IMUNOTERAPIA NO CÂNCER DE PULMÃO: PRINCÍPIOS

Dr. Gilberto de Castro Junior Professor Colaborador Livre-Docente – Faculdade de Medicina da USP. Serviço de Oncologia Clínica - Instituto do Câncer do Estado de São Paulo Centro de Oncologia - Hospital Sírio Libanês Grupo Brasileiro de Oncologia Torácica – GBOT/LACOG São Paulo - BRASIL

Potenciais conflitos de interesses

Apoio em participação de eventos de cunho científico Roche, AstraZeneca, MSD, BMS, Boehringer-Ingelheim,

Novartis, Bayer, Eurofarma

Dr. Gilberto de Castro Junior CRM:84448

Investigador de ensaios clínicos patrocinados Investigador de ensaios clínicos patrocinados

Aulas e apresentações Roche, AstraZeneca, BMS, MSD, Merck Serono,

Eurofarma, Pfizer

Consultorias científicas Roche, AstraZeneca, MSD, Merck Serono, Eurofarma,

Boehringer-Ingelheim, Pfizer

Kandoth et al. Nature 2013

Mutational landscape across 12 major cancer types

Pardoll. Nat Rev Cancer 2012

T cell activation in lymph node

Ribas. NEJM 2015

T cell activation in tumor milieu

Ribas. NEJM 2015

A Via do PD-L1 inibindo a Resposta Imune

MHC

PD-L1

PD-1 PD-1

PD-1 PD-1

Nivolumab PD-1 Receptor Blocking Ab

Recognition of tumor by T cell through MHC/antigen interaction mediates IFNγ release

and PD-L1/2 up-regulation on tumor

Priming and activation of T cells through MHC/antigen & CD28/B7 interactions with

antigen-presenting cells

T-cell receptor

T-cell receptor

PD-L1 PD-L2

PD-L2

MHC

CD28 B7

T cell

NFκB Other

PI3K Dendritic

cell Tumor cell

IFNγ

IFNγR

Shp-2

Shp-2

http://www.ackc.org/nivolumab-mechanism-cartoon/

PD-L1 Expression Associated with Favorable Outcome With Pembrolizumab

1. Garon EB et al. N Engl J Med 2015;372:2018-28.

• TPS ≥50% cutpoint rigorously determined using independent training and validation sets derived from KEYNOTE-0011

• PD-L1 IHC 22C3 pharmDx (Dako) approved in the US as a companion diagnostic for pembrolizumab

Negative TPS 1%–49% TPS ≥50%

20x

40x

Keynote 010: PD-L1 Expression Correlates with improved OS in Advanced NSCLC

Imagens de IHC images de três amostras diagnósticas anti-PDL-1

Marianne J. Ratcliffe et al. Clin Cancer Res 2017;23:3585-3591

PD-L1 expression on tumor cells

Journal of Thoracic Oncology Vol. 12 No. 2: 208-222

Each dot represents the mean score of 3 pathologists

0

10

20

30

40

50

60

70

80

90

100

% T

um

or

Sta

inin

g

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39

Cases

SP263SP14228-822C3

Conclusion 1: 3 assays showed similar staining

characteristics for PD-L1 staining on tumor cells, but

SP142 comparatively showed less tumor cells stained

David P Carbone et al. N Engl J Med 2017; 376:2415-2426

Phase 3 CheckMate 026 Study Design: Nivolumab vs Chemotherapy in First-line NSCLC

Primary endpoint: PFS per BIRC (≥5% PD-L1+)

Secondary endpoints:

•PFS per BIRC (≥1% PD-L1+)

•OS

•ORR

Exploratory objective: Predictive biomarkers for outcomes with nivolumab

Nivolumab 3 mg/kg IV Q2W

n = 271

Randomize 1:1

Key eligibility criteria:

• Stage IV or recurrent NSCLC

• No prior systemic therapy for advanced disease

• No EGFR/ALK mutations sensitive to available targeted inhibitor therapy

• ≥1% PD-L1 expression Chemotherapy

(histology dependent) Maximum of 6 cycles

n = 270

Disease progression or unacceptable toxicity

Disease progression

Crossover nivolumab (optional)

Tumor scans Q6W until week 48 then Q12W

Stratification factors at randomization:

• PD-L1 expression (<5% vs ≥5%)

• Histology (squamous vs non-squamous)

An exploratory analysis was conducted in CheckMate 026 to test the hypothesis that patients with high TMB may derive enhanced benefit from nivolumab

Exploratory TMB Methods CheckMate 026 TMB Analysis: Nivolumab in First-line NSCLC

N3

Whole exome sequencinga

Tumor DNA

Germline DNA (blood)

Somatic missense mutations

Tumor exome data

Germline exome data

TMB

Sample size throughout TMB determination

Patients, n (%) Tumor DNA Germline

DNA

Randomized 541 (100) 541 (100)

Samples available for DNA extractiona 485 (90) 452 (84)

DNA available for sequencing 408 (75) 452 (84)

Successful preparation of next-generation sequencing library

402 (74) 452 (84)

Passed internal quality controlb

320 (59) 432 (80)

Matched tumor-germline exome sequences for TMB analysisc

312 (58)

aSamples were not available for various reasons, including but not limited to lack of patient pharmacogenetic consent, samples exhausted for PD-L1 testing, or poor tissue sampling bInternal quality control failure included factors such as discordance between tumor and germline DNA, too few sequence reads, and low or uneven target region coverage c8 patients with available tumor DNA sequences did not have matched germline DNA sequences

aDNA was sequenced on the Illumina HiSeq 2500 using 2 × 100-bp paired-end reads; an average of 84 and 89 million reads were sequenced per tumor and germline sample, respectively (average 84.6 × and 93 × the mean target coverage, respectively)

David P Carbone et al. N Engl J Med 2017; 376:2415-2426

PFS by Tumor Mutation Burden Tertile CheckMate 026 TMB Analysis: Nivolumab in First-line NSCLC

100

90

80

70

60

50

40

30

20

10

0

0 3 6 9 12

Months

15 18 21 24

PF

S (

%)

High

Low

Medium

Medium n = 49 n = 47

3.6

(2.7, 6.9)

Low n = 62

4.2

(1.5, 5.6)

9.7

(5.1, NR)

Median PFS, months

(95% CI)

High

Nivolumab Arm Chemotherapy Arm

Medium n = 53 n = 60

6.5

(4.3, 8.6)

Low n = 41

6.9

(5.4, NR)

5.8

(4.2, 8.5)

Median PFS, months

(95% CI)

High 100

90

80

70

60

50

40

30

20

10

0

0 3 6 9 12

Months

15 18

High

Low

Medium

21

Data for patients with low and medium TMB were pooled in subsequent analyses

David P Carbone et al. N Engl J Med 2017; 376:2415-2426

Characteristic

High TMB (n = 107)

Low/medium TMB (n = 205)

Median age, years (range) 65 (40, 87) 65 (32, 89)

Female, % 36.4 42.0

ECOG PS, % 0 1/2

31.8

67.3/0.9

32.2

66.3/1.0

Smoking status, % Current smoker Former smoker Never smoker

22.4 73.8 2.8

15.6 70.2 12.7

Disease stage, % Stage IV Recurrent

91.6 7.5

94.1 5.9

Tumor histology, % Squamous Non-squamous

29.0 71.0

19.5 80.5

PD-L1 expression level, % ≥5% ≥25% ≥50%

77 60 45

83 59 40

Baseline Characteristics According to TMB Subgroup- CheckMate 026 TMB Analysis:

Nivolumab in First-line NSCLC

N3

500

300

1000

10

50

100

200

Current Former

Smoking Status

Never Unknown

No

. o

f M

iss

en

se

Mu

tati

on

s 400

David P Carbone et al. N Engl J Med 2017; 376:2415-2426

Anti-PDL1 + anti-CTLA4

Anti-PDL1 + anti-CTLA4

N3

• CTLA-4 inhibition may also induce compensatory signals, including PD-L1 upregulation, that in turn dampen the immune response

• PD-L1 expression reduces T-cell activation by binding to two important regulatory receptors:

- Binding to PD-1 delivers an inhibitory signal that reduces cytokine production and T-cell proliferation

- Binding to CD80, further reduces CD28 co-stimulation of T cells

• PD-L1 blockade may overcome this immune checkpoint, resulting in prolonged T-cell activation

Combinação anti-CTLA4 & anti-PD1

Tumour cell

T cell

PD-L1

TCR

MHC

PD-1 PD-L1 PD-1

CD80

CD80

Inhibition

CD80

Inhibition

Inhibition

Inhibition

Activation

TCR MHC

PD-L1

CD80

CD86

CTLA-4

CD28

Inhibition

Activation

Tumour antigen

Immune

cell

Combinação anti-CTLA4 & anti-PD1

Tumour cell

T cell

PD-L1

PD-1 PD-L1 PD-1

Activation

CD80 CD80

CD80

CD86 CTLA-4

CD28 Activation

Anti-CTLA4

Anti-CTLA4 blocks CTLA-4 binding to CD80 and CD86

–

TCR MHC

Anti-PD(L)1

–

Anti-PD(L)1

Blocking PDL1-PD1 & PDL1-CD80 interaction

PD-L1

CD80

TCR MHC

Tumour antigen

Immune cell

Inhibition

–

–

Combinação anti-CTLA4 & anti-PD1

N3

Das et al. J Immunol 2015

Munn, Mellor. J Clin Invest 2007

Indoleamina 2,3-dioxigenase

• Epacadostat is an oral inhibitor of IDO1

• Phase 1/2 study

- Safety and efficacy of oral epacadostat plus IV nivolumab in pts with advanced tumors

• No DLT was observed in P1 (36 pts)

• Most common TRAEs (≥15%) in pts treated with E 100 mg (n = 70) and E 300 mg (n = 135) were rash (33% and 22%), fatigue (26% and 31%), and nausea (24% and 19%).

• For the 23 SCCHN pts treated with E 300 mg, preliminary DCR was 70% (n = 16).

• Of 30 MEL pts, 8 were treated with E 100 mg and 22 with E 300 mg. ORR 75% (n = 6; all PR) and and DCR 100% (n = 8; 2 SD) (E 100 mg).

Perez et al. ASCO 2017

ECHO-204 trial

• Como identificar os pacientes candidatos?

- Biomarcadores preditivos

• Emergência da resistência ao tratamento

- Diagnóstico e tratamento

• Identificação e tratamento de toxicidades

- Educação da equipe e do paciente

• Custo da medicação e acesso

- Farmacoeconomia: custo-efetividade local

Desafios