Safety and tolerability of andecaliximab as monotherapy and in combination with an anti-PD-1 antibody in Japanese patients with gastric or gastroesophageal junction adenocarcinoma: a phase 1b study

Abstract

Background: Matrix metalloproteinase 9 (MMP9) is implicated in protumorigenic processes. Targeting either stromal or epithelial MMP9 reduces the incidence of metastasis. Andecaliximab is a monoclonal antibody that targets MMP9 with high affinity and selectivity. However, no study has examined whether the inhibition of T-cell programmed death 1 (PD-1) in the presence of andecaliximab increases activated lymphocyte infiltration into the tumor, thereby increasing antitumor activity more than that in anti-PD-1 monotherapy. In this study, we assessed the safety, pharmacokinetics (PK), exploratory biomarkers, and preliminary efficacy of andecaliximab as monotherapy and in combination with nivolumab in Japanese patients with advanced or recurrent gastric or gastroesophageal junction (GEJ) adenocarcinoma.

Methods: This phase 1b study comprised four cohorts enrolling Japanese patients with gastric or GEJ adenocarcinoma. This paper concerns cohorts 1 and 4; cohorts 2 and 3 will be reported subsequently. Cohort 1 enrolled patients with human epidermal growth factor receptor 2 (HER2)-negative tumors (n=8) who received andecaliximab monotherapy (800 mg by intravenous infusion every 2 weeks (Q2W)), and cohort 4 enrolled patients irrespective of their HER2 status (n=10) who received 800 mg of andecaliximab in combination with nivolumab Q2W. Safety, dose-limiting toxicities (DLTs), PK, pharmacodynamics, and biomarkers were assessed in both cohorts.

Results: PK of andecaliximab in Japanese patients with gastric or GEJ adenocarcinoma was similar to that reported in non-Japanese patients with advanced solid tumors. Andecaliximab monotherapy and in combination with nivolumab demonstrated no DLTs in cohort 1 and 4, respectively. Toxicities were manageable and well tolerated in both cohorts. The median progression-free survival was 1.4 months (90% CI, 0.5 to 5.4) and 4.6 months (90% CI, 0.9 to not reached) in cohorts 1 and 4, respectively. The objective response rate was 50% (90% CI, 22% to 78%) in cohort 4, and in some patients, the combination therapy was effective regardless of the biomarker status.

Conclusions: The andecaliximab-nivolumab combination demonstrated a manageable safety profile and promising clinical activity in patients with advanced gastric adenocarcinoma.NCT02862535.

Keywords: combination; drug therapy; gastrointestinal neoplasms.

Conflict of interest statement

Competing interests: None declared.

© Author(s) (or their employer(s)) 2022. Re-use permitted under CC BY. Published by BMJ.

Figures

Figure 1

Pharmacodynamic evaluation of andecaliximab. andecaliximab-free matrix metalloproteinase 9 (ADX-free MMP9) was undetectable on day 15 of cycle 1; therefore, circulating MMP9 was bound to andecaliximab in week 2 after dosing in cohort 1 patients. Circulating free MMP9 levels were below the limits of quantitation on-treatment at trough, indicating that all MMP-9 was bound to andecaliximab for the duration of the Q2W dosing interval. The half-life of andecaliximab is 1 week; therefore, it can inhibit tumor growth as an anti-MMP9 agent. ADX, andecaliximab; MMP9, matrix metalloproteinase 9.

Figure 2

Cohort 4 biomarker status and best overall response. There were five responders: 80% (4/5) were mismatch repair-proficient, 100% (5/5) were Epstein–Barr virus-negative and programmed death ligand-1 (PD-L1) Tumor Proportion Score-negative, and 80% (4/5) were PD-L1 Combined Positive Score (CPS)-positive. However, the proportion of responders among PD-L1 CPS-positive patients was 4/7. ADX, andecaliximab; EBV, Epstein–Barr virus; MMP9, matrix metalloproteinase 9; MMR, mismatch repair; PD, pharmacodynamics; PD-L1, programmed death ligand-1; PR, partial response; SD, stable disease; TC, tumor cell.