and ovarian cancer - Consider option of risk-reducing hysterectomy and bilateral salpingo-oophorectomy, annual office endometrial sampling is an option

Gastric and small bowel cancer - Selected individuals or families or those of Asian descent may consider EGD with extended duodenoscopy

Urothelial cancer - Consider annual urinalysis

Colorectal cancer - Colonoscopy

Endometrial and ovarian cancer - Consider option of risk-reducing hysterectomy and bilateral salpingo-oophorectomy, annual office endometrial sampling is an option

PALB2 Actionable Breast cancer - Recommend breast MRI, discuss option of risk-reducing mastectomy NCCN Guidelines

Colorectal cancer - Colonoscopy

Endometrial and ovarian cancer - Consider option of risk-reducing hysterectomy and bilateral salpingo-oophorectomy, annual office endometrial sampling is an optionEndometrial cancer - Discuss option of risk-reducing hysterectomy, consider annual random endometrial biopsies and/or ultrasoundBreast cancer - Recommend breast MRI, discuss option of risk-reducing mastectomy

Thyroid cancer - Annual thyroid ultrasound

Colorectal cancer - Colonoscopy

Renal cancer - Consider renal ultrasound

RAD51D Actionable Ovarian cancer - Recommend/consider risk-reducing salpingo-oophorectomy NCCN Guidelines

Actionable Mutations

PMS2 Actionable NCCN Guidelines

PTEN Actionable NCCN Guidelines

MSH6 Actionable NCCN Guidelines

MLH1 Actionable NCCN Guidelines

MEN1 ActionableExpert opinion /

consensus statements

BRCA1 Actionable NCCN Guidelines

BRCA2 Actionable NCCN Guidelines

CDKN2A Non-actionableFAM175A Non-actionable

FANCC Non-actionableFLCN Non-actionable

MUTYH Non-actionable

Non-Actionable Mutations

A PHASE III CLINICAL TRIAL OF BEVACIZUMAB WITH IV VERSUS IP CHEMOTHERAPY IN OVARIAN, FALLOPIAN TUBE AND PRIMARY PERITONEAL

CARCINOMA NCI-SUPPLIED AGENT(S): BEVACIZUMAB (NSC #704865, IND #7921) NCT01167712 a GOG/NRG Trial (GOG 252)

Joan L. Walker; Mark F Brady; Paul A DiSilvestro; Keiichi Fujiwara; David Alberts; Wenxin Zheng; Krishnansu Tewari; David E Cohn; Matthew Powell; Linda van Le; Stephen Rubin; Susan A Davidson; Heidi J Gray; Steven Waggoner; Tashanna Myers; Carol Aghajanian; Angeles Alvarez Secord; Robert S Mannel

GOG 252: IP chemo and dose dense Paclitaxel showedimproved OS, both have toxicities; which is best?

Should we use dose dense Paclitaxel?

Should we use IP chemotherapy?

Should we useBevacizumab?

• JGOG 3016 showed improved OS, but not replicated in the US

• GOG 172 showed survival advantage, but was toxic, with only 42% receiving 6 cycles; additional studies were done to address the toxicity:-GOG9916/17 Substituted IP carbo for cisplatin-GOG9921 Reduced IP cisplatin dose

• GOG 218 showed improved PFS with Bev, andfeasibly safe with IP Chemo

Arm 1: Dose dense Paclitaxel

Arm 2: IP Chemo substitute

Arm 3:

Include Bevacizumab

Key questionsfor GOG 252 Indications and contemporary results

Implications forGOG 252 schema

All:

IP chemo, reducedcisplatin dose

Arm 1

Arm 2

Arm 3

GOG 252: Schema

• Stage II-III Epithelial Carcinoma: Ovary, Fallopian Tube, Peritoneal

• Resected to optimal: less than or equal to 1 cm visible tumor by surgeon report

• Exploratory: suboptimal (7%) and Stage IV (5%)

Eligibility

Differences in Dosing in GOG 252 Arm 3 IP Cisplatin compared to GOG 172

• Dose reduction cisplatin(100 down to 75 mg/m2)• Infusion time reduction 135 mg/m2 paclitaxel(3 hr instead of 24h)• All outpatient therapy• Bevacizumab 15 mg/m2 for all arms on cycles 2-22• Comparison arm dose dense paclitaxel with carbo IV AUC 6-

GOG262 (JGOG)

• Second experimental Arm IP carbo and dose dense paclitaxel

Progression Free Survival Optimal Stage II-III(10% stage II)

• Estimated hazard ratios, and logrank tests are adjusted for stage of disease and size ofresidual disease micro vs < 1cm

• CT required every 6 months for surveillance (not required in GOG 114/172)

Arm N Events Median PFS HR [95% CI] Logrank Logrank

IV Carbo 461 303 26.8 months Reference arm P-value Chi square

IP Carbo 464 300 28.7 months 0.947 [0.808-1.11]

0.416 0.661

IP Cisp 456 307 27.8 months 1.01 [0.858-1.18] 0.727 0.122

P r o g r e s s i o n - F r e e S u r v iv a l b y T r e a t m e n t G r o u pS ta g e I I o r I I I O p t i m a l l y D e b u l k e d

T r e a tm e n t G r o u p E v e n t s T o t a l M e d i a n ( m o s ) 1 : C r b ( IV )+ T + B e v 3 0 3 4 6 1 2 6 .8

2 : C r b ( IP ) + T + B e v

3 0 0

4 6 4

2 8 .73 : C i s ( IP ) + T + B e v

3 0 7

4 5 6

2 7 .80 .2

0 .4

0 .6

0 .8

1 .0

Pro

porti

on S

urvi

ving

P

rogr

essi

on-F

ree

3 : C i s ( IP ) + T + B e v 2 : C r b ( IP ) + T + B e v

1 : C r b ( IV ) + T + B e vT r e a tm e n t G r o u p

2 6 .82 8 .72 7 .8

E ve n ts T o ta l M e d i a n ( m o s ) 3 0 3 4 6 1

3 0 0 4 6 43 0 7 4 5 6

Progression Free Survival Optimal Stage II-III

0.0 0 12 24 36 48 60 72

Months on Study

1 461 387 244 169 111 37 02 464 391 262 177 125 39 03 456 372 255 168 120 34 0

P r o g r e s s i o n - F r e e S u r v iv a l b y T r e a t m e n t G r o u pS ta g e III w i th N o G r o s s R e s i d u a l D i s e a s e

• T r e a tm e n t G r o u p

E v e n t s T o ta l M e d i a n ( m o s ) 1 : C r b ( IV ) + T + B e v

1 4 4

2 3 9

3 1 .3• 2 : C r b ( IP ) + T + B e v

1 4 5

2 3 8

3 1 .8• 3 : C i s ( IP ) + T + B e v

1 3 8

2 3 9

3 3 .8

0 .2

0 .4

0 .6

0 .8

1 .0

Pro

porti

on S

urvi

ving

P

rogr

essi

on-F

ree

3 : C i s ( IP ) + T + B e v 2 : C r b ( IP ) + T + B e v

1 : C r b ( IV ) + T + B e vT r e a tm e n t G r o u p

3 1 .33 1 .83 3 .8

1 4 4 2 3 91 4 5 2 3 81 3 8 2 3 9

E ve n ts T o ta l M e d i a n ( m o s )

Progression Free Survival Optimal Stage III NGR

0.0 0 12 24 36 48 60 72

Months on Study

1 239 203 141 97 66 21 02 238 209 152 103 72 21 03 239 204 150 104 76 24 0

Across Study Comparisons for PFSArm Study PFS Median in mos

No visible dx Stage 3PFS median mos1 cm or less visible dx

GOG 114 & 172 IV cisplatin 33.4

GOG 172 IV cisplatin 43.2 18.3

GOG 252 IV carbo 31.3 26.8 (10% stage II)

GOG 114 & 172 IP cisplatin 43.2

GOG 172 IP cisplatin 60.4 23.8

GOG 252 IP carboplatin 31.8 28.7 (10% Stage II)

GOG 252 IP cisplatin 33.8 27.8 (10% Stage II)

Discussion

• Survival for optimal and no residual disease participants will not be available for a few years.

• Dose reductions of paclitaxel and cisplatin as well as cross- over mayhave compromised efficacy.

• Dose dense paclitaxel may have improved efficacy to allow us toabandon IP chemo- must we wait- combine both?

• Bevacizumab interactions could have clouded analysis

Conclusions

• All arms have excessive toxicity• Neurotoxicity is similarly high in all arms• Reserve changes in treatment recommendations until survival

data available for no residual disease high grade serous Stage III participants.

• IP Cisplatin increases bevacizumab associated HTN