HHS Public Access

Author manuscript
Nature. Author manuscript; available in PMC 2022 June 15.
Author Manuscript

Published in final edited form as:

Nature. 2021 December ; 600(7890): 727–730. doi:10.1038/s41586-021-04161-3.

Combined PD-1 and HER2 blockade for HER2-positive gastric

cancer
Yelena Y. Janjigian, MD1, Akihito Kawazoe, MD, PhD2, Patricio Yañez, MD3, Ning Li, MD4,
Sara Lonardi, MD5, Oleksii Kolesnik, MD6, Olga Barajas, MD7, Yuxian Bai, MD8, Lin Shen,
MD, PhD9, Yong Tang, MD10, Lucjan S. Wyrwicz, MD, PhD11, Jianming Xu, MD12, Kohei
Shitara, MD2, Shukui Qin, MD, PhD13, Eric Van Cutsem, MD, PhD14, Josep Tabernero, MD,
Author Manuscript

PhD15, Lie Li, PhD16, Sukrut Shah, PhD16, Pooja Bhagia, MD16, Hyun Cheol Chung, MD,
PhD17
1Memorial Sloan Kettering Cancer Center and Weill Cornell Medical College, New York, NY, USA
2National Cancer Hospital East, Kashiwa, Japan
3Universidad de La Frontera, James Lind Cancer Research Center, Temuco, Chile
4Henan Cancer Hospital, the Affiliated Cancer Hospital of Zhengzhou University, Zhengzhou,
China
5Istituto Oncologico Veneto IOV-IRCCS, Padova, Italy
6Medical Center “Oncolife”, Zaporizhzhia, Ukraine
7Arturo López Pérez Foundation, Santiago, Chile
Author Manuscript

8Harbin Medical University Cancer Hospital, Harbin, China

9Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/Beijing),
Peking University Cancer Hospital & Institute, Beijing, China
10Cancer Hospital Affiliated to Xinjiang Medical University, Xinjiang, China
11Maria Sklodowska-Curie Memorial Cancer Center and Institute of Oncology, Warsaw, Poland
12The Fifth Medical Center, Chinese PLA General Hospital, Beijing, China
13Cancer Center of People’s Liberation Army, Nanjing, China
14University Hospitals Gasthuisberg and KU Leuven, Leuven, Belgium
Author Manuscript

15Vall
d’Hebron Hospital Campus and Institute of Oncology (VHIO), IOB-Quiron, UVic-UCC,
Barcelona, Spain

Correspondence should be addressed to Yelena Y. Janjigian, MD; Memorial Sloan Kettering Cancer Center, 300 E66th Street, Room
1001, New York, NY 10065 USA; Telephone: +1 646-888-4286; janjigiy@mskcc.org; Twitter: @YJanjigianMD.
Author Contributions
Y.Y.J., L.S., K.S., S.Q., E.V.C., J.T., P.B., and H.C.C. conceived, designed, and planned the study. Y.Y.J., A.K., P.Y., N.L., S.L., O.K.,
O.B., Y.B., L.S., Y.T., L.S.W., J.X., and H.C.C. collected data. P.B. provided study oversight. L.L. performed the statistical analysis.
Y.Y.J., L.L., S.S., and P.B. prepared the first draft of the manuscript. Y.Y.J, A.K., P.Y., N.L., S.L., O.K., O.B., Y.B., L.S., Y.T., L.S.W.,
J.X., K.S., S.Q., E.V.C., J.T., L.L., S.S., P.B., and H.C.C. interpreted the results, provided critical review and revision of the drafts, and
approved the decision to submit for publication.
Reprints and permissions information is available at www.nature.com/reprints.
Supplementary Information is available for this paper.
 Janjigian et al. Page 2

16Merck & Co., Inc., Kenilworth, NJ, USA

Author Manuscript

17Yonsei Cancer Center, Yonsei University College of Medicine, Seoul, South Korea

SUMMARY
Human epidermal growth factor receptor 2 (ERBB2/HER2) amplification or overexpression
occurs in approximately 20% of advanced gastric or gastroesophageal junction
adenocarcinomas1–3. Over a decade ago, combination therapy with the anti–HER2 antibody
trastuzumab and chemotherapy became the standard first-line treatment for these patients4.
Although adding the anti–programmed death 1 (PD-1) antibody pembrolizumab to chemotherapy
does not significantly improve efficacy in advanced HER2-negative gastric cancer5, there are
preclinical6–19 and clinical20,21 rationales for adding pembrolizumab in HER2-positive disease.
Here we describe results of the protocol-specified first interim analysis of the randomized, double-
blind, placebo-controlled phase III KEYNOTE-811 study of pembrolizumab plus trastuzumab and
Author Manuscript

chemotherapy for unresectable or metastatic, HER2-positive gastric or gastroesophageal junction

adenocarcinoma (ClinicalTrials.gov, NCT03615326)22. We show that adding pembrolizumab to
trastuzumab and chemotherapy dramatically reduces tumour size, induces complete responses in
some participants, and significantly improves objective response rate.

A pre-existing immune response to tumour antigens is associated with clinical activity

of immune checkpoint blockade23. Enabling such a response through use of tumour-
targeting antibodies for facilitation of cross-priming or chemotherapeutic agents for
inducing immunogenic cell death or reducing immune suppressive populations is a
logical next step in the development of combinatorial anticancer therapies. In preclinical
models and patient-derived samples, trastuzumab increases HER2 internalization and cross-
presentation by dendritic cells6, which in turn stimulates HER2-specific T-cell responses7–9.
Author Manuscript

Trastuzumab has also been shown to upregulate expression of PD-1 and its ligand,
PD-L1, induce expression of tumour-infiltrating lymphocytes, and modulate expression
of major histocompatibility complex class II10–14. Following treatment with oxaliplatin,
dying cancer cells can stimulate dendritic cells, which enhances antigen processing and
presentation and facilitates priming of CD8+ tumour-specific T cells15–17. Coadministration
of immune checkpoint inhibitors and trastuzumab has been shown to enhance HER2-specific
T-cell responses, promote immune cell trafficking, and induce expansion of peripheral
memory T cells7,8,14,18,19. The possible mechanistic interaction of these modalities has been
demonstrated in a HER2-positive transgenic mouse model in which dual inhibition of PD-1
and HER2 resulted in tumour eradication14.

Clinical efficacy and manageable safety for the combination of pembrolizumab,

Author Manuscript

trastuzumab, and chemotherapy in patients with HER2-positive advanced oesophageal,

gastroesophageal, or gastric adenocarcinoma have been demonstrated in two single-
arm, phase II studies20,21. To further study the combination, we conducted the
randomized, double-blind, phase III KEYNOTE-811 trial of pembrolizumab or placebo
in combination with trastuzumab and the investigator’s choice of chemotherapy with 5-
fluorouracil and cisplatin or capecitabine and oxaliplatin in participants with previously
untreated unresectable or metastatic, HER2-positive gastric or gastroesophageal junction

Nature. Author manuscript; available in PMC 2022 June 15.

 Janjigian et al. Page 3

adenocarcinoma (ClinicalTrials.gov, NCT03615326)22. We present results of the protocol-

Author Manuscript

specified first interim analysis of objective response rate for the first 264 participants
enrolled (efficacy population) and safety for all enrolled participants who received at least
one dose of study treatment as of the June 17, 2020, data cutoff (as-treated population).

Of the 434 participants randomly allocated to treatment between October 5, 2018, and June
17, 2020 (intention-to-treat population), 433 received at least one dose of study therapy
—217 of 217 allocated to the pembrolizumab group and 216 of 217 allocated to the
placebo group—and were included in the as-treated population. The efficacy population
included 133 participants in the pembrolizumab group and 131 in the placebo group.
The median (range) study follow-up, defined as the time from randomization to the June
17, 2020, data cutoff was 9.9 months (0.1–19.4) in the intention-to-treat population and
12.0 months (8.5–19.4) in the efficacy population. Calculations of follow-up duration that
account for participant death are found in Extended Data Table 1. The disposition for both
Author Manuscript

populations is shown in Extended Data Table 2. Baseline characteristics were as expected

and generally balanced between groups in both populations (Extended Data Table 3). In
the intention-to-treat population, 81.3% of participants were male, 68.4% had primary
tumours in the stomach, 50.2% had an intestinal histologic subtype, and 80.6% had HER2
immunohistochemistry 3+ tumours. The chosen chemotherapy regimen was capecitabine
plus oxaliplatin for 86.6% of participants and 5-fluorouracil plus cisplatin for the remaining
13.4%.

In the efficacy population, we observed an objective response rate assessed per Response
Evaluation Criteria in Solid Tumors (RECIST), version 1.1, by blinded, independent
central review of 74.4% (95% CI, 66.2–81.6) in the pembrolizumab group and of 51.9%
(95% CI, 43.0–60.7) in the placebo group. This resulted in a statistically significant
Author Manuscript

22.7% improvement in objective response rate in the pembrolizumab group (95% CI, 11.2–
33.7; P=0.00006). We found that treatment differences in key participant subgroups were
generally consistent with that of the total population (Extended Data Fig. 1). Responses in
the pembrolizumab group were also deeper than those in the placebo group (median change
from baseline, –65% vs –49%; ≥80% decrease from baseline, 32.3% vs 14.8%) (Fig. 1),
and complete responses were more frequent (11.3% vs 3.1%) (Table 1). Among responders,
50.5% in the pembrolizumab group and 44.1% in the placebo group had ongoing response at
data cutoff. By Kaplan-Meier estimation, 70.3% of responders in the pembrolizumab group
and 61.4% of responders in the placebo group had response duration of at least 6 months
and 58.4% and 51.1%, respectively, had response duration of at least 9 months; median
response duration was 10.6 months (range, 1.1+ to 16.5+) in the pembrolizumab group and
9.5 months (range, 1.4+ to 15.4+) in the placebo group.
Author Manuscript

In the as-treated population, median (range) treatment duration was 6.2 months (2 days-17.7
months) in the pembrolizumab group and 5.3 months (1 day-17.8 months) in the placebo
group. We observed a similar incidence of adverse events in the pembrolizumab and
placebo groups; adverse events were of grade 3–5 severity in 57.1% of participants in
the pembrolizumab group versus 57.4% in the placebo group and led to discontinuation of
any study treatment in 24.4% versus 25.9% (Table 2). The most common adverse events in
both groups were diarrhoea (52.5% in the pembrolizumab group vs 44.4% in the placebo

Nature. Author manuscript; available in PMC 2022 June 15.

 Janjigian et al. Page 4

group), nausea (48.8% vs 44.4%), and anaemia (41.0% vs 44.0%) (Extended Data Table 4).
Author Manuscript

Adverse events with a possibly immune-mediated cause and/or infusion reactions occurred
in 33.6% and 20.8% of participants, respectively; the most common of these events in
the pembrolizumab group were infusion-related reactions (18.0% vs 13.0% in the placebo
group) and pneumonitis (5.1% vs 1.4%) (Extended Data Table 5). Seven (3.2%) participants
in the pembrolizumab group and 10 (4.6%) in the placebo group died from adverse events;
these were attributed to study treatment by investigators in two (0.9%) participants in each
group and were considered immune-mediated in three (1.4%) and 1 (0.5%) participants,
respectively (Extended Data Table 6).

Discussion
Results of this protocol-specified first interim analysis of KEYNOTE-811 show that
adding pembrolizumab to standard therapy with trastuzumab and chemotherapy results
Author Manuscript

in a statistically significant, clinically meaningful improvement in objective response rate

compared with trastuzumab and chemotherapy alone as first-line therapy for unresectable
or metastatic, HER2-positive gastric or gastroesophageal junction adenocarcinoma.
We also found responses in the pembrolizumab group to be deeper and more
durable. The safety profile of pembrolizumab plus trastuzumab and chemotherapy was
manageable. The observed adverse events were as expected based on previous studies
of pembrolizumab and trastuzumab plus chemotherapy4,5,20,21,24, with no new events
identified for the combination. Based on these findings, the United States Food and
Drug Administration approved pembrolizumab in combination with trastuzumab and
fluoropyrimidine- and platinum-containing chemotherapy for first-line treatment of locally
advanced or unresectable or metastatic HER2-positive gastric or gastroesophageal junction
adenocarcinoma. KEYNOTE-811 is continuing as planned, and results for the primary end
Author Manuscript

points of overall and progression-free survival will be assessed at a later date in accordance
with the statistical analysis plan.

Several phase III trials have studied the addition of PD-1 inhibitors to first-line
chemotherapy in patients with HER2-negative advanced gastric or gastroesophageal
cancer.5,25,26 In these studies, objective response rate in the combination groups ranged
from 49% to 60%, representing an improvement of approximately 10% to 15% by adding
a PD-1 inhibitor. The 74.4% objective response rate observed when adding pembrolizumab
to trastuzumab and chemotherapy in KEYNOTE-811, representing a 22.7% improvement
versus trastuzumab and chemotherapy, supports preclinical data suggesting that there may be
synergy between dual HER2 and PD-1 inhibition7,8,14,18,19.

To support biomarker analyses, we required all participants in KEYNOTE-811 to provide

Author Manuscript

tissue samples at baseline and included optional collection of an on-treatment biopsy

sample; blood samples for biomarker assessment are also captured throughout treatment
to support analysis of genetic, plasma, and serum biomarkers and circulating tumour
DNA. To-date, the only biomarkers we have explored are MSI-high status and PD-L1
expression. Three participants (0.7%) in the intention-to-treat population were known to
have MSI-high tumours, a molecularly defined subset of gastric cancer associated with
a great likelihood of response to PD-1 inhibitors27. This low frequency is not surprising

Nature. Author manuscript; available in PMC 2022 June 15.

 Janjigian et al. Page 5

because HER2-positive tumours are characterized by chromosomal instability and lower

tumour mutational burden28. PD-L1 expression has been shown to correlate with efficacy of
Author Manuscript

pembrolizumab monotherapy in gastric cancer5,24. In KEYNOTE-811, 84.1% of participants

had a PD-L1 combined positive score of ≥1. We observed a greater difference in objective
response rate in participants with PD-L1 combined positive score ≥1, although the 95%
confidence intervals for the combined positive score ≥1 and <1 subgroups overlapped. It will
be interesting to see whether the relative benefit of pembrolizumab plus trastuzumab and
chemotherapy for overall survival will be impacted by PD-L1 expression.

These promising initial findings of KEYNOTE-811 suggest that the combination of

pembrolizumab, trastuzumab, and chemotherapy may be a transformative treatment option
for HER2-positive gastric or gastroesophageal junction adenocarcinoma and support
exploring the combination in patients with earlier stages of disease. As an example of this
approach, an ongoing phase II study is exploring the role of pembrolizumab and trastuzumab
Author Manuscript

as adjuvant therapy for patients with HER2-positive oesophageal or gastric cancer with
minimal residual disease (NCT04510285).

METHODS
Study Design and Participants
KEYNOTE-811 is a global randomized, double-blind, placebo-controlled, phase III study
designed to assess the efficacy and safety of adding pembrolizumab to standard-of-care
therapy with trastuzumab and platinum-based chemotherapy as first-line therapy for
HER2-positive advanced gastric or gastroesophageal cancer (ClinicalTrials.gov identifier,
NCT03615326) being conducted at 168 medical centres in Australia, Brazil, Chile, China,
France, Germany, Guatemala, Ireland, Israel, Italy, Japan, New Zealand, Poland, Russia,
Author Manuscript

South Korea, Spain, Turkey, the United Kingdom, and the United States (Supplementary
Information). The study was designed by academic advisors and employees of the study
sponsor. An external data and safety monitoring committee oversees the trial by periodically
assessing safety and assessing efficacy at prespecified interim analyses. The study protocol
and all amendments were approved by the appropriate ethics body at each participating
study centre (Supplementary Information). All participants provided written informed
consent. This ongoing study is being conducted in accordance with Good Clinical Practice
guidelines.

Full eligibility criteria have been published22. Briefly, eligible participants are aged 18
years or older; have previously untreated, unresectable or metastatic, histologically or
cytologically confirmed adenocarcinoma of the stomach or gastroesophageal junction that is
Author Manuscript

HER2-positive as assessed by central review; have measurable disease per RECIST, version
1.129, as assessed by the investigator; have ECOG performance-status score of 0 or 130; have
life expectancy of more than 6 months; have adequate organ function; and have provided
a tumour sample adequate for assessment of programmed death ligand 1 (PD-L1) and
microsatellite instability (MSI) status.

Nature. Author manuscript; available in PMC 2022 June 15.

 Janjigian et al. Page 6

Randomization and Treatment

Author Manuscript

Participants are randomly allocated in a 1:1 ratio to receive either pembrolizumab at a dose
of 200 mg or placebo (normal saline or dextrose), both administered intravenously once
every 3 weeks, by means of an integrated interactive voice-response and Web-response
system. Randomization is stratified by geographic region (Australia/Europe/Israel/North
America vs Asia vs rest of world), PD-L1 combined positive score (≥1 vs <1), and
investigator’s choice of chemotherapy (5-fluorouracil plus cisplatin vs capecitabine plus
oxaliplatin). In the global cohort, all participants receive trastuzumab at a dose of 6 mg per
kilogram of body weight intravenously once every 3 weeks following an initial loading dose
of 8 mg per kilogram and either 5-flurouracil (800 mg per square meter of body-surface
area administered intravenously on days 1–5 of each 3-week cycle) and cisplatin (80 mg
per square meter administered intravenously once every 3 weeks) or capecitabine (1000
mg per square meter administered orally twice daily on days 1–14 of each 3-week cycle)
Author Manuscript

and oxaliplatin (130 mg per square meter administered intravenously once every 3 weeks).
All treatment is continued for up to 35 cycles (~2 years) or until disease progression,
unacceptable toxic effects, investigator decision, or participant withdrawal of consent. If
toxicity is clearly attributed to one drug, that drug alone may be discontinued.

Assessments
HER2 status is assessed during screening at a central laboratory using
immunohistochemistry (IHC) (Dako HercepTest™) and fluorescence in situ hybridization
(FISH) (Dako HER2 FISH pharmDx™ Kit). HER2 positivity was defined as IHC 3+ or
IHC 2+ with positive ISH or FISH, with ISH and FISH positivity defined as a ratio of
≥2.0 for the number of HER2 copies to the number of signals for CEP17; a ratio of <2.0
was considered positive if the HER2 copy number was >6. PD-L1 expression in formalin-
Author Manuscript

fixed tumour samples is assessed during screening at a central laboratory using the PD-L1
IHC 22C3 pharmDx assay (Agilent Technologies, Inc.) and characterized according to the
combined positive score, calculated as the number of PD-L1–staining cells (tumour cells,
lymphocytes, macrophages) divided by the total number of viable tumour cells, multiplied
by 100; a minimum of 100 viable tumour cells must be present for a sample to be considered
evaluable31. MSI status is determined centrally by polymerase chain reaction using the MSI
Analysis System, version 1.2 (Promega) and characterized as MSI-high if 2 or more markers
are changed compared to normal controls. Tumour imaging by computed tomography
(preferred) or magnetic resonance imaging is scheduled for week 6 and every 6 weeks
thereafter. Response is assessed according to RECIST, version 1.1, by blinded independent
central review for determination of study endpoints and according to iRECIST32 by the
investigator to inform treatment decisions. Adverse events and laboratory abnormalities are
Author Manuscript

collected regularly during study treatment and for up to 30 days thereafter (up to 90 days
for serious adverse events), classified according to the Medical Dictionary for Regulatory
Affairs, version 23.0, and graded according to the National Cancer Institute Common
Terminology Criteria for Adverse Events, version 4.03. Following treatment discontinuation,
participants are assessed for survival every 12 weeks.

Nature. Author manuscript; available in PMC 2022 June 15.

 Janjigian et al. Page 7

Statistical Analysis
Author Manuscript

The dual primary endpoints are progression-free survival assessed per RECIST, version
1.1, by blinded independent central review and overall survival. Secondary end points
are objective response and duration of response assessed per RECIST, version 1.1, by
blinded independent central review and safety. Efficacy is assessed in all randomly allocated
participants according to the allocated treatment. Safety is assessed in the as-treated
population (i.e., all randomly allocated participants who received ≥1 dose of study treatment
according to the treatment received).

The full statistical analysis plan specifies the performance of three interim analyses and
a final analysis. An extension of the graphical method of Mauer and Bretz is used to
control the family-wise type I error rate at a one-sided α=0.025 across all hypotheses and
interim analyses. Planned enrolment in the global cohort is 692 participants and is based
Author Manuscript

on the following assumptions: an enrolment period of 28 months and a ramp-up enrolment

period of 6 months; the duration of progression-free survival and overall survival follows
an exponential distribution; median progression-free survival is 6.7 months in the placebo
group and the true hazard ratio is 0.7; and median overall survival is 13.8 months in the
placebo group and the true hazard ratio is 0.75. Per protocol, the first interim analysis was
to be performed when the first 260 participants enrolled had at least 8.5 months of follow-up
and was to test the objective response hypothesis. With approximately 260 participants,
the study has approximately 90% power to detect a difference in objective response of 25
percentage points (73% vs 48%) favouring the pembrolizumab group at one-sided α=0.002.
Power and interim analysis calculations were performed using the gsDesign R package
version 3.0.1.

Study data were captured in InForm version 4.6 (Oracle). All statistical analyses were
Author Manuscript

performed using SAS version 9.4 (SAS Institute, Inc.). No data from the specified analysis
populations were excluded from analysis. The difference in objective response and its
95% CI were calculated using the Miettinen and Nurminen method stratified by the
randomization stratification factors of geographic region (Australia/Europe/Israel/North
America vs Asia vs rest of world), PD-L1 combined positive score (≥1 vs <1), and
investigator’s choice of chemotherapy (5-fluorouracil plus cisplatin vs capecitabine plus
oxaliplatin) with strata weighting by sample size. Duration of response was estimated using
the Kaplan-Meier method. All reported data are based on the first interim analysis, which
had a data cutoff of June 17, 2020.
Author Manuscript

Nature. Author manuscript; available in PMC 2022 June 15.

 Janjigian et al. Page 8

Extended Data
Author Manuscript
Author Manuscript
Author Manuscript

Extended Data Fig. 1. Treatment difference in objective response in subgroups of the efficacy
population.
Response was assessed per RECIST, version 1.1, by blinded, independent central review.
The estimated treatment difference between the pembrolizumab and placebo groups in the
overall population was calculated using the Miettinen and Nurminen method stratified by
geographic region (Australia/Europe/Israel/North America [Aus/Eur/Isr/NAm] vs Asia vs
rest of world), PD-L1 combined positive score ([CPS]; ≥1 vs <1), and chemotherapy choice
(5-fluorouracil plus cisplatin [FP] vs capecitabine plus oxaliplatin [CAPOX]); differences
in subgroups were calculated using the unstratified Miettinen and Nurminen method. The
efficacy population included the first 264 participants randomly allocated to treatment.
Author Manuscript

The treatment regimen included trastuzumab and chemotherapy in both groups. Diamonds
represent the estimated treatment difference in objective response, the error bars represent
the 95% confidence interval (CI) of the estimated treatment difference, and the region
shaded in dark grey represents the 95% CI for the treatment difference in the overall
population. ECOG, Eastern Cooperative Oncology Group; GEJ, gastroesophageal junction;
IHC, immunohistochemistry; ISH, in situ hybridization.

Nature. Author manuscript; available in PMC 2022 June 15.

 Janjigian et al. Page 9
Author Manuscript

Extended Data Table 1.

Follow-up duration calculated as time from randomization to the date of death or the
database cutoff date, whichever was earliest

Follow-up duration Pembrolizumab group Placebo group

Intention-to-treat population N=217 N=217

Median,* months 8.4 7.7
Range, months 0.1–19.0 0.5–17.9
Efficacy population N=133 N=131
Median,* months 11.1 10.4
Range, months 2.2–19.0 0.5–17.9
*
Author Manuscript

Medians were calculated using the observation time method.

Extended Data Table 2.

Disposition of study treatment

Efficacy population Intention-to-treat population

Pembrolizumab group Placebo group Pembrolizumab group Placebo group
Variable — no. (%) (N=133) (N=131) (N=217) (N=217)

Started 133 130 217 216

Discontinued all study
treatment 79 (59.4) 93 (71.5) 90 (41.5) 112 (51.9)
Adverse event 7 (5.3) 10(7.7) 10(4.6) 14(6.5)
Clinical progression 4 (3.0) 9 (6.9) 6 (2.8) 12 (5.6)
Non-study anticancer 2(1.5) 3 (2.3) 2 (0.9) 3(1.4)
Author Manuscript

therapy
Physician decision 1 (0.8) 2(1.5) 1 (0.5) 2 (0.9)
Radiographic
progression 59 (44.4) 64 (49.2) 65 (30.0) 76 (35.2)
Withdrawal by
participant 6 (4.5) 5 (3.8) 6 (2.8) 5 (2.3)
Continuing any study
treatment 54 (40.6) 37 (28.5) 127 (58.5) 104 (48.1)

All percentages are calculated out of the number of participants who started study treatment. The treatment regimen
included trastuzumab and chemotherapy in both groups.
Extended Data Table 3.

Demographic and disease characteristics at baseline

Author Manuscript

Efficacy population Intention-to-treat population

Pembrolizumab group Placebo group Pembrolizumab group Placebo group
characteristic (N=133) (N=131) (N=217) (N=217)

Age
Median (range) — yr 62 (19–84) 61 (32–83) 62 (19–84) 63 (32–83)
≥65 yr — no. (%) 55 (41.4) 53 (40.5) 89 (41.0) 98 (45.2)
Male sex — no. (%) 112 (84.2) 104 (79.4) 179 (82.5) 174 (80.2)

Nature. Author manuscript; available in PMC 2022 June 15.

 Janjigian et al. Page 10
Author Manuscript

Efficacy population Intention-to-treat population

Pembrolizumab group Placebo group Pembrolizumab group Placebo group
characteristic (N=133) (N=131) (N=217) (N=217)
Region of enrolment — no. (%)
Australia/Europe/lsrael/
North America 41 (30.8) 44 (33.6) 67 (30.9) 67 (30.9)
Asia 40 (30.1) 39 (29.8) 76 (35.0) 75 (34.6)
Rest of world 52 (39.1) 48 (36.6) 74 (34.1) 75 (34.6)
ECOG performance-status score — no. (%)
0 65 (48.9) 59 (45.0) 101 (46.5) 89 (41.0)
1 68 (51.1) 72 (55.0) 116(53.5) 128 (59.0)
Primary location at diagnosis — no. (%)
Gastroesophageal
Author Manuscript

junction 37 (27.8) 42 (32.1) 62 (28.6) 75 (34.6)

Stomach 96 (72.2) 89 (67.9) 155 (71.4) 142 (65.4)
No. of metastatic sites — no. (%)
0–2 71 (53.4) 77 (58.8) 117(53.9) 131 (60.4)
≥3 62 (46.6) 54 (41.2) 100 (46.1) 86 (39.6)
Histologic subtype — no. (%)
Diffuse 28 (21.1) 26(19.8) 48 (22.1) 39(18.0)
Intestinal 81 (60.9) 63 (48.1) 117(53.9) 101 (46.5)
Indeterminate 24(18.0) 42 (32.1) 52 (24.0) 77 (35.5)
Previous gastrectomy or oesophagectomy — no. (%)
Yes 22 (16.5) 25(19.1) 34(15.7) 44 (20.3)
No 111 (83.5) 106 (80.9) 183 (84.3) 173 (79.7)
Author Manuscript

PD-L1 combined positive score — no. (%)

≥1 117(88.0) 112(85.5) 184 (84.8) 181 (83.4)
<1 16(12.0) 19(14.5) 33(15.2) 36(16.6)
HER2 status — no. (%)
IHC 1 + 0 (0.0) 0 (0.0) 1 (0.5) 1 (0.5)
IHC 2+ ISH equivocal 0 (0.0) 0 (0.0) 0 (0.0) 1 (0.5)
IHC 2+ ISH positive 24(18.0) 27 (20.6) 36(16.6) 45 (20.7)
IHC 3+ 109 (82.0) 104 (79.4) 180 (82.9) 170 (78.3)
MSI status — no. (%)
MSI-high 1 (0.8) 1 (0.8) 2 (0.9) 1 (0.5)
Non-MSI-high 120 (90.2) 120 (91.6) 171 (78.8) 174 (80.2)
Unknown 12 (9.0) 10(7.6) 44 (20.3) 42(19.4)
Author Manuscript

Sum of target lesions at baseline — no. (%)

<Median 62 (46.6) 65 (49.6) 98 (45.2) 107 (49.3)
≥Median 62 (46.6) 61 (46.6) 107 (49.3) 101 (46.5)
Missing 9 (6.8) 5 (3.8) 12 (5.5) 9(4.1)
Chosen chemotherapy regimen — no. (%)
Capecitabine and
oxaliplatin 115(86.5) 115(87.8) 189 (87.1) 187 (86.2)

Nature. Author manuscript; available in PMC 2022 June 15.

 Janjigian et al. Page 11
Author Manuscript

Efficacy population Intention-to-treat population

Pembrolizumab group Placebo group Pembrolizumab group Placebo group
characteristic (N=133) (N=131) (N=217) (N=217)
5-fluorouracil and 18(13.5) 16(12.2) 28 (12.9) 30(13.8)
cisplatin

The treatment regimen included trastuzumab and chemotherapy in both groups.

Extended Data Table 4.

Adverse events that occurred in 10% or more of participants in either group of the as-treated
population

Pembrolizumab group (N=217) Placebo group (N=216)

Event — no.(%) Any grade Grade 3–5 Any grade Grade 3–5
Author Manuscript

Diarrhoea 114(52.5) 16(7.4) 96 (44.4) 18 (8.3)

Nausea 106 (48.8) 10(4.6) 96 (44.4) 12 (5.6)
Anaemia 89 (41.0) 19(8.8) 95 (44.0) 20 (9.3)
Decreased appetite 67 (30.9) 5(2.3) 69 (31.9) 9 (4.2)
Vomiting 67 (30.9) 10(4.6) 59 (27.3) 4(1.9)
Platelet count decreased 53 (24.4) 17(7.8) 61 (28.2) 15(6.9)
Fatigue 51 (23.5) 9(4.1) 43(19.9) 6(2.8)
Neutrophil count decreased 51 (23.5) 16(7.4) 53 (24.5) 16(7.4)
Peripheral sensory neuropathy 50 (23.0) 6(2.8) 40(18.5) 3(1.4)
Aspartate aminotransferase increased 45 (20.7) 1 (0.5) 28 (13.0) 1 (0.5)
Weight decreased 41 (18.9) 2 (0.9) 37(17.1) 1 (0.5)
Palmar-plantar erythrodysaesthesia syndrome 40(18.4) 1 (0.5) 35(16.2) 2 (0.9)
Author Manuscript

Neutropenia 34(15.7) 11 (5.1) 33(15.3) 11 (5.1)

Constipation 33(15.2) 1 (0.5) 37(17.1) 0(0.0)
Neuropathy peripheral 33(15.2) 3(1.4) 35(16.2) 5(2.3)
Hypokalaemia 32 (14.7) 11 (5.1) 25(11.6) 14 (6.5)
Alanine aminotransferase increased 31 (14.3) 1 (0.5) 22 (10.2) 1 (0.5)
Infusion related reaction 31 (14.3) 4 (1.8) 20 (9.3) 2 (0.9)
Pyrexia 31 (14.3) 1 (0.5) 26(12.0) 0
Hypoalbuminemia 30(13.8) 0(0.0) 32 (14.8) 4(1.9)
White blood cell count decreased 27(12.4) 4 (1.8) 31 (14.4) 5(2.3)
Malaise 24 (11.1) 1 (0.5) 19(8.8) 2 (0.9)
Stomatitis 24 (11.1) 2 (0.9) 22 (10.2) 4(1.9)
Asthenia 22 (10.1) 5(2.3) 35(16.2) 8 (3.7)
Author Manuscript

Blood bilirubin increased 22 (10.1) 3(1.4) 12 (5.6) 1 (0.5)

Thrombocytopenia 20 (9.2) 7(3.2) 25(11.6) 2 (0.9)

The treatment regimen included trastuzumab and chemotherapy in both groups.

Nature. Author manuscript; available in PMC 2022 June 15.

 Janjigian et al. Page 12
Author Manuscript

Extended Data Table 5.

Adverse events with a possible immune-mediated cause and infusion reactions in the as-
treated population

Pembrolizumab group (N=217) Placebo group (N=216)

Event — no.(%) Any grade Grade 3–5 Any grade Grade 3–5

Infusion-related reactions 39(18.0) 6 (2.8) 28 (13.0) 2 (0.9)

Pneumonitis 11 (5.1) 3(1.4) 3(1.4) 0(0.0)
Colitis 10 (4.6) 6(2.8) 4(1.9) 4(1.9)
Hypothyroidism 10 (4.6) 0 (0.0) 6(2.8) 0 (0.0)
Hyperthyroidism 8 (3.7) 0 (0.0) 7 (3.2) 0 (0.0)
Hypophysitis 3(1.4) 1 (0.5) 0 (0.0) 0 (0.0)
Author Manuscript

Hepatitis 2 (0.9) 2 (0.9) 2 (0.9) 0 (0.0)

Severe skin reactions 2 (0.9) 2 (0.9) 0 (0.0) 0 (0.0)
Nephritis 1 (0.5) 0 (0.0) 0 (0.0) 0 (0.0)
Thyroiditis 1 (0.5) 0 (0.0) 0 (0.0) 0(0.0)
Type 1 diabetes mellitus 1 (0.5) 1 (0.5) 0 (0.0) 0 (0.0)
Uveitis 1 (0.5) 0 (0.0) 1 (0.5) 0 (0.0)
Myocarditis 0 (0.0) 0 (0.0) 1 (0.5) 1 (0.5)

Adverse events with a possible immune-mediated cause and infusion reactions were considered regardless of attribution to
study treatment by the investigator. The specific events are based on a list of terms provided by the sponsor. In addition to
the specific terms listed, related terms were also included.
Extended Data Table 6.

Adverse events leading to death in the as-treated population

Author Manuscript

Pembrolizumab group (N-217) Placebo group (N=216)

Any Treatment Immune † Any cause Treatment Immune †
Event — no. (%) cause related* mediated related* mediated

Pneumonitis 2 (0.9) 1 (0.5) 2 (0.9) 0 0 0

Abdominal 1 (0.5)
infection 0 0 0 0 0
Hepatitis 1 (0.5) 1 (0.5) 1 (0.5) 0 0 0
Multiple organ 1 (0.5) 1 (0.5)
dysfunction
syndrome 0 0 0 0
Myocardial
infarction 1 (0.5) 0 0 0 0 0
Pneumonia 1 (0.5) 0 0 1 (0.5) 0 0
Author Manuscript

Aspiration 0 0 0 1 (0.5) 0 0
Cholangitis 0 0 0 1 (0.5) 1 (0.5) 0
Completed suicide 0 0 0 1 (0.5) 0 0
Craniocerebral
injury 0 0 0 1 (0.5) 0 0
Death 0 0 0 1 (0.5) 0 0
Gastric cancer 0 0 0 1 (0.5) 0 0

Nature. Author manuscript; available in PMC 2022 June 15.

 Janjigian et al. Page 13
Author Manuscript

Pembrolizumab group (N-217) Placebo group (N=216)

Any Treatment Immune † Any cause Treatment Immune †
Event — no. (%) cause related* mediated related* mediated
‡ ‡ ‡
Myocarditis 0 0 0 1 (0.5) 1 (0.5) 1 (0.5)
Respiratory tract
0 0 0 1 (0.5) 0 0
infection
*
As indicated by the investigator.
†
Adverse events with a possible immune-mediated cause were considered regardless of attribution to study treatment by the
investigator. The specific events are based on a list of terms provided by the sponsor. In addition to the specific terms listed,
related terms were also included.
‡
Diagnosis of myocarditis was supported by post-mortem histological findings of a CD3-positive T cell population in the
lymphocytic infiltrate in the myocardium; there were no features of myocardial infarction.

Supplementary Material
Author Manuscript

Refer to Web version on PubMed Central for supplementary material.

Acknowledgements
This study was funded by Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA.
Yelena Y. Janjigian was additionally supported by Memorial Sloan Kettering Cancer Center National Institutes
of Health/National Cancer Institute Cancer Center Support Grant P30 CA008748. We thank the patients and
their families and caregivers for participating in the study; the investigators and site personnel; and the following
employees of Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA: Chie-Schin
Shih, MD, for study oversight and Melanie A. Leiby, PhD, for medical writing and editorial assistance.

Competing Interest Declaration

Y.Y.J. has received research funding to the institution from Merck Sharp & Dohme, the National Cancer Institute,
the United States Department of Defense, Cycle for Survival, Fred’s Team, Rgenix, Bayer, Genentech/Roche,
Author Manuscript

Bristol-Myers Squibb, and Eli Lilly, has served on advisory boards for Rgenix, Merck Serono, Bristol-Myers
Squibb, Eli Lilly, Pfizer, Bayer, Imugene, Merck Sharp & Dohme, Daiichi-Sankyo, Zymeworks, SeaGen, Basilea
Pharmaceutical and AstraZeneca, and has equity in Rgenix. A.K. has received research funding to the institution
from Merck Sharp & Dohme, Ono Pharmaceutical Co., Ltd. and Taiho Pharmaceutical Co., Ltd., and received
honoraria for lectures, presentations, speakers bureaus, or educational events from Ono Pharmaceutical Co., Ltd.
and Taiho Pharmaceutical Co., Ltd. P.Y. has received research funding to the institution and travel support from
Merck Sharp & Dohme. N.L. has received research funding to the institution from Merck Sharp & Dohme.
S.L. has received research funding to the institution from Merck Sharp & Dohme, Amgen, Merck Serono,
Bayer, Roche, Lilly, AstraZeneca, and Bristol-Myers Squib, consulting fees from Amgen, Merck Serono, Lilly,
AstraZeneca, Incyte, Daiichi Sankyo, Bristol-Myers Squibb, and Servier, and payment or honoraria for lectures,
presentations, speakers bureaus, or educational events from Roche, Lilly, Bristol-Myers Squibb, Servier, Merck
Serono, Pierre-Fabre, GSK, and Amgen. O.K. has received research funding to the institution from Merck Sharp &
Dohme. O.B. has received research funding to the institution from Merck Sharp & Dohme, has received honoraria
for serving as a speaker from Eli Lilly, travel support from Bristol-Myers Squibb and Roche, participated as an
advisor for Bristol-Myers Squibb and Roche, and has a leadership role in the Sociedad Chilena de Oncología
Medica. Y.B. has received research funding to the institution from Merck Sharp & Dohme. L.S. has received
research funding to the institution from Merck Sharp & Dohme, Beijing Xiantong Biomedical Technology, Qilu
Pharmaceutical, ZaiLab Pharmaceutical (Shanghai), Jacobio Pharmaceuticals, and Beihai Kangcheng (Beijing)
Author Manuscript

Medical Technology, has received consulting fees from Merck Sharp & Dohme, Merck, Mingji Biopharmaceutical,
Haichuang Pharmaceutical, Harbour BioMed, and Boehringer Ingelheim, has received payment for speakers
bureaus from Hutchison Whampoa, Henrui, ZaiLab, and CSTONE Pharmaceutical, and has participated on an
advisory board for Rongchang Pharmaceutical, Zailab, and CSTONE Pharmaceutical. Y.T. has received research
funding to the institution from Merck Sharp & Dohme. L.S.W. has received research funding to the institution
from Merck Sharp & Dohme. J.X. has received research funding to the institution from Merck Sharp &
Dohme. K.S. has received research funding to the institution from Merck Sharp & Dohme, Astellas, Lilly, Ono,
Sumitomo Dainippon, Daiichi Sankyo, Taiho, Chugai, Medi Science, and Eisai, has received consulting fees from
Astellas, Lilly, Bristol-Myers Squibb, Takeda, Pfizer, Ono, Merck Sharp & Dohme, Taiho, Novartis, AbbVie,
GlaxoSmithKline, Daiichi Sankyo, Amgen, and Boehringer Ingelheim, and has received honoraria for lectures
from Novartis, AbbVie, and Yalkult. S.Q. has received consulting fees from Merck Sharp & Dohme for serving

Nature. Author manuscript; available in PMC 2022 June 15.

 Janjigian et al. Page 14

on an advisory board. E.V.C. has received funding to the institution from Merck Sharp & Dohme, Amgen,
Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Celgene, Ipsen, Lilly, Merck KGaA, Novartis, Roche, and
Author Manuscript

Servier and has received consulting fees from Array, Astellas, AstraZeneca, Bayer, Beigene, Biocartis, Boehringer
Ingelheim, Bristol-Myers Squibb, Celgene, Daiichi Sankyo, Halozyme, GSK, Incyte, Ipsen, Merck Sharp &
Dohme, Merck KGaA, Novartis, Pierre-Fabre, Roche, Servier, Sirtex, and Taiho. J.T. has received funding to
the institution from Merck Sharp & Dohme, has received consulting fees from Array Biopharma, AstraZeneca,
Avvinity, Bayer, Boehringer Ingelheim, Chugai, Daiichi Sankyo, F. Hoffmann-La Roche Ltd., Genentech Inc.,
HalioDX SAS, Hutchison MediPharma International, Ikena Oncology, IQVIA, Lilly, Menarini, Merck Serono,
Merus, Merck Sharp & Dohme, Mirati, Neophore, Novartis, Orion Biotechnology, Peptomyc, Pfizer, Pierre Fabre,
Samsung Bioepis, Sanofi, SeaGen, Servier, Taiho, Tessa Therapeutics, and TheraMyc, and has received payment for
educational collaboration with Imedex, Medscape Education, MJH Life Sciences, PeerView Institute for Medical
Education, and Physicians Education Resource (PER). L.L. receives salary for full-time employment from Merck
Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA. S.S. receives salary for full-time
employment from Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA and
owns stock and/or has stock options in Merck & Co., Inc., Kenilworth, NJ, USA. P.B. receives salary for full-time
employment from Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA and
owns stock and/or has stock options in Merck & Co., Inc., Kenilworth, NJ, USA. H.C.C. has received funding
to the institution from Merck Sharp & Dohme. Y.Y.J, A.K., P.Y., N.L., S.L., O.K., O.B., Y.B., L.S., Y.T., L.S.W.,
J.X., K.S., S.Q., E.V.C., J.T., L.L., S.S., P.B., and H.C.C. received medical writing support for this manuscript (no
payment) from Merck Sharp & Dohme.
Author Manuscript

Data Availability
Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA
(MSD) is committed to providing qualified scientific researchers access to anonymized data
and clinical study reports from the company’s clinical trials for the purpose of conducting
legitimate scientific research. MSD is also obligated to protect the rights and privacy of
trial participants and, as such, has a procedure in place for evaluating and fulfilling requests
for sharing company clinical trial data with qualified external scientific researchers. The
MSD data sharing website (available at: http://engagezone.msd.com/ds_documentation.php)
outlines the process and requirements for submitting a data request. Feasible requests will
be reviewed by a committee of MSD subject matter experts to assess the scientific validity
Author Manuscript

of the request and the qualifications of the requestors. In line with data privacy legislation,
submitters of approved requests must enter into a standard data-sharing agreement with
MSD before data access is granted. Data will be made available for request after product
approval in the US and EU or after product development is discontinued; an exception will
be made for this ongoing trial such that data will be made available for request after the
protocol-specified primary endpoint analyses have been reported. There are circumstances
that may prevent MSD from sharing requested data, including country or region-specific
regulations. If the request is declined, it will be communicated to the investigator. Access
to genetic or exploratory biomarker data requires a detailed statistical analysis plan that is
collaboratively developed by the requestor and MSD subject matter experts; after approval
of the statistical analysis plan and execution of a data-sharing agreement, MSD will either
perform the proposed analyses and share the results with the requestor or will construct
Author Manuscript

biomarker covariates and add them to a file with clinical data that is uploaded to a SAS
portal so that the requestor can perform the proposed analyses.

Main Text References

1. Cancer Genome Atlas Research Network. Comprehensive molecular characterization of gastric
adenocarcinoma. Nature 513, 202–209 (2014). [PubMed: 25079317]
2. Van Cutsem E et al. HER2 screening data from ToGA: targeting HER2 in gastric and
gastroesophageal junction cancer. Gastric Cancer 18, 476–484 (2015). [PubMed: 25038874]

Nature. Author manuscript; available in PMC 2022 June 15.

 Janjigian et al. Page 15

3. Cancer Genome Atlas Research Network. et al. Integrated genomic characterization of oesophageal
carcinoma. Nature 541, 169–175 (2017). [PubMed: 28052061]
Author Manuscript

4. Bang YJ et al. Trastuzumab in combination with chemotherapy versus chemotherapy alone for
treatment of HER2-positive advanced gastric or gastro-oesophageal junction cancer (ToGA): a
phase 3, open-label, randomised controlled trial. Lancet 376, 687–697 (2010). [PubMed: 20728210]
5. Shitara K et al. Efficacy and safety of pembrolizumab or pembrolizumab plus chemotherapy vs
chemotherapy alone for patients with first-line, advanced gastric cancer: the KEYNOTE-062 phase
3 randomized clinical trial. JAMA Oncol 6, 1571–1580, (2020). [PubMed: 32880601]
6. Gall VA et al. Trastuzumab increases HER2 uptake and cross-presentation by dendritic cells. Cancer
Res 77, 5374–5383, (2017). [PubMed: 28819024]
7. Park S et al. The therapeutic effect of anti-HER2/neu antibody depends on both innate and adaptive
immunity. Cancer Cell 18, 160–170, (2010). [PubMed: 20708157]
8. Taylor C et al. Augmented HER-2 specific immunity during treatment with trastuzumab and
chemotherapy. Clin Cancer Res 13, 5133–5143, (2007). [PubMed: 17785568]
9. zum Buschenfelde C. M., Hermann C, Schmidt B, Peschel C & Bernhard H Antihuman epidermal
growth factor receptor 2 (HER2) monoclonal antibody trastuzumab enhances cytolytic activity of
Author Manuscript

class I-restricted HER2-specific T lymphocytes against HER2-overexpressing tumor cells. Cancer

Res 62, 2244–2247, (2002). [PubMed: 11956077]
10. Chaganty BKR et al. Trastuzumab upregulates PD-L1 as a potential mechanism of trastuzumab
resistance through engagement of immune effector cells and stimulation of IFNgamma secretion.
Cancer Lett 430, 47–56, (2018). [PubMed: 29746929]
11. Varadan V et al. Immune signatures following single dose trastuzumab predict pathologic response
to preoperative trastuzumab and chemotherapy in HER2-positive early breast cancer. Clin Cancer
Res 22, 3249–3259, (2016). [PubMed: 26842237]
12. Triulzi T et al. HER2 signaling regulates the tumor immune microenvironment and trastuzumab
efficacy. Oncoimmunology 8, e1512942, (2019). [PubMed: 30546951]
13. Triulzi T et al. Early immune modulation by single-agent trastuzumab as a marker of trastuzumab
benefit. Br J Cancer 119, 1487–1494, (2018). [PubMed: 30478407]
14. Stagg J et al. Anti-ErbB-2 mAb therapy requires type I and II interferons and synergizes with
anti-PD-1 or anti-CD137 mAb therapy. Proc Natl Acad Sci U S A 108, 7142–7147, (2011).
Author Manuscript

[PubMed: 21482773]
15. Apetoh L et al. Toll-like receptor 4-dependent contribution of the immune system to anticancer
chemotherapy and radiotherapy. Nat Med 13, 1050–1059, (2007). [PubMed: 17704786]
16. Ghiringhelli F et al. Activation of the NLRP3 inflammasome in dendritic cells induces IL-1beta-
dependent adaptive immunity against tumors. Nat Med 15, 1170–1178, (2009). [PubMed:
19767732]
17. Tesniere A et al. Immunogenic death of colon cancer cells treated with oxaliplatin. Oncogene 29,
482–491, (2010). [PubMed: 19881547]
18. Mortenson ED, Park S, Jiang Z, Wang S & Fu YX Effective anti-neu-initiated antitumor responses
require the complex role of CD4+ T cells. Clin Cancer Res 19, 1476–1486, (2013). [PubMed:
23363817]
19. Muller P et al. Trastuzumab emtansine (T-DM1) renders HER2+ breast cancer highly susceptible
to CTLA-4/PD-1 blockade. Sci Transl Med 7, 315ra188, (2015).
20. Janjigian YY et al. First-line pembrolizumab and trastuzumab in HER2-positive oesophageal,
gastric, or gastro-oesophageal junction cancer: an open-label, single-arm, phase 2 trial. Lancet
Author Manuscript

Oncol 21, 821–831, (2020). [PubMed: 32437664]

21. Rha SY et al. Targeting HER2 in combination with anti-PD-1 and chemotherapy confers a
significant tumor shrinkage of gastric cancer: A multi-institutional phase Ib/II trial of first-line
triplet regimen (pembrolizumab, trastuzumab, chemotherapy) for HER2-positive advanced gastric
cancer (AGC). J Clin Oncol 38 (suppl 15), abstract 3081 (2020).
22. Chung HC et al. First-line pembrolizumab/placebo plus trastuzumab and chemotherapy in HER2-
positive advanced gastric cancer: KEYNOTE-811. Future Oncol 17, 491–501 (2021). [PubMed:
33167735]

Nature. Author manuscript; available in PMC 2022 June 15.

 Janjigian et al. Page 16

23. Barrueto L et al. Resistance to checkpoint inhibition in cancer immunotherapy. Transl Oncol 13,
100738 (2020). [PubMed: 32114384]
Author Manuscript

24. Shitara K et al. Pembrolizumab versus paclitaxel for previously treated, advanced gastric or
gastro-oesophageal junction cancer (KEYNOTE-061): a randomised, open-label, controlled, phase
3 trial. Lancet 392, 123–133 (2018). [PubMed: 29880231]
25. Boku N et al. Nivolumab plus chemotherapy versus chemotherapy alone in patients with
previously untreated advanced or recurrent gastric/gastroesophageal junction (G/GEJ) cancer:
ATTRACTION-4 (ONO-4538–37) study. Ann Oncol 31, S1192–S1192 (2020).
26. Janjigian YY et al. First-line nivolumab plus chemotherapy versus chemotherapy alone for
advanced gastric, gastro-oesophageal junction, and oesophageal adenocarcinoma (CheckMate
649): a randomised, open-label, phase 3 trial. Lancet 398, 27–40, (2021). [PubMed: 34102137]
27. Chao J et al. Assessment of pembrolizumab therapy for the treatment of microsatellite
instability-high gastric or gastroesophageal junction cancer among patients in the
KEYNOTE-059, KEYNOTE-061, and KEYNOTE-062 clinical trials. JAMA Oncol, doi:10.1001/
jamaoncol.2021.0275 (2021).
28. Janjigian YY et al. Genetic predictors of response to systemic therapy in esophagogastric cancer.
Author Manuscript

Cancer Discov 8, 49–58 (2018). [PubMed: 29122777]

Methods References
29. Eisenhauer EA et al. New response evaluation criteria in solid tumours: revised RECIST guideline
(version 1.1). Eur J Cancer 45, 228–247 (2009). [PubMed: 19097774]
30. Oken MM et al. Toxicity and response criteria of the Eastern Cooperative Oncology Group. Am J
Clin Oncol 5, 649–655 (1982). [PubMed: 7165009]
31. Kulangara K et al. Clinical utility of the combined positive score for programmed death ligand-1
expression and the approval of pembrolizumab for treatment of gastric cancer. Arch Pathol Lab
Med 143, 330–337 (2019). [PubMed: 30028179]
32. Seymour L et al. iRECIST: guidelines for response criteria for use in trials testing
immunotherapeutics. Lancet Oncol 18, e143–e152 (2017). [PubMed: 28271869]
Author Manuscript
Author Manuscript

Nature. Author manuscript; available in PMC 2022 June 15.

 Janjigian et al. Page 17
Author Manuscript
Author Manuscript

Fig. 1. Best percentage change from baseline in the size of target lesions among participants in
Author Manuscript

the efficacy population.

(a) pembrolizumab group; (b) placebo group. Only those participants in the efficacy
population who had RECIST-measurable disease at baseline and at least one evaluable post-
baseline measurement are evaluable for change from baseline (N=124 in the pembrolizumab
group, N=122 in the chemotherapy group). The treatment regimen included trastuzumab and
chemotherapy in both groups. Increases from baseline greater than 100% were truncated at
100%.
Author Manuscript

Nature. Author manuscript; available in PMC 2022 June 15.

 Janjigian et al. Page 18

Table 1.

Summary of confirmed objective response in the efficacy population

Author Manuscript

Variable Pembrolizumab Group Placebo Group

(N=133) (N=131)

Objective response — % (95% CI)* 74.4 (66.2–81.6) 51.9 (43.0–60.7)

Disease control — % (95% CI)† 96.2 (91.4–98.8) 89.3 (82.7–94.0)

Best overall response — no. (%)

Complete response 15 (11.3) 4 (3.1)

Partial response 84 (63.2) 64 (48.9)

Stable disease 29 (21.8) 49 (37.4)

Progressive disease 5 (3.8) 7 (5.3)

Not evaluable‡ 0 (0.0) 2 (1.5)

Not assessed‡ 0 (0.0) 5 (3.8)

Author Manuscript

The treatment regimen included trastuzumab and chemotherapy in both groups.

*
Objective response was defined as the proportion of participants with a best response of complete or partial response. The estimated treatment
difference in objective response between the pembrolizumab and placebo groups was 22.7% (95% CI, 11.2–33.7; P=0.00006) and was calculated
using the Miettinen and Nurminen method stratified by geographic region (Australia/Europe/Israel/North America vs Asia vs rest of world), PD-L1
combined positive score (≥1 vs <1), and chemotherapy choice (5-fluorouracil plus cisplatin vs capecitabine plus oxaliplatin).
†
Disease control was defined as the proportion of participants with a best response of complete response, partial response, or stable disease.
‡
Participants who were not evaluable were those who had at least one postbaseline imaging assessment, none of which could be evaluated for
response according to RECIST, version 1.1. Participants who were not assessed included those who had no postbaseline imaging assessment.
Author Manuscript
Author Manuscript

Nature. Author manuscript; available in PMC 2022 June 15.

 Janjigian et al. Page 19

Table 2.

Summary of adverse events in the as-treated population

Author Manuscript

Event — no. (%) Any Cause Immune-Mediated Events and Infusion Reactions*

Pembrolizumab Group Placebo Group Pembrolizumab Group Placebo Group

(N=217) (N=216) (N=217) (N=216)
Any grade 211 (97.2) 212 (98.1) 73 (33.6) 45 (20.8)

Grade 3–5 124 (57.1) 124 (57.4) 21 (9.7) 7 (3.2)

Serious 68 (31.3) 83 (38.4) 19 (8.8) 7 (3.2)

Led to death 7 (3.2) 10 (4.6) 3 (1.4) 1 (0.5)

Led to discontinuation of any 53 (24.4) 56 (25.9) 12 (5.5) 5 (2.3)

drug

Pembrolizumab or placebo 12 (5.5) 15 (6.9) 7 (3.2) 2 (0.9)

Trastuzumab 12 (5.5) 16 (7.4) 7 (3.2) 2 (0.9)

Author Manuscript

Any chemotherapy 50 (23.0) 44 (20.4) 12 (5.5) 5 (2.3)

All drugs 8 (3.7) 12 (5.6) 6 (2.8) 2 (0.9)

The treatment regimen included trastuzumab and chemotherapy in both groups.

*
Events with a possible immune-mediated cause and infusion reactions were considered regardless of attribution to study treatment or immune
relatedness by the investigator and are based on a list of terms provided by the sponsor.
Author Manuscript
Author Manuscript

Nature. Author manuscript; available in PMC 2022 June 15.