Randomized, open-label, phase 2 study of andecaliximab plus nivolumab versus nivolumab alone in advanced gastric cancer identifies biomarkers associated with survival

Abstract

Background: Matrix metalloproteinase-9 (MMP9) selectively cleaves extracellular matrix proteins contributing to tumor growth and an immunosuppressive microenvironment. This study evaluated andecaliximab (ADX), an inhibitor of MMP9, in combination with nivolumab (NIVO), for the treatment of advanced gastric cancer.

Methods: Phase 2, open-label, randomized multicenter study evaluating the efficacy, safety, and pharmacodynamics of ADX+NIVO versus NIVO in patients with pretreated metastatic gastric or gastroesophageal junction (GEJ) adenocarcinoma. The primary endpoint was objective response rate (ORR). Secondary endpoints included progression-free survival (PFS), overall survival (OS), and adverse events (AEs). We explored the correlation of efficacy outcomes with biomarkers.

Results: 144 patients were randomized; 141 were treated: 81% white, 69% male, median age was 61 years in the ADX+NIVO group and 62 years in the NIVO-alone group. The ORR was 10% (95% CI 4 to 19) in the ADX+NIVO group and 7% (95% CI 2 to 16) in the NIVO-alone group (OR: 1.5 (95% CI 0.4 to 6.1; p=0.8)). There was no response or survival benefit associated with adding ADX. AE rates were comparable in both treatment groups; the most common AEs were fatigue, decreased appetite, nausea, and vomiting. Programmed cell death ligand 1, interferon-γ (IFN), and intratumoral CD8+ cell density were not associated with treatment response or survival. The gene signature most correlated with shorter survival was the epithelial-to-mesenchymal gene signature; high transforming growth factor (TGF)-β fibrosis score was negatively associated with OS (p=0.036). Gene expression analysis of baseline tumors comparing long-(1+ years) and short-term (<1 year) survivors showed that GRB7 was associated with survival beyond 1 year. Human epidermal growth factor receptor 2 (HER2)-positive disease was associated with significantly longer survival (p=0.0077). Median tumor mutation burden (TMB) was 2.01; patients with TMB ≥median had longer survival (p=0.0025) and improved PFS (p=0.016). Based on a model accounting for TMB, TGF-β fibrosis, and HER2, TMB was the main driver of survival in this patient population.

Conclusion: Combination of ADX+NIVO had a favorable safety profile but did not improve efficacy compared with NIVO alone in patients with pretreated metastatic gastric or GEJ adenocarcinoma. HER2 positivity, higher TMB or GRB7, and lower TGF-β were associated with improved outcomes.

Trial registration number: NCT02864381 or GS-US-296--2013.

Keywords: Biomarkers; Clinical Trials; Phase II as Topic; Tumor; antibodies; neoplasm.

Conflict of interest statement

Competing interests: MAS has received institutional funding from Bristol Meyers Squibb, Oncolys Biopharma, and Merck, DCu has received grants from MedImmune, Clovis, Eli Lilly, 4SC, Bayer, Celgene, Leap, and Roche, and has participated in a scientific advisory board for OVIBIO. J-PM has received honoraria from Merck Sharp & Dohme. J-PM, reports honoraria from MSD, EVC reports grants from Amgen, Bayer, Boehringer Ingelheim, Bristol-Meyers Squibb, Celgene, Ipsen, Lily, Merck Sharp & Dohme, Merck KGaA, Novartis, Roche, and Servier, and has served as a consultant for Array, Astellas, AstraZeneca, Bayer, Beigene, Biocartis, Boehringer Ingelheim, Bristol-Myers Squibb, Celgene, Daiichi, Halozyme, GSK, Incyte, Ipsen, Lilly, Merck Sharp & Dohme, Merck KGaA, Novartis, Pierre Fabre, Roche, Servier, Sirtex, Taiho. ZW reports research support from BMS, Gilead Sciences, Novartis, Plexxikon, and has served as a consultant for AstraZeneca, Bayer, Daiichi, Five Prime, Gilead, Lily, Macrogenic, and Merck. MZ has nothing to disclose. DI has nothing to disclose. JL is an employee of Gilead Sciences, and owns stock in Gilead Sciences, and Roche. SDP is an employee of Gilead Sciences, Inc., and owns stock in Gilead Science, and Amgen. CBB, DT, EE, JH, JZ, K-WL, PB, ST are employees of and own stock in Gilead Sciences. DCa has received consulting fees from Genentech, Roche, Eli Lilly, Merck, Daiichi Sankyo, BMS, Ono Pharma USA, Five Prime, Seattle Genetics, Amgen, Taiho Pharmaceutical, Astellas Pharma, Gritstone Bio, Pieris Pharmaceutical, Zymeworks, Basilea Pharmaceutica, QED Group, Arcus Biosciences, Foundation Medicine, Pierian Biosciences, Silverback Therapeutics, Servier Pharmaceuticals, Blueprint Medicines, Tempus, Guardant Health, Archer Biosciences, and Natera, and has received honoraria from Genentech, Roche, Eli Lilly, Merck, Daiichi Sankyo, AstraZeneca, Tempus, and Guardant Health.

© Author(s) (or their employer(s)) 2021. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.

Figures

Figure 1

OS (A) and PFS (B) by treatment arm and change from baseline in CD8+ cell density (Wilcoxon signed-rank test) (C) and gene signature scores (D). (A)*Since the primary endpoint was not met, OS was not analyzed per protocol but for exploratory purposes. (B)*Since the primary endpoint was not met, PFS was not analyzed per protocol but for exploratory purposes. BL, baseline; CR, complete response; IFN, interferon; n.s., not significant; OS, overall survival; PD-L1, programmed death ligand 1; PFS, progression-free survival; PR, partial response; TC+IC, tumor and associated immune cell-positive.

Figure 2

OS (A) and PFS (B) of total treated population, overall survival in all treated patients with PD-L1 (TC+IC)-positive tumors (C) and PD-L1 (TC-only)-positive tumors (D), OS analyses in select subgroups (E). Gray shading denotes probability of survival between upper and lower 95% CI. ADX, andecaliximab; IFN, interferon; OS, overall survival; PBO, placebo; PD-L1, programmed death ligand 1; PFS, progression-free survival; TC+IC, tumor and associated immune cell-positive.

Figure 3

Associations between hallmark EMT ssGSEA score and TGF-β gene signatures in all archival tumors with gene expression (n=80) (A), and between TGF-β fibrosis gene signature scores and Hallmark EMT ssGSEA score (B). Correlation between desmoplasia and TGF-β (C), between desmoplasia and EMT (D). OS for EMT ssGSEA score high versus low (median cut) patients (E), and OS for TGF-β ssGSEA score high versus low (median cut) patients (F). (A)* In all archival tumors with gene expression data (n=80). EMT, epithelial-to-mesenchymal transition; OS, overall survival; ssGSEA, single-sample gene set enrichment analysis; TGF, transforming growth factor.

Figure 4

Plot analysis of gene signatures associated with survival in ERBB2 expression according to documented HER2 status (A), OS by HER2 status (B), chromosome instability signature (almac) according to HER2 status (C)57 58, and OS by TMB high versus low (median cut) (D). HER2, human epidermal growth factor receptor 2; OS, overall survival; TMB, tumor mutation burden.

Figure 5

Association between documented HER2 status and TGF-β fibrosis signature according to HER2 status (A), TMB according to HER2 status (B), correlation between ERBB2 gene expression and TGF-β fibrosis ssGSEA score (C), and a Cox model of OS incorporating TMB, TGF-β fibrosis, and HER2 (D). (C)*All HRs refer to biomarker level above median versus below median. HER2, human epidermal growth factor receptor 2; OS, overall survival; ssGSEA, single-sample gene set enrichment analysis; TGF, transforming growth factor; TMB, tumor mutation burden.