Immunotherapy in GI cancers
Julien TAIEB
Sorbonne Paris Cité
Hopital Européen Georges Pompidou
Inserm U970
J Taieb ESDO Leuven 2017
How to progress in GI cancer treatmentBetter define subgroups
With different prognostic and different treatments
Combine different targeted agents in accordance with the molecular profile
of the patients tumor
Find predictive markers for precision medecine
Induce an efficient immune response from the host
against the tumor
J Taieb ESDO Leuven 2017
Immunité innée et adaptativeInnée :
- Rapide
- Non spécifique
Adaptative :
- Plus lente
- Spécifique
Immunité anti-tumorale
Terme, Taieb et al. Immunooncology 2014
Immature DC
mature DC
TUMOR
CD8+ TLymphocyte
CD4+T Lymphocyte
Lyse
antigens
Peptide / MHC IPeptide / MHC II
Major immunotherapeutic approaches
Checkpoint blockade
Anti-cancervaccines
Adoptive transfer
J Taieb ESDO Leuven 2017
1. Parkhurst MR, et al. Mol Ther 2011;19:620–626.
Anti-CEA T cell transfer
CEA decrease
(maximum decline74–99% vs baseline)
Radiological response Inflammatory colitis
J Taieb ESDO Leuven 2017
Immune check points
Mellman I et al., Science, 2011
J Taieb ESDO Leuven 2017
Immune check points
Mellman I et al., Science, 2011
ActivateThe activators
InhibitThe inhibitors
J Taieb ESDO Leuven 2017
.
Immune check points
Pardoll DM et al., Nat Rev Cancer, 2012
T cellregulation
T cell function
J Taieb ESDO Leuven 2017
In 2017, Efficacy in phase III
Anti-CTLA4
(Ipilimumab, tremelumumab)
• Melanoma• M+ (L1, L2)
• Adjuvant (st III)
Anti-PD(L)1
(Nivolumab, Pembro, Atezolizumab, avelumab, darvalumab…)
• Melanoma M+ (L1, L2) Nivo +Ipili +++
• Melanoma M+ and Lung M+ (L1 & L2)
• Lung M+ (L2) Nivo/Atezolizumab
• Kidney M+ (L2) Nivolumab
• Squamous Cell Cancer of the Head and Neck (L2). Nivolumab
• Bladder (phase 2: Atezolizumab)
PD1/PD-L1
Blockade
reported ORR
SCLC
15%
Mel
30-40%
RCC
15-20%
Bladder
25%
HNSCC
15-25%Gastric
20%
HCC
20%
Hodgkin
65-85%
MSI High
CRC
60%
TNBC
20%NSCLC
15-20%
Ovarian
15%
Colon cancer
-100
0
100
-100
0
100
-100
0
100
-100
0
100Melanoma1
-100
0
100NSCLC2
-100
0
100Gastric6
-100
0
100
-100
0
100H&N3 TNBC5
-100
0
100cHL7
-100
0
100NHL PMBCL8
Urothelial4
Ch
ang
e F
rom
Bas
elin
e in
Tu
mo
r S
ize,
%
-100
0
100
Mesothelioma9
-100
0
100
Anal14
-100
0
100
-100
0
100
SCLC11
-100
0
100NPC13
-100
0
100Biliary Tract15
-100
0
100Colorectal16
Esophageal12
-100
0
100
Ovarian10
-100
0
100ER+/HER2– BC17 Cervical18 Thyroid19 Salivary20
-100
0
100
J Taieb ESDO Leuven 2017
Checkpoint blockers Efficacy signal
in MSI-H colorectal cancer
Le DT, et al. N Engl J Med. 2015;372(26):2509-2520
Radiographic responses*
*RECIST-based radiographic response**Adjusted for elapsed time since the initial diagnosis
Treatment with pembrolizumab (anti-PD-1 antibody)(n=11 mismatch repair-deficient CRC, n=21 mismatch-repair proficient CRC, n=9 mismatch-repair deficient non-CRC)
OS in CRC
Adjusted OS HR for mismatch-repair deficient vs proficient CRC: 0.18, P = .05
Immune-related ORR in mismatch-repair deficient vs proficient CRC: 40% vs 0%
J Taieb ESDO Leuven 2017
Response to Pembrolizumab for MSI-H/dMMR
• 11 patients achieved a CR and were taken off therapy after 2 years of treatment.
• No evidence of cancer PD has been observed in those patients with a median time off therapy of 8.3 months.
Le DT, Science 2017
Best responseMSI-H CRC
N = 61a
MSI-H
Non-CRC
N = 77b
n % (95% CI) n % (95% CI)
ORR (95% CI), % 17 28% (17-41) 29 38% (27-49)
Complete response 0 0 2 3% (0-9)
Partial response 17 28% (17-41) 27 35% (25-47)
Stable disease 14 23% (13-36) 16 21% (12-32)
Progressive disease 28 46% (33-59) 24 31% (21-43)
DCR (CR+PR+SD) 31 51% (38-64) 45 58% (47-70)
Median duration of response (range), monthsc Not reached (2.9+ to 12.5+) Not reached (2.4+ to 9.2+)
Median time to response (range), months 4.0 (2-10) 2.1 (1-4)
Pembrolizumab for MSI-H/dMMR CRC and Non-CRC
Diaz LA, ASCO 2017, J Clin Oncol 35l(suppl 15), 3618J Taieb ESDO Leuven 2017
Keynote 16
Le DT et al., ASCO 2016, CSS 103
• ASCO update
• Tolerability :• Rash prurit 25% Diarrhea/Colitis 11%• Arthralgia 15% Thyroïditis/Hypothyroïdia 11%, • Fatigue 9%• grade 3-4 toxicities : 14%
including pancreatitis 4% and diarrhea /colitis / thrombocytopenia / leucopenia /anemiain 2% each
• Efficacy :• MSI : RR 28% with 11%CR ; DCR 51%
MSS : RR 0% ; DCR 16%
KEYNOTE 164 is ongoing in MSI-H (phase II, 120 pretreated pts)
J Taieb ESDO Leuven 2017
May 23, 2017
FDA GRANTS ACCELERATED APPROVAL TO PEMBROLIZUMAB
FOR FIRST TISSUE/SITE-AGNOSTIC INDICATION
• Adult and pediatric patients with unresectable or metastatic, MSI-H or dMMR solid tumors that have
progressed following prior treatment and who have no satisfactory alternative treatment options
• or with MSI-H or dMMR colorectal cancer that has progressed following treatment with a
fluoropyrimidine, oxaliplatin, and irinotecan
Non MSI GI cancer will be treatable by anti-PD1 in the future => MSI testing has to be performed in daily practice
J Taieb ESDO Leuven 2017
Phase 2 CheckMate 142 Study Design: MSI-H Cohort
19
Nivo 3 mg/kg (Q2W)
mStage 1
MSI-H
• Second-line colon MSI-H
• ≥ 1 prior treatment for metastatic disease
• ≥ 1 target lesion
• ECOG PS of 0-1
≥ 7/19 Nivo 3 mg/kg (Q2W)
mStage 2Responsesa
Responsesa
≥ 7/19
• Nivo 3 mg/kg + Ipi 1 mg/kg (Q3W x 4 doses)
• Then Nivo 3 mg/kg (Q2W)
cStage 2b
3–6/19
• Nivo 3 mg/kg +
Ipi 1 mg/kg
(Q3W x 4 doses)
• Then Nivo 3 mg/kg(Q2W)
cStage 1
Responsesa
aIn patients with centrally confirmed MSI-H statusbCurrently enrolling
cStage 1 = combination therapy stage 1; cStage 2 = combination therapy stage 2; Ipi = ipilimumab; mStage 1 = monotherapy stage 1; mStage 2 = monotherapy stage 2; Nivo = nivolumab; Q2W = every 2 weeks; Q3W = every 3 weeks
J Taieb ESDO Leuven 2017
Patient Demographics and Disease
Characteristics
Overman M, Lancet Oncol 2017
Overall Response and Disease ControlCheckMate 142 Nivolumab alone
Overman M, Lancet Oncol 2017
Median time to response is 2,8 (1,4-3,2)
Only 3 patients with ORR experiencing progression
Median duration of response are not yet reached and are responder are alive at the time of analysis
Changes in Tumor Burden CheckMate 142 Nivolumab alone
Overman M, Lancet Oncol 2017
• Median TTR: 2.8 (1.1–14.0) months
• Median DOR: not reached
• 85% (39/46) of responses ongoing
Tt discontinuations
38 pts( 51%) :
27 (36%) for PD
6 (8%) Toxicity
5 (7%) other
Progression-Free Survival
Median PFS, mo 14,3 months
PFS rate, % (95% CI)
12 months 50 % (38, 61)
Overman M, Lancet Oncol 2017
Overall Survival
Median OS, mo (95%
CI)
NR (18, NE)
OS rate, % (95% CI)
12 months 73% (62, 82)
23 death/74 patients
36 investigator assessed PD
Median Follow up 12 months
CheckMate 142 Nivolumab alone
Investigator-assessed ORR and disease control (n=74)
CheckMate 142 Nivolumab alone
Overman M, Lancet Oncol 2017
dMMR/MSI-H per Local Assessment (N = 74)a
Patients, n (%) Grade 1 or 2 Grade 3 or 4
Any Event 36 (49%) 15 (21%)
TRAE reported in ≥ 10% of patientsFatigueDiarrheaPruritus HypothyroidismLipase increasedRashColitisAdrenal insufficiencyIncrease ASAT/ALAT
16 (22.%)15 (20%)10 (14%)7 (10%)3 (4%)
8 (11%)00
5 (7%)
1 (1%)1 (1%)
00
6 (8%)0
1 (1%)1(%)
1 (1%)
.
Safety SummaryCheckMate 142 Nivolumab alone
Overman M, Lancet Oncol 2017
ConclusionsCheckMate 142: nivolumab alone
• Nivolumab monotherapy provided durable responses and disease control in patients with dMMR/MSI-H CRC
• Responses and disease control were observed regardless of tumor PD-L1 expression, BRAF or KRAS mutation status, or a clinical history of Lynch syndrome
• Patient-reported outcome analyses showed clinically meaningful improvements in functioning, symptoms, and QOL
– Patients who continued treatment for ≥ 19 weeks attained a level of health that would be considered as equal to or exceeding the general health of many populations
• Nivolumab was well tolerated with no new safety signals observed
• These results suggest that nivolumab could be considered a new standard of care for patients with dMMR/MSI-H CRC and warrants further investigation in patients with other dMMR/MSI-H tumors
Overman M, Lancet Oncol 2017
FDA grants nivolumab accelerated approval for MSI-H or dMMR colorectal cancer
On July 31, 2017, the U.S. Food and Drug Administration granted accelerated approval to nivolumab (OPDIVO, Bristol-Myers Squibb Company) for the treatment of patients 12 years and older with mismatch repair deficient (dMMR) and microsatellite instability high (MSI-H) metastatic colorectal cancer that has progressed following treatment with a fluoropyrimidine, oxaliplatin, and
irinotecan. The approval was based on data from Study CA209142 (CHECKMATE 142; NCT 02060188)
Combination of nivolumab (NIVO) + ipilimumab (IPI) in the treatment of
patients (pts) with deficient DNA mismatch repair (dMMR)/high
microsatellite instability (MSI-H) metastatic colorectal cancer (mCRC):
CheckMate 142 Study
André T et al; ASCO 2017; abst 3531
Study Design: CheckMate 142 Study
Primary endpoint: ORR per investigator assessment (RECIST v1.1)
Current analysis included all patients (n=84) who received their first dose ≥ 6 months prior to the data cut-off
Median (range) time from first dose to data cut-off: 8.6 (6.3 – 19.4) months
Patients
NIVO 3 mg/kg +
IPI 1 mg/kg Q3W
(4 doses and then NIVO 3 mg/kg Q2W)
NIVO 3 mg/kg +
IPI 1 mg/kg Q3W
(4 doses and then NIVO 3 mg/kg Q2W)
If ≥ 7/19 confirmed
responders,
continue enrollmentNIVO 3 mg/kg Q2W NIVO 3 mg/kg Q2W
If ≥ 7/19 confirmed
responders, continue
enrollment
• Histologically
confirmed
metastatic/
recurrent CRC
• dMMR/MSI-H per
local laboratory
• ≥ 1 prior line of
therapy
Stage 2Stage 1
MONOTHERAPY
ARM
COMBINATIONARM
n = 74 pts
n = 84 pts
André T et al; ASCO 2017, Abst 3531
dMMR/MSI-H
(n = 84)
Age
Median (range), years
< 65 years, n (%)
57 (21-81)
61 (73)
ECOG performance status, n (%)
0
1
31 (37)
53 (63)
Mutation status, n (%)
BRAF/KRAS wild type
BRAF mutated
KRAS mutated
22 (26)
21 (25)
30 (36)
Prior lines of therapy, n (%)
0
1
2
3
≥ 4
1 (1)
17 (20)
31 (37)
23 (27)
12(14)
Prior radiotherapy, n (%) 17 (20)
Patient Demographics and Disease Characteristics
CheckMate 142 Study: nivolumab + ipilimumab
André T et al; ASCO 2017, Abst 3531
NIVO + IPI
(n = 84)
Investigator-
Assessed
ORR, n (%)
[95% CI]
46 (55)
[43.5, 65.7]
Best overall response, n (%)
CR
PR
SD
PD
ND/reported
2 (2)
44 (52)
26 (31)
9 (11)
3 (4)
Disease control for ≥ 12 weeks, n (%)a 66 (79)
Overall Response and Disease Control:
CheckMate 142 Study: nivolumab + ipilimumab
NIVO Monotherapy1
(n = 74)
Investigator-Assessed
23 (31)
[20.8, 42.9]
0
23 (31)
29 (39)
18 (24)
4 (5)
51 (69)
aPatients with CR, PR, or SD for ≥ 12 weeks André T et al; ASCO 2017, Abst 3531
Overman M, Lancet Oncol 2017
Best Change in Target Lesion Size
NIVO + IPI (n = 84) NIVO Monotherapy1 (n = 74)
100
75
50
25
0
-25
-50
-75
-100
100
Best
Red
ucti
on
Fro
m B
aselin
ein
Targ
et
Lesio
n (
%) 75
50
25
0
-25
-50
-75
-100
Best
Red
ucti
on
Fro
m B
aselin
ein
Targ
et
Lesio
n (
%)
80% of patients had reduction in tumor burden from baseline 62% of patients had reduction in tumor burden from baseline
Confirmed CR or PR per investigator % Change truncated at 100 André T et al; ASCO 2017, Abst 3531
Overman M, Lancet Oncol 2017
34
100
80
60
40
20
0
100
80
60
40
20
B
Be
st
red
uc
tio
n f
rom
ba
se
lin
e
in t
arg
et
les
ion
(%
)
KRAS/BRAF wild type KRAS mutationBRAF mutation Unknown
Results
Figure 2. Association of best reduction in target lesion size (B) BRAF and KRAS mutation status in patients with dMMR/MSI-H mCRC treated with NIVO + IPI
35
C
100
80
60
40
20
0
100
80
60
40
20
Be
st
red
uc
tio
n f
rom
ba
se
lin
e
in t
arg
et
les
ion
(%
)
Lynch syndrome (Yes) UnknownLynch syndrome (No)
Figure 2. Association of best reduction in target lesion size with(C) Lynch syndrome in patients with dMMR/MSI-H mCRC treated with NIVO + IPI
Results
Changes in Tumor Burden
CheckMate 142 Study: nivolumab + ipilimumab
• Median TTR: 2.8 (1.1–14.0) months• Median DOR: not reached• 85% (39/46) of responses ongoing
Ch
ange
in t
arge
t le
sio
n fr
om
bas
elin
e, (%
)
Time since start of treatment (week)
0 7872666054484236302418126 84 90
100
75
50
25
-25
-50
-75
-100
0
1st occurrence of new lesion
CR or PR
On treatment
André T et al; ASCO 2017, Abst 3531
Overman M, Lancet Oncol 2017
No. at Risk
1.0
0.9
0.8
0.5
0.4
0.3
0.2
0.0
Pro
ba
bilit
y o
f P
rog
res
sio
n-F
ree
Su
rviv
al
0.6
0.1
0 3 6 9
Time (months)
12 15 18 21
65 35 17 13 8 1 084
Progression-Free SurvivalCheckMate 142 Study: nivolumab + ipilimumab
Median PFS, mo (95% CI) NR (11.5, NE)
PFS rate, % (95% CI)
6 months 77 (66.5, 85.1)
9 months 77 (66.5, 85.1)
NE = not estimable; NR = not reached André T et al; ASCO 2017, Abst 3531
Median OS, mo (95% CI) NR (NE, NE)
OS rate, % (95% CI)
6 months 89 (80.2, 94.2)
9 months 88 (78.1, 93.1)
Overall SurvivalCheckMate 142 Study: nivolumab + ipilimumab
NE = not estimable; NR = not reached
Pro
ba
bilit
y o
f S
urv
iva
l
0 3 6 9 12 15 18 21
84 77 73 40 22 19 13 0
1.0
0.9
0.8
0.5
0.4
0.3
0.2
0.0
0.6
0.1
0.7
Time (months)
No. at Risk
André T et al; ASCO 2017, Abst 3531
Checkmate 142 Treatment-Related Adverse Events in ≥ 15% of Patients With MSI-H
Event, n (%)
Nivolumab3 mg/kg(n = 70)
Nivolumab 3 mg/kg + Ipilimumab 1 mg/kg
(n = 30)
Any grade Grade 3–4 Any grade Grade 3–4
Any event 41 (58.6)a 10 (14.3) 25 (83.3) 8 (26.7)
Fatigue 13 (18.6) 1 (1.4) 6 (20.0) 0
Diarrhea 10 (14.3) 1 (1.4) 13 (43.3) 0
Pruritus 8 (11.4) 0 5 (16.7) 1 (3.3)
Nausea 5 (7.1) 0 6 (20.0) 0
Pyrexia 3 (4.3) 0 7 (23.3) 0
Any event leading to discontinuation
4 (5.7) 2 (2.9) 4 (13.3) 4 (13.3)
39
aOne Grade 5 event of sudden death
J Taieb ESDO Leuven 2017
NIVO + IPI(n = 84)
Patients, n (%) Any Grade Grade 3 or 4Any TRAE 57 (68) 24 (29)Serious TRAEs 15 (18) 14 (17)
Discontinuation due to TRAEs 11 (13) 8 (9)
TRAEs reported in ≥ 10% of patientsDiarrhea 20 (24) 1 (1)Fatigue 14 (17) 1 (1)Aspartate aminotransferase increase 14 (17) 8 (9)Pyrexia 13 (16) 0Pruritus 13 (16) 2 (2)Alanine aminotransferase increase 12 (14) 7 (8)Nausea 12 (14) 0Hyperthyroidism 11 (13) 0Hypothyroidism 11 (13) 0
Safety Check Mate 142 Study: nivolumab + ipilimumab
• No treatment-related deaths were reported
André T et al; ASCO 2017, Abst 3531
NIVO + IPI provided durable responses, sustained disease control,
and encouraging survival data in pretreated patients with dMMR/MSI-H mCRC
–ORR of 55%, with 79% of patients achieving disease control for ≥ 12 weeks
–85% of responses ongoing, and the median DOR was not yet reached
–88% of patients alive at 9 months, 77% without progression
NIVO + IPI demonstrated a manageable safety profile; 29% of patients had grade 3/4 TRAEs
This trial is ongoing and patients will continue to be followed for survival
41 patients in this study were treated in St Antoine hospital and translational research are planned (we have primary and metastatic tissue for
all)
A French study will began in Septempter 2017 (GERCOR group) with nivolumab and ipilumumab (55 patients) in the same population and we
will collect archival tissue for all patients
ConclusionsCheck Mate 142 Study: nivolumab + ipilimumab
dMMR/
MSI-H
COL-11
Reproduced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Colon Cancer V.2.2017. © 2017 National Comprehensive Cancer Network, Inc. All rights reserved. The NCCN Guidelines® and illustrations herein may not
be reproduced in any form for any purpose without the express written permission of NCCN. To view the most recent and complete version of the NCCN Guidelines, go online to NCCN.org. The NCCN Guidelines are a work in progress that may be refined as often
as new significant data becomes available. NCCN makes no warranties of any kind whatsoever regarding their content, use or application and disclaims any responsibility for their application or use in any way.
Summary of Efficacy in Patients With MSSNivolumab ± Ipilimumab in Metastatic CRC
43
Nivolumab 1 mg/kg + Ipilimumab 3 mg/kg
(n = 10)
Nivolumab 3 mg/kg + Ipilimumab 1 mg/kg
(n = 10)
ORR, n (%) 1 (10) 0
Median PFS, mo (95% CI) 2.28 (0.62, 4.40) 1.31 (0.89, 1.71)
Median OS, mo (95% CI) 11.53 (0.62, NE) 3.73 (1.22, 5.62)
NE = not estimable
…vs NR OS and 14 PFS in MSI-H pts treated with nivolumab alone…vs median not reached in MSI-H pts treated with the combo
Phase Ib: Atezolizumab + cobimetinibin MSS pretreated mCRC
Bendell JVC. et al., ASCO 2016, OS 3502
• Tolerability :
• No DLT, Diarrhea : 70%, rash: 40%, Fatigue : 52%• Grade 3-4 toxicities : 34% incl. diarrhea : 9%
• Efficacy :
• Response 17% (4 OR, 5 SD) : 3 MSS/1 statut MSI ukn, 4 to 7 mo
-100-90-80-70-60-50-40-30-20-10
0102030405060708090
100
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23Ch
ange
in s
um
of
lon
gest
dia
met
ers
fro
mb
asel
ine,
%
Time on study (mo)
PDSDPR/CRDiscontinued atazolizumabNew lesion
Tumor Biopsy : no correlation with PDL1 at D0
Conclusion• Trials are ongoing in mCRC and in the adjuvant setting (soon)
• Colon cancer probably less easy than others
• MSI-H tumors : a good target
• Others may be: PolE, PolD, MSS with immune infiltrates…
• Combination with targeted agents: • + chemotherapy, sequence?• + radiotherapy? • + targeted agents: anti-angiogenics; MEK…?
J Taieb ESDO Leuven 2017
Anti-CTLA-4 or anti-PD-1 or anti-PD-L1
Perioperative First-line Second-lineThird-line +
refractory to standard
Ipilimumab (BMS): anti-CTLA-4
Nivolumab (ONO/BMS):anti-PD-1
Pembrolizumab (MSD):anti-PD-1
KEYNOTE-177 Ph IIIPembro vs CT
MSI-H
Durvalumab (AZ): anti-PD-L1
Atezolizumab (Roche): anti-PD-L1
Phase III Adjuvant NCIFOLFOX 6 m+/- atezo 1y
MSI-H
Ph IIIFOLFOX+BVs Atezo
Vs Atezo+FOLFOX+BMSI-H
Phase IIIAtezo+ cobimetinib vs Atezo vs
regorafenibMSS
Avelumab (Merck/Pfizer)anti-PD-L1
SAMCOR Phase II
Avelumab vs CTMSI-H
After ASCO: even more trials for colorectal cancer
1. Study designs available at: www.clinicaltrials.gov (accessed September 2017). Agents have not yet received EMA approval for treatment of indication listed
Gastric cancer
Cu
mu
lati
ve s
urv
ival
0.0
0.2
0.4
0.6
1.0
0.8
PD-L1 Positive(n=67)
PD-L1 Negative(n=65)
48 72 96 1680 24 144120
1. Kim JW et al. Gastric Cancer 2016;19:42–2; 2. Böger C et al. Oncotarget 2016;7:24269–83; 3. Saito H et al. J Surg Oncol 2013;107:517–22; 4. Takano S et al. Surg Today 2016;46:1341–7; 5. Zhang L et al. Int J Clin Exp Pathol. 2015;8:11084–91.
Rationale for immuno-oncology in GC• Pathology of GC associated
with immune system evasion
• Immunosuppressive proteins expressed on immune cells in patients with GC patients, including key checkpoint inhibitors:1–4
• Cytotoxic T-lymphocyte-associated protein-4 (CTLA-4)
• PD-1
Overall survival
PD-L1 in gastric carcinoma5
Study Design and Endpoints
BOR, best overall response; DCR, disease control rate; DOR, duration of response; ECOG PS, Eastern Cooperative Oncology Group performance status; IV; intravenous; ORR, objective response rate; OS, overall survival; PFS, progression-free survival; Q2W, every 2 weeks; R, randomization; RECIST, Response Evaluation Criteria In Solid Tumors; TTR, time to tumor response.
R2:1
Nivolumab3 mg/kg IV Q2W
Placebo
Key eligibility criteria:
• Age ≥ 20 years
• Unresectable advanced or recurrent gastric or gastroesophageal junction cancer
• Histologically confirmed adenocarcinoma
• Prior treatment with ≥ 2 regimens and refractory to/intolerant of standard therapy
• ECOG PS of 0 or 1
Primary endpoint:• OS
Secondary endpoints:• Efficacy (PFS, BOR, ORR, TTR,
DOR, DCR)• Safety
Exploratory endpoint:• Biomarkers
Stratification based on:
• Country (Japan vs Korea vs Taiwan)
• ECOG PS (0 vs 1)
• Number of organs with metastases (< 2 vs ≥ 2)
• Patients were permitted to continue treatment beyond initial RECIST v1.1–defined disease progression, as assessed by the investigator, if receiving clinical benefit and tolerating study drug
49ASCO GI 2017Kang et al. ASCO GI 2017
Overall Survival
Time (months)
Pro
bab
ility
of
Surv
ival
(%)
22181614121086420
0
10
20
30
40
50
60
70
80
90
100
Hazard ratio, 0.63 (95% CI, 0.50–0.78)
P < 0.0001
0351019395795142275330
0133410163253121163
Nivolumab
Placebo
At risk:
20
193
82
Patients, nEvents,
nMedian OS
[95% CI], months12-Month OS Rate [95%
CI], %
Nivolumab 330 225 5.32 [4.63–6.41] 26.6 [21.1–32.4]
Placebo 163 141 4.14 [3.42–4.86] 10.9 [6.2–17.0]
50ASCO GI 2017Kang et al. ASCO GI 2017
Overall Survival by Subgroup
Subgroup Hazard Ratio [95% CI]
Histological type (Lauren classification)
Intestinal type
Diffuse type
Mixed
Unknown
0.59 [0.41–0.85]
0.82 [0.57–1.17]
0.37 [0.13–1.04]
0.56 [0.37–0.84]
Number of organs with metastasis
< 2
≥ 2
0.70 [0.46–1.08]
0.61 [0.48–0.78]
Peritoneal metastasis
No
Yes
0.63 [0.50–0.81]
0.74 [0.48–1.15]
Liver metastasis
No
Yes
0.63 [0.50–0.80]
0.67 [0.42–1.07]
Measurable lesion
No
Yes
0.70 [0.43–1.14]
0.63 [0.50–0.80]
Number of previous regimens
2
3
≥ 4
0.82 [0.50–1.35]
0.87 [0.61–1.22]
0.44 [0.31–0.61]
Subgroup Hazard Ratio [95% CI]
All 0.64 [0.52–0.80]
Country
Japan
Korea
Taiwan
0.63 [0.46–0.85]
0.70 [0.51–0.96]
0.46 [0.23–0.92]
Age, years
< 65
≥ 65
0.75 [0.57–0.98]
0.53 [0.38–0.74]
Sex
Male
Female
0.58 [0.45–0.75]
0.83 [0.56–1.23]
ECOG PS
0
1
0.59 [0.40–0.87]
0.67 [0.52–0.86]
Prior gastrectomy
No
Yes
0.69 [0.49–0.98]
0.60 [0.46–0.79]
Primary sites
Gastric (fundus, corpus, antrum, and pylorus)
Gastroesophageal junction
Unknown
0.69 [0.55–0.87]
0.44 [0.20–0.97]
0.52 [0.26–1.06]
0 1 2 3
Favors nivolumab Favors placebo
Hazard ratio [95% CI]
0 1 2 3
Favors nivolumab Favors placebo
Hazard ratio [95% CI]
51ASCO GI 2017Kang et al. ASCO GI 2017
Maximum Reduction in Tumor Burden From BaselineNivolumab Placebo
Max
imu
m R
edu
ctio
n F
rom
Bas
elin
e in
Tar
get
Lesi
on
s (%
)
-100
-80
-60
-40
-20
0
20
40
60
80
100
-100
-80
-60
-40
-20
0
20
40
60
80
100a
a Patients with a change in tumor burden that exceeds 100%.
a
Patients with Tumor reduction: 37.3% Patients with Tumor reduction: 12.4%
52ASCO GI 2017Kang et al. ASCO GI 2017
31% ORR
0
10
20
30
40
50
60
70
80
90
100
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28
Pro
bab
ilit
y o
f S
urv
ival
(%
)
0
10
20
30
40
50
60
70
80
90
100
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28
ONO-4538
Placebo
PD-L1 <1% PD-L1 ≥1%
Hazard ratio, 0.58 (95% CI,
0.24–1.38)
Median OS, months (95% CI)
Nivolumab (n=16) 5.2 (2.8–9.4)
Placebo (n=10) 3.8 (0.8–5.0)
Pro
bab
ilit
y o
f S
urv
iva
l (%
)
114 100 75 56 49 42 37 24 15 11 7 4 3 1 0
52 40 27 22 16 14 11 6 5 4 3 2 2 2 0
16 15 10 7 5 4 4 2 2 0 0 0 0 0 0
10 8 4 2 1 1 1 0 0 0 0 0 0 0 0
Nivolumab
Placebo
Months MonthsNo. at Risk
Median OS, months (95% CI)
Nivolumab (n=114) 6.1 (4.8–8.6)
Placebo (n=52) 4.2 (3.0–6.9)
Hazard ratio, 0.71 (95% CI,
0.50–1.01)
No strong signal for PDL1 predictive value
Overall survival by PD-L1 expression <1% vs ≥1%
PD-L1 evaluable patients (N=192)
53
Methods
54
aIV oxaliplatin 130 mg/m2 on day 1 followed by 20 days off and oral S-1 or oral capecitabine twice daily for 14 days followed by 7 days off. S-1 initial dose was 40 mg/m2/dose. Capecitabine initial dose was 1,000 mg/m2/dose
ECOG PS = Eastern Cooperative Oncology Group performance status; IV = intravenous; Q3W = every 3 weeks; RECIST = Response Evaluation Criteria in Solid Tumors
Figure 1. ATTRACTION-04 study design (part 1)
Results
55
Table 2. Summary of treatment-related adverse events
56
Figure 3. Best reduction in tumor burden
Results
Authors’ Conclusions
57
• Part 1 of this two-part randomized study met its primary and secondary endpoints; nivolumab + chemotherapy regimens (SOX or CapeOX)
had manageable safety profiles and demonstrated clinically meaningful antitumor activity
– Discontinuations due to treatment-related AEs were low (≤10% of patients) and no treatment-related deaths were reported
– The incidence and type of treatment-related AEs were consistent with AEs known to be associated with chemotherapy regimens and
with nivolumab5–9
– Objective responses were observed in more than two-thirds of patients, occurred early, and were durable, with encouraging PFS
• No substantial differences in clinical activity and safety were observed between the nivolumab + SOX and nivolumab + CapeOX treatment
regimens
• These results suggest that nivolumab in combination with SOX or CapeOX chemotherapy in patients with unresectable advanced or
recurrent gastric or gastroesophageal junction cancer may be a first-line therapy option
– Part 2 of this study is currently recruiting patients
4014: Nivolumab ± ipilimumab in pts with advanced (adv)/metastatic chemotherapy-refractory (CTx-R) gastric (G), esophageal (E), or gastroesophageal junction (GEJ) cancer: CheckMate 032 study
*Administered for 4 cycles followed by nivolumab 3 mg/kg IV q2wJanjigian YY, et al. J Clin Oncol 2017;35(Suppl):Abstr 4014
Study objective
• To evaluate the long-term survival, efficacy, and safety of nivolumab alone and in combination with ipilimumab in the oesophagogastric cohort of the CheckMate 032 study
PD/death/
toxicity
Nivolumab 3 mg/kg IV q2w
(n=59)Key patient inclusion criteria
• Advanced/metastatic oesophagogastric cancer
• Progression on ≥1 prior chemotherapy
• Western population
(n=160)
PRIMARY ENDPOINT
• ORR (RECIST v1.1)SECONDARY ENDPOINTS
• OS, PFS, TTR, DoR, safety, PD-L1 tumour expression
PD/death/
toxicity
Nivolumab 1 mg/kg + ipilimumab 3 mg/kg
IV q3w*
(n=49)
Nivolumab 3 mg/kg + ipilimumab 1 mg/kg
IV q3w*
(n=52)
PD/death/
toxicity
4014: Nivolumab ± ipilimumab in pts with advanced (adv)/metastatic chemotherapy-refractory (CTx-R) gastric (G), esophageal(E), or gastroesophageal junction (GEJ) cancer: CheckMate 032 study – Janjigian YY, et al
Janjigian YY, et al. J Clin Oncol 2017;35(Suppl):Abstr 4014
1.0
0.8
0.6
0.4
0.2
0
Pro
ba
bili
ty o
f su
rviv
al
0 3 6 9 12 15 18 21 24 27 30 33
Time, months
1.0
0.8
0.6
0.4
0.2
0Pro
ba
bili
ty o
f p
rogre
ssio
n-f
ree s
urv
iva
l
0 3 6 9 12 15 18 21 24 27 30
Time, months
mPFS, months
(95%CI)
PFS rate, %
6-month 12-month
Nivolumab 3 mg/kg 1.4 (1.2, 1.5) 17 8
Nivolumab 1 mg/kg +
Ipilimumab 3 mg/kg1.4 (1.2, 3.8) 24 17
Nivolumab 3 mg/kg +
Ipilimumab 1 mg/kg1.6 (1.4, 2.6) 12 10
mOS, months
(95%CI)
OS rate, %
12-month 18-month
Nivolumab 3 mg/kg 6.2 (3.4, 12.4) 39 25
Nivolumab 1 mg/kg +
Ipilimumab 3 mg/kg6.9 (3.7, 11.5) 35 28
Nivolumab 3 mg/kg +
Ipilimumab 1 mg/kg4.8 (3.0, 8.4) 24 13
OS PFS
4014: Nivolumab ± ipilimumab in pts with advanced (adv)/metastatic chemotherapy-refractory (CTx-R) gastric (G), esophageal(E), or gastroesophageal junction (GEJ) cancer: CheckMate 032 study – Janjigian YY, et al
Janjigian YY, et al. J Clin Oncol 2017;35(Suppl):Abstr 4014
Key results (cont.)
Best reduction in target lesions
Nivolumab 3 mg/kg
PD-L1-evaluable patients, 38 of 53
Nivolumab 3 mg/kg +
Ipilimumab 1 mg/kg
PD-L1-evaluable patients, 34 of 41
Nivolumab 1 mg/kg +
Ipilimumab 3 mg/kg
PD-L1-evaluable patients, 38 of 42
100
75
50
25
0
-25
-50
Be
st re
du
ctio
n fro
m b
ase
line
in
ta
rge
t le
sio
ns, %
100
75
50
25
0
-25
-50
100
75
50
25
0
-25
-50
PD-L1 <1% PD-L1 ≥1% PD-L1 not evaluable/missing
† †
*
*
* *
*
*
*
†
**
** * *
* * *
†
*
*
*
KEYNOTE-059: Efficacy and Safety of Pembrolizumab Alone or in Combination With Chemotherapy in Patients With Advanced Gastric or Gastroesophageal Cancer
Study Design
Treat for up to 35 cycles(~2 years), or until
progression or intolerable toxicity
Follow-up for survival bytelephone until death,
withdrawal, or study end
Cohort 3No prior therapy
PD-L1 positive
Cohort 2No prior therapy
PD-L1 positive or negative
Cohort 1≥2 prior lines of chemotherapy PD-L1
positive or negative
Pembrolizumab 200 mg Q3W +Cisplatin 80mg/m2 Q3W +5-FU 800 mg/m2 Q3W or
Capecitabine 1000 mg/m2 BID Q3Wa
Pembrolizumab200 mg Q3W
Pembrolizumab200 mg Q3W
Response assessment per RECIST v1.1: First scan 9 weeks after cycle 1, then every 6 weeks for year 1 and every 9 weeks thereafterPrimary end points: Safety (all cohorts); ORR by central review per RECIST v1.1 (cohort 1: all patients and patients with PD-L1-positive expression); ORR by central review per RECIST v1.1 (cohort 3)PD-L1 positive: combined positive score (CPS) ≥1 (previously reported as and equivalent to CPS ≥1%), where CPS=number of PD-L1-positive cellsb (tumor cells, lymphocytes, and macrophages) divided by the total number of tumor cells x 100
aCapecitabine was administered only in JapanbPD-L1 IHC 22C3 pharmaDx (Agilent Technologies, Carpinteria, CA,USA)Wainberg et al., Abstract LBA28, ESMO 2017
KEYNOTE-059 cohort 1: Efficacy and safety of pembrolizumab monotherapy in patients with previously treated advanced gastric cancer.
Study Design
aCapecitabine was administered only in Japan
Treat for up to 35 cycles(~2 years), or until
progression or intolerable toxicity
Follow-up for survival bytelephone until death,
withdrawal, or study end
Cohort 3No prior therapy
PD-L1 positive
Cohort 2No prior therapy
PD-L1 positive or negative
Cohort 1≥2 prior lines of chemotherapy PD-L1
positive or negative
Pembrolizumab 200 mg Q3W +Cisplatin 80mg/m2 Q3W +5-FU 800 mg/m2 Q3W or
Capecitabine 1000 mg/m2 BID Q3Wa
Pembrolizumab200 mg Q3W
Pembrolizumab200 mg Q3W
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017
KEYNOTE-059 cohort 1: Efficacy and safety of pembrolizumab monotherapy in patients with previously treated advanced gastric cancer.
Characteristic N=259
Median age (range), years 62 (24-89)
Male, n (%) 198 (76)
Race, n (%)WhiteAsianOther
200 (77)41 (16)
7 (3)
ECOG, n (%)01
107 (41)151 (58)
Number of prior therapies, n (%)23≥4
134 (52)75 (29)50 (19)
Characteristic N=259
Location of primary tumor, n (%)GEJGastric
134 (52)124 (48)
Prior gastrectomy, n (%)TotalPartialNo
49 (19)17 (7)
193 (75)
PD-L1 expression , n (%)PositiveNegativeUnknown
148 (57)109 (42)
2 (1)
HER2 status, n (%)PositiveNegativeIntermediate
63 (24)194 (75)
2 (1)
Baseline Characteristics
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017
KEYNOTE-059 cohort 1: Efficacy and safety of pembrolizumab monotherapy in patients with previously treated advanced gastric cancer.
Response
Responseb
All PatientsN = 259
PD-L1 Positivea
N = 148PD-L1 Negative
N = 109
% 95% CI % 95% CI % 95% CI
ORR 12 8-17 16 11-23 6 3-13
DCRc 27 22-33 34 26-42 19 12-28
DOR
CR 3 1-6 3 1-8 3 1-8
PR 9 6-13 13 8-19 4 1-9
SD 16 12-21 18 12-25 15 9-23
PD 56 49-62 53 44-61 60 50-69
aCPS ≥ 1bOnly confirmed responses were includedcCR + PR + SD≥2 months
• Median (range) follow-up: 5.6 (0.5-24.7) months• 134 patients received pembrolizumab as third-line therapy; ORR was 16%, and DCR was 31%• 125 patients received pembrolizuman as fourth plus-line therapy; ORR was 7%, and DCR was 23%
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017
KEYNOTE-059 cohort 1: Efficacy and safety of pembrolizumab monotherapy in patients with previously treated advanced gastric cancer.
aOnly patients with measurable disease per RECIST v1.1 by central review at baseline who had ≥1 postbaseline assessment wereincluded (n=223); assessment was nonevaluable in 1 patientbLongitudinal change in the sum of the longest target lesion diameters from baseline in responders (n=31)+No progressive disease at last disease assessment
Best Percentage Change and Longitudinal Change in Target Lesion Size
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017
KEYNOTE-059 cohort 1: Efficacy and safety of pembrolizumab monotherapy in patients with previously treated advanced gastric cancer.
PFS and OS in All Patients
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017
KEYNOTE-059 cohort 1: Efficacy and safety of pembrolizumab monotherapy in patients with previously treated advanced gastric cancer.
PFS and OS by PD-L1 Expression
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017
KEYNOTE-059 cohort 1: Efficacy and safety of pembrolizumab monotherapy in patients with previously treated advanced gastric cancer.
Treatment-Related Adverse Events
Event, n (%) N = 259
Any 159 (61)
Grades 3-5Anemia, grade 3Fatigue, grade 3Dehydration, grade 3
46 (18)7 (3)6 (2)3 (1)
Serious 29 (11)
Led to discontinuationa 7 (3)
Led to deathAcute kidney injuryPleural effusion
2 (1)1 (1)1 (1)
• Median (range) duration of exposure was 2.1 (0.0-23,7) months
aAbnormal hepatic function, bile duct stenosis, encephalitis, increase blood bilirubin level, hyperglycemia, acutekidney injury, and pneumonitis
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017
KEYNOTE-059 cohort 1: Efficacy and safety of pembrolizumab monotherapy in patients with previously treated advanced gastric cancer.
Immune-Mediated Adverse Eventsa and Infusion-Related Reactions
• There were no grade 4/5 immune-mediated or infusion reactions
aBased on a list of terms specified by the sponsor and included regardless of attribution to studytreatment or immune relatedness by the investigatorb2 (1%) patients experienced grade 3 rash; 1 (<1%) patient experienced grade 3 adverse events: uveitis, hepatitis, jaundice, encephalitis, and maculopapular rash
Event, n (%)N=259
All Grades in >2 Patients Grades 3b
Any 50 (19) 13 (5)
Hypothyroidism 24 (9) 1 (<1)
Hyperthyroidism 9 (4) 0
Colitis 4 (2) 3 (1)
Infusion-related reactions 4 (2) 0
Pneumonitis 4 (2) 2 (1)
Thyroiditis 3 (1) 1 (<1)
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017
aCapecitabine was administered only in Japan
KEYNOTE-059 Cohort 2: Safety and efficacy of pembrolizumab plus5-fluorouracil and cisplatin for first-line treatment of advanced gastric cancer
Study Design
Treat for up to 35 cycles(~2 years), or until
progression or intolerable toxicity
Follow-up for survival bytelephone until death,
withdrawal, or study end
Cohort 3No prior therapy
PD-L1 positive
Cohort 2No prior therapy
PD-L1 positive or negative
Cohort 1≥2 prior lines of chemotherapy PD-L1
positive or negative
Pembrolizumab 200 mg Q3W +Cisplatin 80mg/m2 Q3W +5-FU 800 mg/m2 Q3W or
Capecitabine 1000 mg/m2 BID Q3Wa
Pembrolizumab200 mg Q3W
Pembrolizumab200 mg Q3W
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017
Characteristic N=25
Median age (range), years, y 64 (21-82)
Male, n (%) 16 (24)
Race, n (%)WhiteAsian
8 (32)17 (68)
ECOG PS, n (%)01
15 (60)10 (40)
Characteristic N=25
Location of primary tumor, n (%)GEJGastric
5 (20)20 (80)
Prior gastrectomy, n (%)TotalPartialNo
4 (16)1 (4)
20 (80)
PD-L1 expression , n (%)PositiveNegativeUnknown
16 (64)8 (32)1 (4)
HER2 negative, n (%) 25 (100)
Baseline Characteristics
KEYNOTE-059 Cohort 2: Safety and efficacy of pembrolizumab plus5-fluorouracil and cisplatin for first-line treatment of advanced gastric cancer
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017
Response
Responseb
All PatientsN = 25
PD-L1 Positivea
N = 16PD-L1 Negative
N = 8
% 95% CI % 95% CI % 95% CI
ORR 60 39-79 69 41-89 38 9-76
DCRc 80 59-93 75 48-93 75 35-97
BOR
CR 4 0-20 0 0-22 13 0-53
PR 56 35-76 69 41-89 25 3-65
SD 32 15-54 19 4-46 50 16-84
PD 4 0-20 6 0-30 0 0-37
aCPS ≥1 bOnly confirmed responses were includedcCR + PR + SD≥6 months
Median (range) follow-up in cohort 2: 13.8 (1.8-24.1) months
KEYNOTE-059 Cohort 2: Safety and efficacy of pembrolizumab plus5-fluorouracil and cisplatin for first-line treatment of advanced gastric cancer
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017
Best Percentage Change and Longitudinal Change in Target Lesion Size
aOnly patients with measurable disease per RECIST v1.1 by central review at baseline who had ≥1 postbaseline assessment wereincluded (n=25); assessment was nonevaluable for 1 patientbLongitudinal change in the sum of the longest target lesion diameters from baseline in patients with ≥1 postbaseline assessment(n=25)+No progressive disease at last disease assessment
KEYNOTE-059 Cohort 2: Safety and efficacy of pembrolizumab plus5-fluorouracil and cisplatin for first-line treatment of advanced gastric cancer
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017
KEYNOTE-059 Cohort 2: Safety and efficacy of pembrolizumab plus5-fluorouracil and cisplatin for first-line treatment of advanced gastric cancer
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017
PFS and OS
Treatment-Related Adverse Events
Event, n (%) N = 25
Any 25 (100)
Grades 3/4NeutropeniaStomatitisAnemiaDecreased platelet countDecreased appetiteFatigue
19 (76)6 (24)5 (20)2 (8)2 (8)2 (8)2 (8)
Led to discontinuation 3 (12)
Led to death 0
• Median (range) duration of exposure was 7.1 (0.8-23.8) months• 3 (12%) patients discontinued treatment because of chemotherapy-related
adverse events (stomatitis [grade 3], hypoacusis [grade 2], increasedcreatinine level [grade 1])
• No patients discontinued treatment because of pembrolizumab-relatedadverse events
KEYNOTE-059 Cohort 2: Safety and efficacy of pembrolizumab plus5-fluorouracil and cisplatin for first-line treatment of advanced gastric cancer
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017
Immune-Mediated Adverse Eventsa
• There were no grade 4/5 immune-mediated or infusion reactions
aBased on a list of terms specified by the sponsor and included regardless of attribution to study treatment orimmune relatedness by the investigator
Event, n (%)
N=25
All Grades in >2 Patients
Grades 3 in All Patients
Any 12 (48) 4 (16)
Hyperthyroidism 4 (16) 0
Palmar-plantar erythrodysesthesia
2 (8) 2 (8)
Nephrotic syndrome 1 (4) 1 (4)
Rash 1 (4) 1 (4)
Maculopapular rash 1 (4) 1 (4)
KEYNOTE-059 Cohort 2: Safety and efficacy of pembrolizumab plus5-fluorouracil and cisplatin for first-line treatment of advanced gastric cancer
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017
KEYNOTE-059 Cohort 3: Safety and efficacy of pembrolizumab monotherapy for first-line treatment of patientswith PD-L1-positive advanced gastric/gastroesophageal cancer
Study Design
Treat for up to 35 cycles(~2 years), or until
progression or intolerable toxicity
Follow-up for survival bytelephone until death,
withdrawal, or study end
Cohort 3No prior therapy
PD-L1 positive
Cohort 2No prior therapy
PD-L1 positive or negative
Cohort 1≥2 prior lines of chemotherapy PD-L1
positive or negative
Pembrolizumab 200 mg Q3W +Cisplatin 80mg/m2 Q3W +5-FU 800 mg/m2 Q3W or
Capecitabine 1000 mg/m2 BID Q3Wa
Pembrolizumab200 mg Q3W
Pembrolizumab200 mg Q3W
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017 aCapecitabine was administered only in Japan
Characteristic N=31
Median age (range), years, y 62 (32-75)
Male, n (%) 19 (61)
Race, n (%)WhiteAsian
16 (52)15 (48)
ECOG PS, n (%)01
14 (45)17 (55)
No. of prior therapies, n (%) 0
Characteristic N=31
Location of primary tumor, n (%)GEJGastric
12 (39)19 (61)
Prior gastrectomy, n (%)TotalPartialNo
9 (29)3 (10)
19 (61)
PD-L1 positive, n (%) 31 (100)
HER2 negative, n (%) 31 (100)
Baseline Characteristics
KEYNOTE-059 Cohort 3: Safety and efficacy of pembrolizumab monotherapy for first-line treatment of patientswith PD-L1-positive advanced gastric/gastroesophageal cancer
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017
Responsea
N = 31
% 95% CI
ORR 26 12-45
DCRb 36 19-55
BOR
CR 7 1-21
PR 19 8-38
SD 29 14-48
PD 39 22-58
aOnly confirmed responses were includedbCR + PR + SD≥6 monthsData cutoff: April 21, 2017
Median (range) follow-up in cohort 3: 17.5 (1.7-20.7) months
Response
KEYNOTE-059 Cohort 3: Safety and efficacy of pembrolizumab monotherapy for first-line treatment of patientswith PD-L1-positive advanced gastric/gastroesophageal cancer
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017
aOnly patients with measurable disease per RECIST v1.1 by central review at baseline who had ≥1 postbaseline assessment wereincluded (n=31); assessment were nonevaluable/not evailable in 3 patientsbLongitudinal change in the sum of the longest target lesion diameters from baseline in patients with CR or PR (n=30)+No progressive disease at last disease assessment
Best Percentage Change and Longitudinal Change in Target Lesion Size
KEYNOTE-059 Cohort 3: Safety and efficacy of pembrolizumab monotherapy for first-line treatment of patientswith PD-L1-positive advanced gastric/gastroesophageal cancer
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017
PFS and OS
KEYNOTE-059 Cohort 3: Safety and efficacy of pembrolizumab monotherapy for first-line treatment of patientswith PD-L1-positive advanced gastric/gastroesophageal cancer
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017
Treatment-Related Adverse Events
• Median (range) duration of exposure was 2.8 (0.7-20.3) months• 1 patient each experienced grade 3 adverse events (neutropenia, diffuse
uveal melanocytic proliferation, colitis, bile duct obstruction, decreasedneutrophils, dehydration, hyponatremia, and rash
• 1 patient died because of a treatment-related adverse events during safetyfollow-up (pneumonitis)
adeath occurred during safety follow-up
KEYNOTE-059 Cohort 3: Safety and efficacy of pembrolizumab monotherapy for first-line treatment of patientswith PD-L1-positive advanced gastric/gastroesophageal cancer
Event, n (%) N = 31
Any 24 (77)
Grades 3-5 7 (23)
Led to discontinuation 0
Led to deathPneumonitisa
1 (3)1 (3)
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017
Immune-Mediated Adverse Eventsa
aBased on a list of terms specified by the sponsor and included regardless of attribution to studytreatment or immune relatedness by the investigator
Event, n (%)
N=31
All Grades in >2 Patients
Grades 3 Grade 5
Any 10 (32) 2 (7) 1 (3)
Pneumonitis 4 (13) 0 1 (3)
Colitis 1 (3) 1 (3) 0
Rash 1 (3) 1 (3) 0
KEYNOTE-059 Cohort 3: Safety and efficacy of pembrolizumab monotherapy for first-line treatment of patientswith PD-L1-positive advanced gastric/gastroesophageal cancer
Data cutoff: April 21, 2017
Wainberg et al., Abstract LBA28, ESMO 2017
Updated KEYNOTE-059 results
• Pembrolizumab continues to demonstrate, in patients with advanced G/GEJ cancer,
• Promising antitumor activity and durable response as monotherapy in patients whose disease hasprogressed after ≥2 prior lines of therapy
• Encouraging antitumor activity in combination with chemotherapy in previously untreated patients
• Encouraging antitumor activity as monotherapy in previously untreated patients with PD-L1-positive tumors
• Responses were regardless of PD-L1 expression, but higher in patients with PD-L1-positive tumors in cohorts 1 and 2
• Safety was manageable and consistent with that previous reports: no new safety signals
Wainberg et al., Abstract LBA28, ESMO 2017
KEYNOTE-059 Cohort 3: Safety and efficacy of pembrolizumab monotherapy for first-line treatment of patientswith PD-L1-positive advanced gastric/gastroesophageal cancer
Summary and Conclusion (I)
• Results support further development of pembrolizumab for advanced G/GEJ cancer
• KEYNOTE-061 (NCT02370498) is an ongoing, phase 3, randomized clinical study of pembrolizumab versus paclitaxel in patients
with advanced G/GEJ cancer whose disease progressed after first-line therapy with platinum and fluoropyrimidine
• KEYNOTE-062 (NCT02494583) is an ongoing, phase 3, randomized clinical study of pembrolizumab alone or in combination with
chemotherapy versus chemotherapy alone as first-line therapy for advanced PD-L1 positive G/GEJ cancer
Wainberg et al., Abstract LBA28, ESMO 2017
KEYNOTE-059 Cohort 3: Safety and efficacy of pembrolizumab monotherapy for first-line treatment of patientswith PD-L1-positive advanced gastric/gastroesophageal cancer
Summary and Conclusion (II)
Anti-CTLA-4 or anti-PD-1 or anti-PD-L1
Perioperative First-line Second-lineThird-line +
refractory to standard
Ipilimumab (BMS): anti-CTLA-4
Phase III 649/648Nivo Ipi vs CTX
Nivolumab (ONO/BMS):anti-PD-1
Phase III Adjuvant 577 Nivo vs placeboESO after RCTX
Phase III 649/648Nivo Ipi vs CTX
ONO-473 Phase III
Nivo vs Taxanes
ONO-4538-07Phase II
Nivo
Pembrolizumab (MSD):anti-PD-1
KEYNOTE-062 Ph IIIPembro vs Pembro, Cis,
5-FU vs Cis, 5-FU
KEYNOTE-181 Phase IIIPembro vs standard of care
KEYNOTE-061 Phase IIIPembro vs paclitaxel
KEYNOTE-180Phase IIPembro
Durvalumab (AZ): anti-PD-L1
Phase IIAdjuvant
Phase IIDurvalumab vs placebo
Phase Ib/IIDurva vs Treme vs Durva+Treme
Phase Ib and IIDurva+Treme
Atezolizumab (Roche): anti-PD-L1
Phase IIPerioperative FOLFOX/ FLOT
+/- atezolizumab
Avelumab (Merck/Pfizer)anti-PD-L1
JAVELIN GASTRIC 100Phase III
Maintenance after FOLFOX
(Phase I) and Phase IIIAvelumab
JAVELIN GASTRIC 300
After ESMO: even more trials for oesophagogastriccancer
1. Study designs available at: www.clinicaltrials.gov (accessed September 2017). Agents have not yet received EMA approval for treatment of indication listed
HCC
• Disease assessment imaging (CT or MRI) every 6 weeks
• Interim analysis data cutoff date: August 8, 2016
– Median follow-up was 13.3 months in the dose-escalation phase and 10.5 months in the dose-expansion phase
Dose Escalation
0.1–10 mg/kg
N = 48
Dose Expansion
3 mg/kg
N = 214
Uninfected (n = 23)
HCV infected (n = 10)
HBV infected (n = 15)
Uninfected (n = 113)
HCV infected (n = 50)
HBV infected (n = 51)
Sorafenib Experienced (2L)(n = 37)
Sorafenib Naive (1L)(n = 11)
Sorafenib Experienced (2L)(n = 145)
Sorafenib Naive (1L)(n = 69)
Study Endpoints
Primary
• Safety and tolerability (escalation)
• Objective response ratea (expansion)
Secondary
• Objective response ratea (escalation)
• Disease control rate
• Time to response
• Duration of response
• Overall survival
Other
• Biomarker assessments
• Patient-reported outcomesb
a RECIST v1.1; b Baseline and every 6 weeks through week 25 using the EQ-5D utility index and visual analog scale (VAS).
All
Pa
tie
nts
(N
= 2
62
)
CheckMate 040 Study Design
88ASCO GI 2017
Safety: Dose-Expansion Phase
Patients, n (%)
Uninfected (n = 113)
HCV Infected(n = 50)
HBV Infected(n = 51)
All Dose Expansion(N = 214)
Any Grade
Grade 3/4
Any Grade
Grade 3/4
Any Grade
Grade 3/4
Any Grade
Grade 3/4
Any treatment-related AE (TRAE) 84 (74) 22 (19) 40 (80) 15 (30) 35 (69) 3 (6) 159 (74) 40 (19)
TRAEs (≥ 5%)
Fatigue 34 (30) 2 (2) 8 (16) 1 (2) 7 (14) 0 49 (23) 3 (1)
Pruritus 18 (16) 0 14 (28) 1 (2) 13 (25) 0 45 (21) 1 (<1)
Rash 16 (14) 2 (2) 9 (18) 0 8 (16) 0 33 (15) 2 (1)
Diarrhea 19 (17) 2 (2) 5 (10) 0 3 (6) 1 (2) 27 (13) 3 (1)
Nausea 10 (9) 0 6 (12) 0 1 (2) 0 17 (8) 0
Dry mouth 9 (8) 0 2 (4) 0 2 (4) 0 13 (6) 0
Decreased appetite 6 (5) 0 2 (4) 1 (2) 3 (6) 0 11 (5) 1 (<1)
Laboratory TRAEs (≥ 5%)
AST increase 9 (8) 4 (4) 6 (12) 5 (10) 1 (2) 0 16 (7) 9 (4)
ALT increase 7 (6) 2 (2) 7 (14) 3 (6) 3 (6) 0 17 (8) 5 (2)
89ASCO GI 2017
NR, not reached.
TTR, median (range), mo 1.9 (1.4–5.6)
DOR, median (range), mo 17.1 (7.2–32.5+)
TTR, median (range), mo 2.7 (1.2–9.6)
DOR, median (range), mo NR (1.4–9.8+)
Uninfected(n = 14)
HCV(n = 7)
HBV(n = 6)
Dose Escalation Dose Expansion
Time to Response and Duration of Response
Sorafenib Experienced (2L)
Uninfected(n = 3)
HCV(n = 2)
HBV(n = 1)
0.1 mg/kg
1 mg/kg
0.3 mg/kg
10 mg/kg
1 mg/kg
3 mg/kg
Investigator Assessment
CR
Censored with ongoing response
Last dose
PR
Last dose when off treatment
90ASCO GI 2017
Overall SurvivalSorafenib Experienced (2L)
OS Rate (95% CI), % Dose Escalation (n = 37) Dose Expansion (n = 145)
6 months 67 (49–80) 82 (74–87)
9 months 67 (49–80) 71 (63–78)a
12 months 58 (40–72) NC
18 months 46 (29–62) NCa Data cutoff August 8, 2016. NC, not available/not calculated.
0.2
Dose Expansion: Median OS (95% CI), mo = 13.2 (13.2–NR)
Dose Escalation: Median OS (95% CI), mo = 15.0 (5.0–20.2)
Pro
bab
ility
of
Surv
ival
Months
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.1
0.0
0 3 6 9 12 15 18 21 24 27 30 33 36 39
91ASCO GI 2017
Response by PD-L1 Expression
• Responses were observed irrespective of PD-L1 expression on tumor cells
Sorafenib Experienced (2L)
Dose Expansion
Ch
an
ge i
n T
arg
et
Le
sio
n
Siz
e F
rom
Ba
se
lin
e (
%)
Patients
100
50
0
-50
-100
PD-L1 < 1% PD-L1 ≥ 1%
ORR, n/N (%)
95% CI, %
4/26 (15.4),
4.4–34.9
2/9 (22.2),
2.8–60
PD-L1 < 1% PD-L1 ≥ 1%
ORR, n/N (%)
95% CI, %
17/99 (17.2)
10.3–26.1
8/25 (32.0),
14.9–53.5
≥ 1% NAPD-L1: < 1%
-100Patients
100
50
0
-50
≥ 1% NAPD-L1: < 1%
Investigator Assessment
Dose Escalation
92ASCO GI 2017
Other GI cancer
Attraction 01 trialNivolumab monotherapy for advanced pretreated oesophageal cancer
94
Best Overall Response/ORR Investigator Central Review
Response N % (95% CI) N % (95% CI)
Complete response (CR) 2 3.1 (0.9, 10.7) 3 4.7 (1.6, 12.9)
Partial response (PR) 12 18.8 (11.1, 30.0) 8 12.5 (6.5, 22.8)
Stable disease (SD) 20 31.3 (21.2, 43.4) 16 25.0 (16.0, 36.8)
Progressive disease (PD) 29 45.3 29 45.3
NE 1 1.6 8* 12.5
ORR (CR+PR) 14 21.9 (13.5, 33.4) 11 17.2 (9.9, 28.2)
*Including subjects who have no target lesion
64 patients were assessable for the primary endpoint as one patient was excluded due to having multiple primary cancers10
Summary of Clinical Activity (N=64): As of Nov. 17, 2016
95
Results
Figure 2. Progression-Free Survival and Overall Survival (N=64): As of Nov. 17, 2016
CI = confidence interval
Nivolumab in L2+ Anal cancerNCI9673: Primary Endpoint of Response Rate
Lancet Oncology 2017 Cathy Eng
RESULTATS• Taux de réponse :
• 9/37 = 24% (IC95=15-33) en réponse dont 2 réponses complètes, et 1 réponse partielle chez un patient VIH
• 7/9 des répondeurs (78%) ont maintenu cette réponse jusqu'à l'analyse des données (temps médian 5.8mois IC95=3.9-8.1, durée la plus longue à 10.4mois)
• Réponse radiologique de -70% (IC95=57-90)
• Taux de stabilisation : • 17/37 = 47% (IC95=60-63) en maladie stable,
soit au total 72% (IC95=53-84) de contrôle de la maladie à la 1ère évaluation
Bryan S. Balancing cost-effectiveness with other values: the NICE experience - Gesundheitswesen 2009; 71: S30-S33
Conclusion
J Taieb WCGIC Barcelona 2016