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

Joseph Chao; Charles S Fuchs; Kohei Shitara; Josep Tabernero; Kei Muro; Eric Van Cutsem; Yung-Jue Bang; Ferdinando De Vita; Gregory Landers; Chia-Jui Yen; Ian Chau; Anneli Elme; Jeeyun Lee; Mustafa Özgüroğlu; Daniel Catenacci; Harry H Yoon; Erluo Chen; David Adelberg; Chie-Schin Shih; Sukrut Shah; Pooja Bhagia; Zev A Wainberg

doi:10.1001/jamaoncol.2021.0275

. 2021 Apr 1;7(6):895–902. doi: 10.1001/jamaoncol.2021.0275

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

Joseph Chao ^1,^✉, Charles S Fuchs ², Kohei Shitara ³, Josep Tabernero ⁴, Kei Muro ⁵, Eric Van Cutsem ⁶, Yung-Jue Bang ⁷, Ferdinando De Vita ⁸, Gregory Landers ⁹, Chia-Jui Yen ¹⁰, Ian Chau ¹¹, Anneli Elme ¹², Jeeyun Lee ¹³, Mustafa Özgüroğlu ¹⁴, Daniel Catenacci ¹⁵, Harry H Yoon ¹⁶, Erluo Chen ¹⁷, David Adelberg ¹⁷, Chie-Schin Shih ¹⁷, Sukrut Shah ¹⁷, Pooja Bhagia ¹⁷, Zev A Wainberg ¹⁸

¹Department of Medical Oncology and Therapeutics Research, City of Hope Comprehensive Cancer Center, Duarte, California

²Yale Cancer Center, Smilow Cancer Hospital, New Haven, Connecticut

³National Cancer Center Hospital East, Kashiwa, Japan

⁴Vall d’Hebron University Hospital and Institute of Oncology, Baselga Oncological Institute–Quiron, University of Vic–Central University of Catalonia, Barcelona, Spain

⁵Aichi Cancer Center Hospital, Nagoya, Japan

⁶University Hospital Leuven and KU Leuven, Leuven, Belgium

⁷Seoul National University College of Medicine, Seoul, Republic of Korea

⁸University of Study of Campania Luigi Vanvitelli, Naples, Italy

⁹The Oncology Centre, Durban, South Africa

¹⁰National Cheng Kung University Hospital, Taiwan, People’s Republic of China

¹¹Royal Marsden Hospital, Sutton, Surrey, United Kingdom

¹²North Estonia Medical Center Foundation, Tallinn, Estonia

¹³Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea

¹⁴Cerrahpaşa Medical Faculty, Cerrahpaşa School of Medicine, Istanbul University–Cerrahpaşa, Istanbul, Turkey

¹⁵Section of Hematology/Oncology, Department of Medicine, University of Chicago, Chicago, Illinois

¹⁶Mayo Clinic, Rochester, Minnesota

¹⁷Merck Sharp & Dohme, a subsidiary of Merck & Co, Kenilworth, New Jersey

¹⁸David Geffen School of Medicine, University of California, Los Angeles

Accepted for Publication: December 23, 2020.

Published Online: April 1, 2021. doi:10.1001/jamaoncol.2021.0275

^✉

Corresponding Author: Joseph Chao, MD, Department of Medical Oncology and Therapeutics Research, City of Hope Comprehensive Cancer Center, 1500 E Duarte Rd, Duarte, CA 91010 (jchao@coh.org).

Author Contributions: Dr Chao had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Chao, Fuchs, Shitara, Tabernero, Shah, Wainberg.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Chao, Shitara, Lee, Özgüroğlu, Wainberg.

Critical revision of the manuscript for important intellectual content: Chao, Fuchs, Shitara, Tabernero, Muro, Van Cutsem, Bang, De Vita, Landers, Yen, Chau, Elme, Özgüroğlu, Catenacci, Yoon, Chen, Adelberg, Shih, Shah, Bhagia, Wainberg.

Statistical analysis: Chen, Adelberg.

Administrative, technical, or material support: Tabernero, Muro, Van Cutsem, Bang, Chau, Shih, Wainberg.

Supervision: Chao, Fuchs, Tabernero, De Vita, Yen, Catenacci, Chen, Shih, Shah, Bhagia, Wainberg.

Other—patient recruitment: Van Cutsem.

Conflict of Interest Disclosures: Dr Chao reported receiving manuscript-writing assistance from Merck Sharp & Dohme during the conduct of the study and receiving grants from Brooklyn ImmunoTherapeutics and Merck and personal fees from Amgen, AstraZeneca, Boston Biomedical, Daiichi Sankyo, Foundation Medicine, MacroGenics, Merck, Ono Pharmaceutical, and Taiho Pharmaceutical outside the submitted work. Dr Fuchs reported receiving personal fees from Agios Pharmaceuticals, Amylin Pharmaceuticals, Bain Capital, CytomX Therapeutics, Daiichi Sankyo, Eli Lilly, Entrinsic Health, EvolveImmune Therapeutics, Genentech, Merck, Taiho Pharmaceutical, and Unum Therapeutics; owning stock in CytomX Therapeutics and Entrinsic Health; cofounding EvolveImmune Therapeutics; serving as the director of CytomX Therapeutics and EvolveImmune Therapeutics; and providing expert testimony for Amylin Pharmaceuticals and Eli Lilly outside the submitted work. Dr Shitara reported receiving grants from Astellas Pharma, Chugai Pharmaceutical, Daiichi Sankyo, Eli Lilly, Merck, Medi Science, Ono Pharmaceutical, Sumitomo Dainippon Pharma, and Taiho Pharmaceutical and personal fees from AbbVie, Astellas Pharma, Bristol Myers Squibb, Eli Lilly, GlaxoSmithKline, Novartis, Pfizer, Takeda Pharmaceutical, and Yakult Honsha outside the submitted work. Dr Tabernero reported receiving personal fees from Array BioPharma, AstraZeneca, Bayer, BeiGene, Biocartis, Boehringer Ingelheim, Chugai Pharmaceutical, Eli Lilly, F. Hoffmann-La Roche, Foundation Medicine, Genentech, Genmab, HalioDx, Halozyme Therapeutics, Imugene, Inflection Biosciences, Ipsen Biopharmaceuticals, Kura Oncology, Menarini, Merck Serono, Merck Sharp & Dohme, Merrimack Pharmaceuticals, Merus, Molecular Partners, Novartis, Peptomyc, Pfizer, Pharmacyclics, ProteoDesign, Rafael Pharmaceuticals, Roche Diagnostics, Sanofi, SeaGen (formerly Seattle Genetics), Servier Laboratories, Symphogen, Taiho Pharmaceutical, and VCN Biosciences outside the submitted work. Dr Muro reported receiving grants from Merck Sharp & Dohme during the conduct of the study and receiving grants from Amgen, Astellas Pharma, Daiichi Sankyo, Merck Sharp & Dohme, Merck Serono, Ono Pharmaceutical, Parexel, Pfizer, Sanofi, Solasia Pharma, and Taiho Pharmaceutical and personal fees from Amgen, AstraZeneca, Bayer, Bristol Myers Squibb, Chugai Pharmaceutical, Eli Lilly, Ono Pharmaceutical, Sanofi, Takeda Pharmaceutical, and Taiho Pharmaceutical outside the submitted work. Dr Van Cutsem reported receiving grants from Amgen, Bayer, Bristol Myers Squibb, Boehringer Ingelheim, Celgene, Eli Lilly, Ipsen Biopharmaceuticals, Merck KGaA, Merck Sharp & Dohme, Novartis, Roche, and Servier Laboratories and serving on the advisory boards of Array BioPharma, AstraZeneca, Bayer, Biocartis, Bristol Myers Squibb, Celgene, Daiichi Sankyo, Eli Lilly, GlaxoSmithKline, Halozyme Therapeutics, Incyte, Ipsen Biopharmaceuticals, Merck KGaA, Merck Sharp & Dohme, Novartis, Pierre Fabre, Roche, Servier Laboratories, Sirtex Medical, and Taiho Pharmaceutical outside the submitted work. Dr Bang reported receiving grants from Astellas Pharma, AstraZeneca, Bayer, BeiGene, Boehringer Ingelheim, Boston Biomedical, Bristol Myers Squibb, CKD Pharmaceuticals, Curis, Daiichi Sankyo, Eli Lilly, Five Prime Therapeutics, Genentech, Genexine, GlaxoSmithKline, GC Pharma, MacroGenics, Merck Serono, Merck Sharp & Dohme, Novartis, Ono Pharmaceutical, Pfizer, Taiho Pharmaceutical, and Takeda Pharmaceutical and serving as a consultant or advisor for Astellas Pharma, AstraZeneca, Bayer, BeiGene, Bristol Myers Squibb, Daiichi Sankyo, Eli Lilly, Genentech, Genexine, GC Pharma, Hanmi Pharmaceutical, Merck Serono, Merck Sharp & Dohme, Novartis, Samyang Biopharm, and Taiho Pharmaceutical outside the submitted work. Dr De Vita reported serving as a consultant or advisor for Celgene and Eli Lilly outside the submitted work. Dr Chau reported receiving grants from Merck Sharp & Dohme during the conduct of the study and receiving grants from Eli Lilly, Janssen-Cilag, and Sanofi Oncology and personal fees from AstraZeneca, Bayer, Bristol Myers Squibb, Eli Lilly, Five Prime Therapeutics, Merck Serono, Merck Sharp & Dohme, Oncologie, Pierre Fabre, and Roche outside the submitted work. Dr Elme reported receiving grants from Merck Sharp & Dohme during the conduct of the study and receiving grants from Roche and personal fees from Amgen, Astra Zeneca, Ipsen Biopharmaceuticals, Merck Sharp & Dohme, and Roche outside the submitted work. Dr Özgüroğlu reported receiving grants from Merck Sharp & Dohme during the conduct of the study and receiving personal fees from Astellas Pharma, AstraZeneca, Bristol Myers Squibb, Janssen Pharmaceuticals, Novartis, Roche, and Sanofi outside the submitted work. Dr Catenacci reported receiving grants from Merck Sharp & Dohme and personal fees from Astellas Pharma, Bristol Myers Squibb, Daiichi Sankyo, Eli Lilly, Five Prime Therapeutics, Foundation Medicine, Genentech, Gritstone Oncology, Guardant Health, Merck, Pieris Pharmaceuticals, Taiho Pharmaceutical, and Tempus Labs during the conduct of the study. Dr Yoon reported receiving grants from Merck and personal fees from BeiGene, Bristol Myers Squibb, and MacroGenics outside the submitted work. Dr Wainberg reported receiving grants from Bristol Myers Squibb, Five Prime Therapeutics, Merck Serono, Novartis, and Ipsen Biopharmaceuticals and personal fees from AstraZeneca, Bayer, Daiichi Sankyo, Eli Lilly, MacroGenics, and Merck outside the submitted work. No other disclosures were reported.

Funding/Support: This study and assistance with medical writing were funded by Merck Sharp & Dohme, a subsidiary of Merck, and supported by grant 5K12CA001727-23 from the National Institutes of Health (Dr Chao).

Role of the Funder/Sponsor: Employees of Merck Sharp & Dohme were involved in the design and conduct of the study and in the collection, management, analysis, and interpretation of the data. Drs Chen, Adelberg, Shih, Shah, and Bhagia, employees of Merck, were involved in the review and approval of the manuscript and the decision to submit the manuscript for publication.

Disclaimer: The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Additional Contributions: Jonathan Cheng, MD, an employee of Merck Sharp & Dohme, provided critical review. Luana Atherly-Henderson, PhD, CMPP, an employee of Merck Sharp & Dohme, provided medical writing assistance. We thank the patients and their families and caregivers along with all of the primary investigators and their site personnel.

^✉

Corresponding author.

PMCID: PMC8017478 PMID: 33792646

Key Points

Question

Is treatment with pembrolizumab therapy vs chemotherapy associated with improvements in antitumor activity among patients who have advanced microsatellite instability–high gastric or gastroesophageal junction cancer regardless of the line of therapy in which it was received?

Findings

In this post hoc cohort study of 1614 patients in which 84 had confirmed microsatellite instability–high advanced gastric or gastroesophageal junction cancer, each of whom was enrolled in one of the KEYNOTE-059, KEYNOTE-061, or KEYNOTE-062 clinical trials, treatment with pembrolizumab therapy alone or in combination with chemotherapy was associated with prolonged overall and progression-free survival and provided durable responses vs chemotherapy alone among patients who had received 2 or more previous lines of therapy in KEYNOTE-059, 1 previous line of therapy in KEYNOTE-061, or no previous therapy in KEYNOTE-062.

Meaning

The study’s findings indicate that incorporation of pembrolizumab with or without chemotherapy may be more beneficial than chemotherapy alone for the treatment of patients who have microsatellite instability–high advanced or metastatic gastric or gastroesophageal junction cancer across all lines of therapy.

Abstract

Importance

Immunotherapy has been associated with improved outcomes among patients who have received previous treatment for microsatellite instability–high (MSI-H) tumors.

Objective

To evaluate the antitumor activity of pembrolizumab therapy vs chemotherapy among patients with MSI-H advanced gastric or gastroesophageal junction (G/GEJ) cancer regardless of the line of therapy in which it was received.

Design, Setting, and Participants

This post hoc analysis of the phase 2 KEYNOTE-059 (third-line treatment or higher) single-arm trial and the phase 3 KEYNOTE-061 (second-line treatment) and KEYNOTE-062 (first-line treatment) randomized trials included patients with advanced G/GEJ cancer from 52 sites in 16 countries enrolled in KEYNOTE-059, 148 sites in 30 countries enrolled in KEYNOTE-061, and 200 sites in 29 countries enrolled in KEYNOTE-062. Patients were enrolled from March 2, 2015, to March 26, 2016, in KEYNOTE-059; from June 4, 2015, to July 26, 2016, in KEYNOTE-061; and from September 18, 2015, to May 26, 2017, in KEYNOTE-062, with data cutoff dates of August 8, 2018; October 26, 2017; and March 26, 2019; respectively.

Interventions

Pembrolizumab monotherapy in KEYNOTE-059, pembrolizumab monotherapy or chemotherapy (paclitaxel) in KEYNOTE-061, and pembrolizumab monotherapy, pembrolizumab plus chemotherapy (cisplatin and 5-fluorouracil or capecitabine), or chemotherapy alone in KEYNOTE-062.

Main Outcomes and Measures

Response was assessed centrally using Response Evaluation Criteria in Solid Tumours (RECIST), version 1.1; MSI-H status was determined centrally by polymerase chain reaction testing.

Results

At data cutoff, 7 of 174 patients (4.0%) in KEYNOTE-059, 27 of 514 patients (5.3%) in KEYNOTE-061, and 50 of 682 patients (7.3%) in KEYNOTE-062 had MSI-H tumors. Among those with MSI-H tumors, the median overall survival was not reached (NR) for pembrolizumab in KEYNOTE-059, KEYNOTE-061, and KEYNOTE-062 or for pembrolizumab plus chemotherapy in KEYNOTE-062. The median progression-free survival (PFS) for pembrolizumab was NR (95% CI, 1.1 months to NR) in KEYNOTE-059 and 17.8 months (95% CI, 2.7 months to NR) in KEYNOTE-061 (vs 3.5 months [95% CI, 2.0-9.8 months] for chemotherapy). In KEYNOTE-062, the median PFS was 11.2 months (95% CI, 1.5 months to NR) for pembrolizumab, NR (95% CI, 3.6 months to NR) for pembrolizumab plus chemotherapy, and 6.6 months (95% CI, 4.4-8.3 months) for chemotherapy. The objective response rate (ORR) for pembrolizumab was 57.1% in KEYNOTE-059 and 46.7% (vs 16.7% for chemotherapy) in KEYNOTE-061. In KEYNOTE-062, the ORR was 57.1% for pembrolizumab , 64.7% for pembrolizumab plus chemotherapy, and 36.8% for chemotherapy.

Conclusions and Relevance

Findings from this study indicate that MSI-H status may be a biomarker for pembrolizumab therapy among patients with advanced G/GEJ cancer regardless of the line of therapy in which it was received.

Trial Registration

ClinicalTrials.gov Identifiers: NCT02335411, NCT02370498, and NCT02494583

This cohort study assesses the antitumor activity of pembrolizumab therapy vs chemotherapy for the treatment of microsatellite instability–high advanced gastric or gastroesophageal junction cancer among patients enrolled in the KEYNOTE-059, KEYNOTE-061, and KEYNOTE-062 clinical trials.

Introduction

The phenotype for microsatellite instability–high (MSI-H) is a distinct tumor subclass that is highly susceptible to immunotherapy.¹ These tumors typically have high tumor mutational burden and high levels of programmed cell death 1 ligand 1 (PD-L1) expression, leading to improved response to anti–programmed cell death 1 protein (anti–PD-1) therapy.¹ Approximately 5% to 20% of gastric tumors are MSI-H,² warranting further understanding of the benefits of anti–PD-1 therapy across treatment lines in this subgroup. The anti–PD-1 monoclonal antibody pembrolizumab has provided durable antitumor immunity in patients with advanced gastric or gastroesophageal junction cancer in the phase 2 KEYNOTE-059 clinical trial (1 arm; third-line treatment or higher),³ the phase 3 KEYNOTE-061 randomized clinical trial (2 arms; second-line treatment),⁴ and the phase 3 KEYNOTE-062 randomized clinical trial (3 arms; first-line treatment).⁵ In this post hoc analysis, we examined the antitumor activity of pembrolizumab therapy among patients with MSI-H advanced gastric or gastroesophageal junction cancer who were enrolled in these studies.

Methods

Study Design, Treatment, and Participants

This post hoc analysis of the phase 2 KEYNOTE-059 and phase 3 KEYNOTE-061 and KEYNOTE-062 clinical trials included patients with advanced gastric or gastroesophageal junction cancer from 52 sites in 16 countries enrolled in KEYNOTE-059, 148 sites in 30 countries enrolled in KEYNOTE-061, and 200 sites in 29 countries enrolled in KEYNOTE-062. Institutional review board approval was obtained from all study centers, and all patients provided written informed consent.

Eligibility criteria and study treatment for the KEYNOTE-059, KEYNOTE-061, and KEYNOTE-062 clinical trials were previously reported.^3,4,5 In brief, patients in the KEYNOTE-059, KEYNOTE-061, and KEYNOTE-062 studies had received 2 or more previous lines of therapy, 1 previous line of therapy, or no previous therapy, respectively. With regard to study treatment, patients in the KEYNOTE-059 study (third-line treatment or higher) received pembrolizumab monotherapy (200 mg every 3 weeks for up to 2 years), patients in the KEYNOTE-061 study (second-line treatment) received pembrolizumab monotherapy or chemotherapy (paclitaxel) alone, and patients in the KEYNOTE-062 study (first-line treatment) received pembrolizumab monotherapy, pembrolizumab plus chemotherapy (cisplatin and 5-fluorouracil or capecitabine), or chemotherapy alone. Patients received treatment until disease progression or unacceptable toxic effects. Evaluations of study objectives, including overall survival (OS), progression-free survival (PFS), objective response rate (ORR), and duration of response, were previously reported.^3,4,5

Biomarker Analysis

Tumor PD-L1 expression was evaluated in pretreatment tissue samples through the PD-L1 IHC 22C3 pharmDx assay (Agilent) using the combined positive score. Tumors that were PD-L1 positive had combined positive scores of 1 or higher. Microsatellite instability–high status was determined centrally through polymerase chain reaction testing using the MSI analysis system, version 1.2 (Promega).

Response was assessed centrally using Response Evaluation Criteria in Solid Tumours (RECIST), version 1.1. Patients with evaluable tumors had available matching tissue and blood samples that met the sufficiency requirements of the MSI analysis. Tumors were considered to be MSI-H if 2 or more markers were changed compared with normal controls.

Statistical Analysis

All patients who received 1 or more doses of the study treatment (KEYNOTE-059) and all randomized patients (KEYNOTE-061 and KEYNOTE-062) were included in the analysis. Patients were enrolled from March 2, 2015, to March 26, 2016, in KEYNOTE-059; from June 4, 2015, to July 26, 2016, in KEYNOTE-061; and from September 18, 2015, to May 26, 2017, in KEYNOTE-062. Data cutoff dates were August 8, 2018, for KEYNOTE-059; October 26, 2017, for KEYNOTE-061; and March 26, 2019, for KEYNOTE-062. Kaplan-Meier estimates were provided for OS, PFS, and duration of response. For ORR, point estimates and 95% CIs were provided using the exact binomial Clopper-Pearson method. All statistical analyses were performed using SAS software, version 9.4 (SAS Institute), and full descriptions of analytic methods were previously published.^3,4,5

Results

A total of 259 patients (median age, 62 years [range, 24-89 years]; 198 men [76.4%]) were enrolled in KEYNOTE-059, 592 patients (median age, 61 years [range, 20-87 years]; 410 men [69.3%]) were enrolled in KEYNOTE-061, and 763 patients (median age, 62 years [range, 20-87 years]; 554 men [72.6%]) were enrolled in KEYNOTE-062 (eTable 1 in the Supplement). Overall, 7 of 174 patients (4.0%) in KEYNOTE-059, 27 of 514 patients (5.3%) in KEYNOTE-061, and 50 of 682 patients (7.3%) in KEYNOTE-062 with evaluable tumors had MSI-H gastric or gastroesophageal junction cancer (Table 1). At data cutoff, the median follow-up period was 5.6 months (range, 0.5-37.6 months) for KEYNOTE-059, 7.9 months (range, 0.2-27.7 months) for KEYNOTE-061, and 11.3 months (range, 0.2-41.2 months) for KEYNOTE-062.

Table 1. Baseline Characteristics of Patients With Microsatellite Instability–High Advanced Gastric or Gastroesophageal Junction Cancer by Line of Therapy.

Characteristic	Patients, No. (%)
	KEYNOTE-059^a	KEYNOTE-061^b		KEYNOTE-062^c
	Pembrolizumab	Pembrolizumab	Chemotherapy	Pembrolizumab	Pembrolizumab plus chemotherapy	Chemotherapy
Total patients, No.	7	15	12	14	17	19
Age, median (range), y	62 (51-89)	67 (36-76)	63 (43-75)	62 (39-74)	67 (59-83)	69 (31-85)
Male sex	6 (85.7)	7 (46.7)	8 (66.7)	7 (50.0)	13 (76.5)	13 (68.4)
Location
Australia, Europe, and North America	2 (28.6)	10 (66.7)	7 (58.3)	10 (71.4)	11 (64.7)	14 (73.7)
Asia	1 (14.3)	4 (26.7)	3 (25.0)	4 (28.6)	4 (23.5)	2 (10.5)
Other region	4 (57.1)	1 (6.7)	2 (16.7)	0	2 (11.8)	3 (15.8)
ECOG performance status
0	4 (57.1)	5 (33.3)	4 (33.3)	7 (50.0)	7 (41.2)	7 (36.8)
1	3 (42.9)	10 (66.7)	8 (66.7)	7 (50.0)	10 (58.8)	12 (63.2)
Cancer type
Metastatic	6 (85.7)	14 (93.3)	11 (91.7)	13 (92.9)	15 (88.2)	16 (84.2)
Stomach	5 (71.4)	11 (73.3)	10 (83.3)	13 (92.9)	13 (76.5)	16 (84.2)
Gastroesophageal junction adenocarcinoma	2 (28.6)	4 (26.7)	2 (16.7)	1 (7.1)	4 (23.5)	3 (15.8)
Diffuse subtype adenocarcinoma	3 (42.9)	4 (26.7)	2 (16.7)	4 (28.6)	9 (52.9)	6 (31.6)
Intestinal subtype adenocarcinoma	3 (42.9)	0	4 (33.3)	8 (57.1)	5 (29.4)	7 (36.8)
Mixed/unknown subtype	1 (14.3)	11 (73.3)	6 (50.0)	2 (14.3)	3 (17.6)	6 (31.6)
Previous lines of therapy
0	0	0	0	14 (100)	17 (100)	19 (100)
1	0	15 (100)	12 (100)	0	0	0
≥2	7 (100)	0	0	0	0	0
PD-L1 CPS
≥1	5 (71.4)	13 (86.7)	11 (91.7)	14 (100)	17 (100)	19 (100)
≥10	2 (28.6)	8 (53.3)	5 (41.7)	11 (78.6)	11 (64.7)	10 (52.6)

Open in a new tab

Abbreviations: CPS, combined positive score; ECOG, Eastern Cooperative Oncology Group; PD-L1, programmed cell death 1 ligand 1.

^{^a}

A total of 174 of 259 patients enrolled in KEYNOTE-059 who had tumors that were evaluable for microsatellite instability–high status were included.

^{^b}

A total of 514 of 592 patients enrolled in KEYNOTE-061 who had tumors that were evaluable for microsatellite instability–high status were included.

^{^c}

A total of 682 of 763 patients enrolled in KEYNOTE-062 who had tumors that were evaluable for microsatellite instability–high status were included.

Among patients in KEYNOTE-059 (with updated data cutoff date from the original publication), no patients were receiving ongoing therapy, and 3 of 7 patients (42.9%) had completed 2 years of pembrolizumab monotherapy (eTable 2 in the Supplement). Among those in KEYNOTE-061, a larger proportion of patients who received pembrolizumab monotherapy had completed therapy (3 of 15 patients [20%]) or were receiving ongoing therapy (4 of 15 patients [26.7%]) compared with patients who received chemotherapy alone (0 patients in both categories). In KEYNOTE-062, a larger proportion of patients who received pembrolizumab monotherapy or pembrolizumab plus chemotherapy had completed therapy (6 of 14 patients [42.9%] and 6 of 17 patients [35.3%], respectively) or were receiving ongoing therapy (1 of 14 patients [7.1%] and 2 of 17 patients [11.8%], respectively) compared with patients who received chemotherapy alone (1 of 19 patients [5.3%] completed therapy, and 1 of 19 patients [5.3%] were receiving ongoing therapy) (eTable 2 in the Supplement).

Data from the period after discontinuation of clinical trial anticancer therapy among patients with MSI-H tumors are summarized in eTable 3 in the Supplement. A small proportion of patients were receiving post–clinical trial immunotherapy after disease progression.

Among patients with MSI-H tumors, the median OS for pembrolizumab monotherapy was not reached (ie, >50% of patients were still alive at data cutoff) in KEYNOTE-059 (95% CI, 1.1 months to not reached) or KEYNOTE-061 (95% CI, 5.6 months to not reached) compared with a median OS of 8.1 months (95% CI, 2.0-16.7 months) for chemotherapy alone in KEYNOTE-061. In KEYNOTE-062, the median OS was not reached for both pembrolizumab monotherapy (95% CI, 10.7 months to not reached) and pembrolizumab plus chemotherapy (95% CI, 3.6 months to not reached) compared with a median OS of 8.5 months (95% CI, 5.3-20.8 months) for chemotherapy alone (Table 2; Figure; eFigure 1 in the Supplement).

Table 2. Outcomes in Patients With Advanced Gastric or Gastroesophageal Junction Cancer by Line of Therapy.

Outcome	KEYNOTE-059^a	KEYNOTE-061^b		KEYNOTE-062^c
Outcome	Pembrolizumab	Pembrolizumab	Chemotherapy	Pembrolizumab	Pembrolizumab plus chemotherapy	Chemotherapy
All patients
Total patients, No.	259	296	296	256	257	250
Objective response rate, % (95% CI)	11.6 (8.0-16.1)	11.1 (7.8-15.3)	12.5 (9.0-16.8)	14.8 (10.7-19.8)	48.6 (42.4-54.9)	37.2 (31.2-43.5)
Best overall response rate, %
Complete	3.5	2.4	2.4	3.5	7.4	5.6
Partial	8.1	8.8	10.1	11.3	41.2	31.6
Stable disease	16.2	22.6	46.3	27.0	28.8	41.6
Progressive disease	56.0	51.7	22.3	41.0	7.4	8.4
Duration of response, median (range), mo	16.1 (2.4-34.9^d)	18.0 (1.4^d-26.0^d)	5.5 (1.3^d-17.7)	13.7 (1.4^d-33.6^d)	6.8 (1.4^d-34.7^d)	6.8 (1.4^d-30.4^d)
Survival, median (95% CI), mo
Progression-free	2.0 (2.0-2.0)	1.5 (1.4-1.6)	4.1 (3.2-4.2)	2.0 (1.5-2.8)	6.9 (5.8-7.3)	6.4 (5.7-7.1)
Overall	5.5 (4.2-6.7)	6.7 (5.4-8.9)	8.3 (7.7-8.8)	10.6 (7.7-13.8)	12.5 (10.8-13.9)	11.1 (9.2-12.8)
Estimated overall survival rate, % (95% CI)
12 mo	25 (NA)	34 (29-39)	28 (23-33)	47 (41-53)	53 (47-59)	46 (39-52)
24 mo	13 (NA)	18 (13-23)	9 (6-13)	27 (21-32)	24 (19-30)	19 (15-24)
Patients with MSI-H tumors
Total patients, No.	7	15	12	14	17	19
Objective response rate, % (95% CI)	57.1 (18.4-90.1)	46.7 (21.3-73.4)	16.7 (2.1-48.4)	57.1 (28.9-82.3)	64.7 (38.3-85.8)	36.8 (16.3-61.6)
Best overall response rate, %
Complete	28.6	6.7	8.3	7.1	35.3	10.5
Partial	28.6	40.0	8.3	50.0	29.4	26.3
Stable disease	14.3	40.0	58.3	21.4	17.6	42.1
Progressive disease	0	6.7	0	14.3	0	10.5
Duration of response, median (range), mo	NR (20.0^d-26.8^d)	NR (5.5-26.0^d)	NR (2.2^d-12.2^d)	21.2 (1.4^d-33.6^d)	NR (1.6^d-34.5^d)	7.0 (2.0-30.4^d)
Survival, median (95% CI), mo
Progression-free	NR (1.1-NR)	17.8 (2.7-NR)	3.5 (2.0-9.8)	11.2 (1.5-NR)	NR (3.6-NR)	6.6 (4.4-8.3)
Overall	NR (1.1-NR)	NR (5.6-NR)	8.1 (2.0-16.7)	NR (10.7-NR)	NR (3.6-NR)	8.5 (5.3-20.8)
Estimated overall survival rate, % (95% CI)
12 mo	71 (NA)	73 (44-89)	25 (6-50)	79 (47-92)	71 (43-87)	47 (24-67)
24 mo	57 (NA)	59 (31-79)	NA	71 (41-88)	65 (38-82)	26 (10-57)

Open in a new tab

Abbreviations: MSI-H, microsatellite instability–high; NA, not available; NR, not reached.

^{^a}

Patients enrolled in KEYNOTE-059 had 2 or more previous lines of therapy.

^{^b}

Patients enrolled in KEYNOTE-061 had 1 previous line of therapy.

^{^c}

Patients enrolled in KEYNOTE-062 had no previous therapy.

^{^d}

Indicates ongoing response.

Figure. — Combination indicates pembrolizumab plus chemotherapy; MSI-H, microsatellite instability–high. A, All patients in KEYNOTE-061 clinical trial. B, Patients with MSI-H tumors in KEYNOTE-061 clinical trial. C, All patients in KEYNOTE-062 clinical trial. D, Patients with MSI-H tumors in KEYNOTE-062 clinical trial. The tick marks on each survival curve indicate censored patients.

The estimated 12-month OS rates for pembrolizumab monotherapy among patients with MSI-H tumors were 71% (95% CI, not available) for KEYNOTE-059 and 73% (95% CI, 44%-89%) for KEYNOTE-061 (compared with 25% [95% CI, 6%-50%] for chemotherapy alone in KEYNOTE-061). In KEYNOTE-062, the estimated 12-month OS rates were 79% (95% CI, 47%-92%) for pembrolizumab monotherapy, 71% (95% CI, 43%-87%) for pembrolizumab plus chemotherapy, and 47% (95% CI, 24%-67%) for chemotherapy alone. In KEYNOTE-059 and KEYNOTE-061, the estimated 24-month OS rates for pembrolizumab monotherapy were 57% (95% CI, not available) and 59% (95% CI, 31%-79%), respectively (24-month OS rate not available for chemotherapy alone in KEYNOTE-061). In KEYNOTE-062, the estimated 24-month OS rates were 71% (95% CI, 41%-88%) for pembrolizumab monotherapy, 65% (95% CI, 38%-82%) for pembrolizumab plus chemotherapy, and 26% (95% CI, 10%-57%) for chemotherapy alone.

The median PFS among patients with MSI-H tumors was not reached (95% CI, 1.1 months to not reached) for pembrolizumab monotherapy in KEYNOTE-059 compared with 17.8 months (95% CI, 2.7 months to not reached) for pembrolizumab monotherapy and 3.5 months (95% CI, 2.0-9.8 months) for chemotherapy alone in KEYNOTE-061. In KEYNOTE-062, the median PFS was 11.2 months (95% CI, 1.5 months to not reached) for pembrolizumab monotherapy, not reached (95% CI, 3.6 months to not reached) for pembrolizumab plus chemotherapy, and 6.6 months (95% CI, 4.4-8.3 months) for chemotherapy alone (Table 2; eFigure 1 and eFigure 2 in the Supplement).

Among patients with MSI-H gastric or gastroesophageal junction cancer, the ORR for pembrolizumab monotherapy in KEYNOTE-059 was 57.1% (95% CI, 18.4%-90.1%) (Table 2). In KEYNOTE-061, the ORR was 46.7% (95% CI, 21.3%-73.4%) for pembrolizumab monotherapy vs 16.7% (95% CI, 2.1%-48.4%) for chemotherapy alone. In KEYNOTE-062, the ORR was 57.1% (95% CI, 28.9%-82.3%) for pembrolizumab monotherapy, 64.7% (95% CI, 38.3%-85.8%) for pembrolizumab plus chemotherapy, and 36.8% (95% CI, 16.3%-61.6%) for chemotherapy alone. The median duration of response was not reached for pembrolizumab monotherapy in both KEYNOTE-059 (range, 20.0-26.8 months) and KEYNOTE-061 (range, 5.5-26.0 months) and not reached for chemotherapy alone (range, 2.2-12.2 months) in KEYNOTE-061. In KEYNOTE-062, the median duration of response was 21.2 months (range, 1.4+ to 33.6 months, with + indicating no progressive disease at last assessment) for pembrolizumab monotherapy, not reached (range, 1.6+ to 34.5+ months) for pembrolizumab plus chemotherapy, and 7.0 months (range, 2.0-30.4+ months) for chemotherapy alone.

Outcomes in patients with non–MSI-H tumors or nonevaluable MSI-H tumors (ie, the confirmed cases with MSI-H tumors were removed) are summarized in eTable 4 in the Supplement. The distribution of baseline characteristics was similar across these patients (eTable 5 in the Supplement), patients with MSI-H tumors (Table 1), and patients with nonevaluable MSI-H tumors (eTable 6 in the Supplement).

Discussion

The results of this post hoc analysis indicate that pembrolizumab or pembrolizumab plus chemotherapy provided durable antitumor activity vs chemotherapy alone among patients with MSI-H gastric or gastroesophageal junction cancer regardless of the line of therapy in which it was received. This finding is exemplified by the greater number of patients with MSI-H cancer who were receiving ongoing therapy or had completed 2 years of pembrolizumab therapy. These data are consistent with those reported in previous studies of patients with MSI-H tumors,^6,7,8 and they illustrate the benefits of pembrolizumab therapy in patients with MSI-H gastric or gastroesophageal junction cancer across all lines of therapy. To our knowledge, the 31 patients with MSI-H tumors in the KEYNOTE-062 clinical trial who received pembrolizumab monotherapy or pembrolizumab plus chemotherapy represent the largest international gastric cancer cohort receiving first-line immune checkpoint inhibitor therapy reported to date.

The worse OS outcomes observed in the first-line setting among patients with MSI-H gastric or gastroesophageal junction cancer who received chemotherapy are consistent with other case series suggesting that patients with MSI-H gastric cancer respond worse to chemotherapy alone.^9,10,11,12 The early divergence of the survival curves between patients with MSI-H tumors who received first-line pembrolizumab therapy vs chemotherapy suggests that earlier introduction of pembrolizumab is beneficial in patients with this molecular subtype of gastric or gastroesophageal junction cancer. Primary resistance to immune checkpoint inhibitors in patients with MSI-H advanced gastric or gastroesophageal junction cancer remains an active area of investigation, and data have suggested that intratumoral heterogeneous loss of mismatch repair enzymes may account for reduced response to pembrolizumab monotherapy.¹³ In the KEYNOTE-062 study, 12- and 24-month OS rates were favorable for both the pembrolizumab and pembrolizumab plus chemotherapy arms among patients with MSI-H tumors, and PFS numerically favored the combination of therapies; however, this conclusion remains hypothesis generating. The recent phase 3 KEYNOTE-177 study of patients with MSI-H metastatic colorectal cancer reported significant improvements in PFS with first-line pembrolizumab therapy vs chemotherapy alone.¹⁴ In addition, the phase 3 CheckMate 649 study, which included a subgroup of patients with MSI-H gastric or gastroesophageal adenocarcinoma, reported improved OS with first-line nivolumab therapy plus chemotherapy vs chemotherapy alone, further supporting the robustness of MSI-H status to estimate benefits for immune checkpoint blockade.¹⁵ These data support routine testing for MSI-H status in patients with advanced gastric cancer. The addition of pembrolizumab to first-line chemotherapy in patients with MSI-H gastric or gastroesophageal junction cancer is being validated in the ongoing phase 3 KEYNOTE-859 clinical trial (NCT03675737).

Limitations

This study has limitations. The small numbers of patients with MSI-H tumors limited statistical interpretation, such as the performance of multivariable analysis. In addition, MSI-H status was not available for all patients, which may have introduced selection bias. However, the distribution of baseline characteristics among patients with nonevaluable tumors was similar to that of the overall study population.

Conclusions

The findings of this analysis support MSI-H status as a biomarker for pembrolizumab therapy among patients with advanced gastric or gastroesophageal junction cancer and warrant its prospective validation in ongoing first-line studies.

Supplement.

eTable 1. Baseline Characteristics of Patients with Advanced Gastric or Gastroesophageal Junction Cancer by Line of Therapy in KEYNOTE-059, KEYNOTE-061, KEYNOTE-062

eTable 2. Treatment Disposition in Patients with MSI-H Advanced Gastric or Gastroesophageal Junction Cancer by Line of Therapy in KEYNOTE-059, KEYNOTE-061, KEYNOTE-062

eTable 3. Postdiscontinuation Anticancer Therapy in Patients with MSI-H Advanced Gastric or Gastroesophageal Junction Cancer by Line of Therapy in KEYNOTE-061 and KEYNOTE-062

eTable 4. Efficacy Outcomes in Patients With Non–MSI-H or MSI-H Nonevaluable Advanced Gastric or Gastroesophageal Junction Cancer by Line of Therapy in KEYNOTE-059, KEYNOTE-061, KEYNOTE-062

eTable 5. Baseline Characteristics of Patients With Non–MSI-H or MSI-H Nonevaluable Advanced Gastric or Gastroesophageal Junction Cancer by Line of Therapy in KEYNOTE-059, KEYNOTE-061, KEYNOTE-062

eTable 6. Baseline Characteristics of Patients With Nonevaluable MSI-H Status

eFigure 1. Kaplan-Meier Estimates of Overall and Progression-Free Survival in Patients With Advanced Gastric or Gastroesophageal Junction Cancer in KEYNOTE-059

eFigure 2. Kaplan-Meier Estimates of Progression-Free Survival in Patients With Advanced Gastric or Gastroesophageal Junction Cancer by Line of Therapy in KEYNOTE-061 and KEYNOTE-062

Click here for additional data file.^{(386.2KB, pdf)}

References

1.Yoshino T, Pentheroudakis G, Mishima S, et al. JSCO-ESMO-ASCO-JSMO-TOS: international expert consensus recommendations for tumour-agnostic treatments in patients with solid tumours with microsatellite instability or NTRK fusions. Ann Oncol. 2020;31(7):861-872. doi: 10.1016/j.annonc.2020.03.299 [DOI] [PubMed] [Google Scholar]
2.Hause RJ, Pritchard CC, Shendure J, Salipante SJ. Classification and characterization of microsatellite instability across 18 cancer types. Nat Med. 2016;22(11):1342-1350. doi: 10.1038/nm.4191 [DOI] [PubMed] [Google Scholar]
3.Fuchs CS, Doi T, Jang RW, et al. Safety and efficacy of pembrolizumab monotherapy in patients with previously treated advanced gastric and gastroesophageal junction cancer: phase 2 clinical KEYNOTE-059 trial. JAMA Oncol. 2018;4(5):e180013. doi: 10.1001/jamaoncol.2018.0013 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Shitara K, Ozguroglu M, Bang YJ, et al. ; KEYNOTE-061 Investigators . Pembrolizumab versus paclitaxel for previously treated, advanced gastric or gastro-oesophageal junction cancer (KEYNOTE-061): a randomised, open-label, controlled, phase 3 trial. Lancet. 2018;392(10142):123-133. doi: 10.1016/S0140-6736(18)31257-1 [DOI] [PubMed] [Google Scholar]
5.Tabernero J, Van Cutsem E, Bang YJ, et al. Pembrolizumab with or without chemotherapy versus chemotherapy for advanced gastric or gastroesophageal junction (G/GEJ) adenocarcinoma: the phase III KEYNOTE-062 study. J Clin Oncol. 2019;37(18 suppl):LBA4007. [Google Scholar]
6.Le DT, Durham JN, Smith KN, et al. Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade. Science. 2017;357(6349):409-413. doi: 10.1126/science.aan6733 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Le DT, Kim TW, Van Cutsem E, et al. Phase II open-label study of pembrolizumab in treatment-refractory, microsatellite instability-high/mismatch repair-deficient metastatic colorectal cancer: KEYNOTE-164. J Clin Oncol. 2020;38(1):11-19. doi: 10.1200/JCO.19.02107 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Marabelle A, Le DT, Ascierto PA, et al. Efficacy of pembrolizumab in patients with noncolorectal high microsatellite instability/mismatch repair-deficient cancer: results from the phase II KEYNOTE-158 study. J Clin Oncol. 2020;38(1):1-10. doi: 10.1200/JCO.19.02105 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Janjigian YY, Sanchez-Vega F, Jonsson P, et al. Genetic predictors of response to systemic therapy in esophagogastric cancer. Cancer Discov. 2018;8(1):49-58. doi: 10.1158/2159-8290.CD-17-0787 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Smyth EC, Wotherspoon A, Peckitt C, et al. Mismatch repair deficiency, microsatellite instability, and survival: an exploratory analysis of the Medical Research Council Adjuvant Gastric Infusional Chemotherapy (MAGIC) trial. JAMA Oncol. 2017;3(9):1197-1203. doi: 10.1001/jamaoncol.2016.6762 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Kim SY, Choi YY, An JY, et al. The benefit of microsatellite instability is attenuated by chemotherapy in stage II and stage III gastric cancer: results from a large cohort with subgroup analyses. Int J Cancer. 2015;137(4):819-825. doi: 10.1002/ijc.29449 [DOI] [PubMed] [Google Scholar]
12.Kubota Y, Kawazoe A, Sasaki A, et al. The impact of molecular subtype on efficacy of chemotherapy and checkpoint inhibition in advanced gastric cancer. Clin Cancer Res. 2020;26(14):3784-3790. doi: 10.1158/1078-0432.CCR-20-0075 [DOI] [PubMed] [Google Scholar]
13.Kim ST, Cristescu R, Bass AJ, et al. Comprehensive molecular characterization of clinical responses to PD-1 inhibition in metastatic gastric cancer. Nat Med. 2018;24(9):1449-1458. doi: 10.1038/s41591-018-0101-z [DOI] [PubMed] [Google Scholar]
14.Andre T, Shiu KK, Kim TW, et al. Pembrolizumab versus chemotherapy for microsatellite instability–high/mismatch repair deficient metastatic colorectal cancer: the phase 3 KEYNOTE-177 study. J Clin Oncol. 2020;38(18 suppl):LBA4. doi: 10.1200/JCO.2020.38.18_supple.LBA4 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Moehler M, Shitara K, Garrido M, et al. LBA6_PR nivolumab (nivo) plus chemotherapy (chemo) versus chemo as first-line (1L) treatment for advanced gastric cancer/gastroesophageal junction cancer (GC/GEJC)/esophageal adenocarcinoma (EAC): first results of the CheckMate 649 study. Ann Oncol. 2020;31(suppl 4):S1191. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials