Primary CNS Lymphoma

Andrés J. M. Ferreri

Unit of Lymphoid MalignanciesDepartment of Onco-Hematology

San Raffaele Scientific Institute, Milano, Italy

Management difficulties

- High proportion of elderly pts

- Poor PS at presentation

- Biopsy not performed

- Palliative treatment

- Therapeutic consensus is lacking

- A few centers with adequate expertise

- Many pts can not be referred to other centers

• Several patients receive steroids for months before biopsy:

– Confounding effect on neuroimaging

– Delayed and unsuitable biopsy (52% inter-observer variability)

– Diabetes and other metabolic disorders

– Immunodepression (severe infections)

– Half of cases of early PD are related to interruptions due to toxicity

• CNS tissues exposed to lymphoma infiltration by months:

– Tissue damage results in poor PS and disabling symptoms

– Loss of autonomy and poor treatment tolerability

– CR and cure do not result in neurological and PS improvement

– Therapeutic interruptions due to poor, irreversible conditions

– Negative effects on trials accrual

Early Diagnosis is the Best Therapy

Ferreri AJM. EHA meeting 2016

PCNSL suspicion

Current strategy= low diagnosis sensitivity

• Neuroimaging: T1, T2, flair, DWI, enhancement, spectroscopy

• Site: corpus callosum , basal ganglia, periventricular areas, …

• Response to steroids

Neuroimaging

PCNSL GBM

Response to Corticosteroids

•Multiple sclerosis•Acute disseminated encephalomyelitis•Cerebral infarction•Neurosarcoidosis •Germinoma •Renal cell carcinoma metastases•Prolactinoma•Hemangioma

Response to steroidsLymphoma

Response to Steroids

Bp: Glioblastoma multiformeBp: no tumor

Early Reliable Suspicion

Reliable molecular and biological parameters that can be easily incorporated in routine practice.

Some chemokines (CXCL13) can be used as diagnostic & prognostic tools.

IL-10 concentration in the CSF is a useful diagnostic and prognostic biomarker.

Some miRNA (21, 19b, 92a) are expressed in the CSF of PCNSL patients, with a diagnosis sensitivy and specificity >95%

Recurrent mutations of CD79B (83%) and MYD88 (76%) in tissue samples.

MYD88mutations can be detected in the vitreous and plasma (CSF?).

The combined use of ADC, CSF CXCL13, and IL-10 results in increased diagnostic performance in CNSL.

Rubenstein J, et al. Blood 2013; Fisher L, et al. CCR 2009; Nguyen-Them L, et al. EJH 2016; Baraniskin A, et al. Blood 2011; Bonzheim I, et al. Blood 2015;

Fontanilles et al. ASH 2015; Nakamura T, et al. Neuropathol Appl Neurobiol 2016; Mabray MC, et al. AJNR 2016

Modern Approach

Updated from Ferreri AJM, ASCO 2012

INDUCTION CONSOLIDATION

HD-MTX poly

WBRT

Others

Age & PS

Comorbidity

Prognostic score

Histotype (DLBCL)

Therapeutic Dilemma

Ferreri AJM, Blood 2011

the dilemma posed by PCNSL treatment is the choice between strategies designed to intensify therapy to improve cure rate and treatment de-escalation strategies to avoid neurotoxicity.

Efficacy Neurot oxicity

Chemotherapy

Its efficacy is limited by several factors including the biology and microenvironment of this malignancy, which is “protected” by the BBB.

BBB penetration

Doses CNS availability

Examples

Good conventional good steroids, alkylating ag.

Low to moderate

high good MTX, araC

Poor conventional

(-limiting tox)

low anthracyclines, vinca-alkaloids

CHOP regimen

Mead GM, et al. Cancer 2000

WBRT 40 + 14 Gy; n=15

WBRT + CHOP; n=38

Pharmacokinetics Triphasic plasmatic clearance

Good BBB penetration at HD

Schedule Infusion duration 3 hours

Infusion timing every 2 wks = 3 wks

Dose 3 g/m2

HD-MTX

CNS availability 1 g/m2 tumoricidal levels in the brain

3 g/m2 tumoricidal levels in the CSF

24-hr inf. tumoricidal levels in the CSF

Tolerability 8 g/m2 45% dose reductions

3.5 g/m2 good compromise

Ferreri AJM. Blood 2011

MTX + Alkylator + Rituximab

INDUCTION CONSOLIDATION N° ORR 2-year PFS

Rituximab

Methotrexate

Procarbazine

Vincristine1

low-dose WBRT 52 79% 57%

Rituximab

Methotrexate

Procarbazine

Vincristine2

TBC - ASCT 33

(≤ 65 ys)

94% 79%

Rituximab

Methotrexate

Temozolomide3

Non-myeloablative

HD-cytarabine

HD-etoposide

44 77% 59%

Rituximab

Methotrexate

Temozolomide4

Hyperfract WBRT

+ TMZ maintenance

53

(<60 yo: 62%)

57% 64%

1Morris PG, et al. JCO 2013; 2Omuro A, et al. Blood 2015; 3Rubenstein JL, et al. JCO 2013; 4Glass J, et al. JCO 2016

The IELSG #32 trial

Ferreri AJM, et al. 13-ICML, Lugano 2015

WBRT 40 Gy

± boost 9 Gy

CR – PR - SD PD – tox

SC harvest

®WBRT 36 Gy

± boost 9 Gy

BCNU 400 mg/m2 d.1

Thiotepa 5 mg/Kg x 2/d; d.2-3

+ APBSCT

PCNSL [≤ 65 ys. + PS 0-3] or [65-70 ys. + PS ≤2]

®4 c. MTX 3.5 g/m2 d.1

araC 2 g/m2 x 2/d, d. 2-3

every 3 weeks

4 c. rituximab 375 mg/m2 d-5 & 0

MTX 3.5 g/m2 d.1

araC 2 g/m2 x 2/d, d. 2-3

every 3 weeks

4 c. rituximab 375 mg/m2 d-5 & 0

MTX 3.5 g/m2 d.1

araC 2 g/m2 x 2/d, d. 2-3

Thiotepa 30 mg/m2 d.4

every 3 weeks

Arms Activity

Ferreri AJM, et al. Lancet Haematol 2016

0.74

0.0007

0.02

0.69

0.00001

0.05

Activity: Arm and IELSG risk

Logit CR OR

IELSG risk score 0,13 0,09

Arm 0,0004 0,000004

PFS and OS

median follow-up: 30 months (12-66)

Ferreri AJM, et al. Lancet Haematol 2016

Chemotherapy: Elderly Patients

HD-MTX improved outcome in selected pts (biased results).

The age upper limit to define elderly pts remains uncertain.

Ferreri AJM. Blood 2011

Elderly Pts: PHRC 2006 Trial

Omuro A, et al. Lancet Haematol 2015

Arm A M-PVA

Arm B M-TMZ

3 cycles/ 28 d

MTX 3,5 g/m2 d1

Vincristine 1,4 mg/m2 D1

Procarbazine 100 mg/m2/d D1-

7

D7D1 D14 D21 D28

Vincristine 1,4 mg/m2 D1

Methylprednisolone

60 mg/d D1-5

Cytarabine 3 g/m2/d1-2After 3rd Cycle

MTX 3,5 g/m2 d1

TMZ 150 mg/m2/d D1-5 If no tox= TMZ 150 mg/m2/d D15-19 , cycle 2 & 3

D7D1 D14 D21 D28

Methylprednisolone

60 mg/d D1-5

MTX 3,5 g/m2 d1 MTX 3,5 g/m2 d1

3 cycles/28 d

AGE ≥ 60 YEARS

PHRC 2006 Trial

Omuro A, et al. Lancet Haematol 2015

MPV-A

(n= 47)

M-TMZ

(n= 48)

p

CR

PR

SD

PD

62%

20%

2%

16%

45%

26%

7%

22%

0.11

ORR 82% 71% 0.23

AGE ≥ 60 YEARS

Elderly pts: PRIMAIN Trial (n= 108)

Kindly provided by G. Illerhaus and B. Kasenda

TABLE 1: Best response from 4 MRI examinations as per protocol. 108 patients.

Best response Values

CR 46 (42.6%)

PR 34 (31.5%)

PD 12 (11.1%)

SD 1 (0.9%)

Missing 15 (13.9%)

AGE ≥ 65 YEARS

Sanctuaries CSF and eyes (intrathecal and intravitreal chemo).

IT/IV chemo efficacy has not been prospectively confirmed. Most trials do not include IT/IV drug delivery.

IT is associated with additional risk of infective complications, neurotoxicity and chemical meningitis.

HD-MTX ( 3 g/m2) treats adequately meninges.

IVi: is active, but toxic (visual acuity deterioration in 27%).

Impact on OS???Ferreri AJM, et al. Neurology 2002Ferreri AJM, et al. J Clin Oncol 2003Pels H, et al. J Clin Oncol 2003

Weigel R, et al. Clin Neurol Neurosurg 2004Batchelor T, et al. Clin Cancer Res 2003

Smith JR, et al. Ophthalmology 2002

High-dose Ifosfamide

Mappa, HemOnc 2013; Arellano-Rodrigo, EJH 2003; Choquet, Lugano 2013 Abs #3664; Motomura, L&L 2011; Choi, IJH 2013

R-IE (n= 22) VIA (n= 16) ICE (n= 17) ICED (n=25) De-VIC

Line Salvage Salvage Salvage Salvage First

ITX (g/m2/d)x[days] 2 x [3] 1 x [5] 5 x [1] 1.5 x [5] 1,5 x [3]

Other drugs R, VP16 araC; VP16 CBDCA; VP16 CBDCA; VP16 CBDCA; VP16

Previous chemo MA CHOD, MA MA M -

Pre-irradiated pts 55% 100% NR 27% -

Median age 60 (39-72) 54 (31-69) 62 (28-84) 58 (20-73) 61 (19-79)

Refractory (mPFS) 50% (8 mo) 6% (19 mo) 24% (12 mo) 36% (12 mo) 0

Dose reduction 0% NR 24% NR NR

NF (TRM) 14% (5%) 50% (0%) 53% (6%) NR (8%) 10% (0%)

ASCT 20% 0% 35% 52% NA

CRR 27% 37% 76% (ASCT) 48% 62% (2c.)

mPFS 4.0 mo 4.5 mo 2.6 mo 11 mo 37 mo

mOS 6.0 mo 6.0 7.3 mo 27 mo 48 mo

Salvage Single-Agent in Trials

Regimen N ORR m TTP G3-4 N G3-4 T TD

RituximabBatchelor T, et al. Neurology 2011

12 42% 8 0% 0% 0%

TemozolomideReni M, et al. Br J Cancer 2007

36 31% 7+ 6% 3% 0%

TopotecanVoloschin A, et al. JNO 2008

15 40% 3 73% 20% 0%

TopotecanFischer L, et al. Ann Oncol 2006

27 33% 9 25% 11% 13%

PemetrexedRaizer JJ, et al. Cancer 2012

11 55% 6 63% 50% 13%

TemsirolimusKorfel A, et al. JCO 2016

37 54% 2 20% 22% 14%

Chia-Ching W, et al. BJH 2014

Ponzoni M, et al. Ann Oncol 2014

• 13 PCNSL; 7 SCNSL

• Ibrutinib 560 mg and 840 mg was well tolerated.

• No DLT

• Meaningful concentrations in the CSF.

• ORR= 75%

• Median PFS= 5 months.

• Ongoing MTX-IBT-R trial

Ibrutinib at ASH 2016

Grommes C, et al. ASH 2016

• 8 c. LENA 20-25 mg + ritux => 12 c LENA 10 mg

• 50 pts with rrPCNSL

• ORR= 67% (36% after induction) EARLY

• 24% of pts received 1 course

• 50% of pts received 2 courses

• 36% completed induction

• 1-yr PFS: 20%

• Dose reduction in 42%

• Prior vs. LENA median PFS= 4 vs. 8 months

Lenalidomide at ASH 2016

Ghesquieres H, et al. ASH 2016

Reactive Perivascular T-cell Infiltrate

CD20

CD3

NEG (n= 34)

POS (n= 16)

Reactive Perivascular T-cell Infiltrate

Ferreri AJM. Frye-Halloran Conference, Boston 2004

SALVAGE THERAPY: NIVOLUMAB

CA209-647: Multicenter open-label, two-cohort, single-arm phase IIPATIENT POPULATION: Rel/Ref PCNSL and Rel/Ref PTLEnd Points: Safety and Efficacy of NivolumabStatus: 58 c./14 countries; 3 Italian centers. 1 pt enrolled. OSR will be open 9/11/16

FOUR ANECDOTAL CASES OF CR

Modern Approach

Updated from Ferreri AJM, ASCO 2012

INDUCTION CONSOLIDATION

HD-MTX poly

WBRT

Others

Age & PS

Comorbidity

Prognostic score

Histotype (DLBCL)

Response

Quality response

Induction

Observation

WBRT

HDC/ASCT

Non-myeloablative

Maintenance

Radiation Field and Doses

+

+

RESPONSE

COMPLETE REMISSION

30-36 Gy

PARTIAL RESPONSE

36-40 Gy

10 Gy

PROGRESSIVE DISEASE

40-45 Gy

10 Gy

30 Gy

36

Neurotoxicity

Correa D, et al. Neuro-Oncol 2009

Poor QoL and

>50% were not working due to illness

Reducing Neurotoxicity Risk

Ferreri AJM, et al. Blood 2011

To avoid consolidation RT (only CRs).

To improve radiation parameters.

To replace RT with other strategies.

Consolidation RT withdrawal?

Thiel E, et al. Lancet Oncol 2011

G-PCNSL-SG-1 trial

551 pts with newly diagnosed PCNSL were enrolled from 75 German Centers and treated between 2000 and 2009

G-PCNSL-SG-1 trial: results

Low-dose WBRT

Morris PG, et al. JCO 2013; Correa DD, et al. JNO 2009; Kim BH, et al. Cancer Res Treat. 2014

Consolidative HDC/ASCT

MTX± others

ThiotepaBus, CTX

60 72 1421

Alimohamed N, et al. L&L 2012

56 (34-69)PS>1: 70%

24 100

MTXaraC, TTP

ThiotepaBCNU + RT

140 81 330

Kasenda B, et al. Ann Oncol 2012

54 (27-64)70 (30-100)

37 77

MTXaraC

BEAM 28 20 028

Abrey L, et al. JCO 2003

53 (25-71)70 (30-100)

18 50

MVpBP+itx/araC

BEAM + RT 34 60 425

Colombat P, et al. BMT 2006

51 (21-60)PS3-4: 32%

44 68

MTXaraC, TTP

ThiotepaBCNU ± RT

72 77 013

Kasenda B, et al. Ann Oncol 2012

54 (38-67)90 (30-100)

54 85

MTXaraC

Bus, CTXVP16 ± RT

25 30 011

Yoon DH, et al. BMT 2011

52 (33-65)PS1: 91%

73 100

MTX ThiotepaBusulfan

13 15 45 1323

Montemurro M, et al. Ann Oncol 2007

55 (18-70)70 (30-100)

70

TRM(%)

Induction Conditioning F-up(mo)

2-yr EFS(%)

N° Age m(r)PS m(r)

CRR(%)

ASCT(%)

Activity of HDC/ASCT

Kindly provided by G. Illerhaus, Stuttgart, Germany

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

after2nd MTX

after4th MTX

after1st AraC/TT

after2nd AraC/TT

afterHD (d30)

offStudy

n.d.

PD

SD

PR

CR

(central neuroradiological review)

ASCT vs. Alternatives

Ferreri AJM & Illerhaus G. Blood 2016

IELSG32: WBRT vs. ASCT

PRECIS: WBRT vs. ASCT

IELSG43 (MATRix): ASCT vs. NMC

ALLIANCE: ASCT vs. NMC

The IELSG #32 trial

Ferreri AJM, et al. 13-ICML, Lugano 2015

WBRT 40 Gy

± boost 9 Gy

CR – PR - SD PD – tox

SC harvest

®WBRT 36 Gy

± boost 9 Gy

BCNU 400 mg/m2 d.1

Thiotepa 5 mg/Kg x 2/d; d.2-3

+ APBSCT

PCNSL [≤ 65 ys. + PS 0-3] or [65-70 ys. + PS ≤2]

®4 c. MTX 3.5 g/m2 d.1

araC 2 g/m2 x 2/d, d. 2-3

every 3 weeks

4 c. rituximab 375 mg/m2 d-5 & 0

MTX 3.5 g/m2 d.1

araC 2 g/m2 x 2/d, d. 2-3

every 3 weeks

4 c. rituximab 375 mg/m2 d-5 & 0

MTX 3.5 g/m2 d.1

araC 2 g/m2 x 2/d, d. 2-3

Thiotepa 30 mg/m2 d.4

every 3 weeks

TOXICITY (PP)

Grade 1-2 3 4 5 1-2 3 4 5

Neutropenia 3 ( 5%) 1 (2%) 3 (5%) 0 (0%) 2 ( 3%) 0 ( 0%) 41 (71%) 0 (0%)

Thrombocytopenia 0 ( 0%) 1 (2%) 1 (2%) 0 (0%) 0 ( 0%) 3 ( 5%) 42 (72%) 0 (0%)

Anaemia 7 (13%) 2 (4%) 0 (0%) 0 (0%) 21 (36%) 17 (29%) 2 ( 3%) 0 (0%)

FN/infections 0 ( 0%) 1 (2%) 0 (0%) 0 (0%) 1 ( 2%) 12 (21%) 3 ( 5%) 2 (3%)

Hepatotoxicity 3 ( 5%) 1 (2%) 0 (0%) 0 (0%) 15 (26%) 3 ( 5%) 1 ( 2%) 0 (0%)

Nephrotoxicity 2 ( 4%) 0 (0%) 0 (0%) 0 (0%) 6 (10%) 0 ( 0%) 0 ( 0%) 0 (0%)

Cardiotoxicity 0 ( 0%) 0 (0%) 0 (0%) 0 (0%) 5 ( 9%) 1 ( 2%) 0 ( 0%) 0 (0%)

Coagulopathy/DVT 2 ( 4%) 0 (0%) 3 (5%) 0 (0%) 6 (10%) 1 ( 2%) 0 ( 0%) 0 (0%)

Gastrointestinal 8 (15%) 0 (0%) 0 (0%) 0 (0%) 24 (41%) 11 (19%) 0 ( 0%) 0 (0%)

Mucositis 2 ( 4%) 0 (0%) 0 (0%) 0 (0%) 15 (26%) 12 (21%) 3 ( 5%) 0 (0%)

Erythema 7 (13%) 0 (0%) 0 (0%) 0 (0%) 0 ( 0%) 0 ( 0%) 0 ( 0%) 0 (0%)

Acute neurotoxicity 7 (13%) 3 (5%) 0 (0%) 0 (0%) 4 ( 7%) 0 ( 0%) 0 ( 0%) 0 (0%)

WBRT (n= 55) ASCT (n= 58)

NEUROTOXICITY 57 patients (30 WBRT and 27 ASCT) were assessable for effects of treatments

on cognitive functions and QoL.

Pre-morbid IQ estimation by NART test showed a mean value ± standard error

of 100 ± 7 for WBRT patients and 105 ± 8 for ASCT patients (p= 0.40).

Neuropsychological tests performed immediately after treatment conclusion

showed a fast improvement in the majority of assessed cognitive functions and

QoL, which was more evident for memory and language in arm E.

A significant impairment of some attention/executive functions among patients

treated with WBRT was recorded, while results of tests addressing language

and memory were stable after this therapy.

Patients treated with ASCT exhibited improved functions at most of the

performed test, and improved QoL figures.

ACTIVITY

Afterinduction

32 CR (54%)

Arm D

(WBRT)CR (95%)

Afterinduction

31 CR (53%)

Arm E

(ASCT)CR (93%)

EVENTS: 45 ARM D (%) ARM E (%)

Relapse 16 (27%) 18 (31%)

PD during consolidation 2 ( 3%) 2 ( 3%)

Toxic death 0 ( 0%) 2 ( 3%)

Death off-therapy (NED) 2 ( 3%) 3 ( 5%)

EVENTSMEDIAN FOLLOW-UP: 40 MONTHS (24-76)

EFFICACY: OS

102 (47%) pts are alive:

- 44 (75%) in arm D

- 37 (63%) in arm E

- 21 pts excluded from R2.

Causes of death Arm D Arm E Non R2

lymphoma 11 15 62

treatment toxicity 0 2 13

toxicity during salvage therapy 1 1 0

neurocognitive decline while NED 0 1 1

late infective complications 3 1 2

car accident 0 0 1

acute erythroid leukaemia 0 1 0

sudden death (> 1 year) 0 1 0

unknown 0 0 1

ITT

PP

Schorb E, et al. BMC Cancer 2016

Non-Myeloablative Chemo

Rubenstein J, et al. JCO 2013

Alliance/CALGB 50202 trial

MTX (8)RituximabTMZ x 8 c.

44 pts(age: 12-76)

CR (66%)araC (8)96-hr VP16

No neurotox

Median f-up: 4.9 ys

21 failures 17 deaths

TRM (sepsis) 2%

Nordic Trial: TMZ maintenance

Pulczynski EJ, et al. Haematologica 2015

All PCNSL pts ≥ 70 years old

Trial registration

Eligible for therapy (PS ≤3 – Not eligible for ASCT)

Elegible for HD-MTX Inelegible for HD-MTX

Ineligible for therapy

WBRT 2340 cGy+ Temozolomide

+ Rituximab

Temozolomidex 12 mo

RIT-MTX-PCZ x 2 c.(PRIMAIN trial)

Lenalidomidex 24 mo

PCZx 6 mo

Strategies for Future Studies

To potentiate early diagnosis

To identify new active drugs

To amply our biological and molecular knowledge

To establish reliable prognostic factors & potential targets

To enhance drug bioavailability

To improve radiation therapy

To reduce neurotoxicity and improve patients’ QoL

To improve international cooperation

Ferreri AJM, et al. JCO 2003; Ferreri AJM, et al. JCO 2013

Trend in Survival

Shields MS, et al. BJH 2016 Zeremski V, et al. Ann Hematol 2016

From SHARED IDEAS to FACTS

PCNSL

Young pts

First line

PCNSL

Elderly pts

First line

SCNSL

Young pts

First line

PCNSL

Young pts

Salvage

PCNSL

Elderly pts

Salvage

SCNSL

Young pts

Salvage

IELSG #42 (MARIETTA) trial

IELSG #32 trial IELSG #43 (MATRIX) trial

SCNSL1 Trial

IELSG #XX

(FIORELLA) trialTIER trial

European PCNSL Collaborative Group

• Our patients and their families

• National Coordinators and DMSC Offices

• Hematologists, oncologists, neuro-radiologists, radiation oncologists, pathologists,

researchers, psychologists, data managers and research nurses of participating centers

• Colleagues, data managers, co-chairs and friends of the International Extranodal

Lymphoma Study Group (IELSG)

• Institutions supporting our trials: Agenzia Italiana del Farmaco, Cancer Research UK,

Oncosuisse and Swiss National Foundation

• Colleagues, data managers and friends of the Fondazione Italiana Linfomi (FIL)

• Colleagues and friends of the European PCNSL Collaborative Group (EPCG)

• Colleagues and friends of the International PCNSL Collaborative Group (IPCG)

• Unit of Lymphoid Malignancies of the San Raffaele Scientific Institute, Milano

Acknowledgments