Helicobacter pylori and immunotherapy for gastrointestinal cancer

Abstract

Helicobacter pylori infection is associated with the risk of gastrointestinal (GI) cancers; however, its impact on immunotherapy for GI cancers remains uncertain. In this study, we included 10,122 patients who underwent ¹³C-urea breath tests. Among 636 patients with Epstein-Barr virus-negative microsatellite-stable gastric cancer (GC) who were treated with anti-PD-1/PD-L1 therapy, H. pylori-positive patients exhibited significantly longer immune-related progression-free survival (irPFS) compared with H. pylori-negative patients (6.97 months versus 5.03 months, p < 0.001, hazard ratio [HR] 0.76, 95% confidence interval [CI] 0.62-0.95, p = 0.015). Moreover, the H. pylori-positive group demonstrated a trend of 4 months longer median immune-related overall survival (irOS) than the H. pylori-negative group. H. pylori-positive GC displayed higher densities of PD-L1⁺ cells and nonexhausted CD8⁺ T cells, indicative of a "hot" tumor microenvironment. Transcriptomic analysis revealed that H. pylori-positive GC shared molecular characteristics similar to those of immunotherapy-sensitive GC. However, H. pylori-positive patients with DNA mismatch repair-deficient (dMMR)/microsatellite instability-high (MSI-H) colorectal adenocarcinoma and esophageal squamous cell carcinoma (ESCC) had shorter irPFS compared with H. pylori-negative patients (16.13 months versus not reached, p = 0.042, HR 2.26, 95% CI 1.13-4.50, p = 0.021 and 5.57 months versus 6.97 months, p = 0.029, HR 1.59, 95% CI 1.14-2.23, p = 0.006, respectively). The difference in irOS between H. pylori-positive and -negative patients had the same trend as that between dMMR/MSI-H colorectal adenocarcinoma and ESCC patients. We also identified a trend of shorter irPFS and irOS in H. pylori-positive liver cancer and pancreatic cancer patients. In summary, our findings supported that H. pylori infection is a beneficial factor for GC immunotherapy by shaping hot tumor microenvironments. However, in dMMR/MSI-H colorectal adenocarcinoma and ESCC patients, H. pylori adversely affects the efficacy of immunotherapy.

Graphical abstract

Figure 1

The inclusion and exclusion procedures for main hypotheses (A–C) The inclusion and exclusion steps of EBV-negative MSS GC patients, dMMR/MSI-H colorectal adenocarcinoma patients, and esophageal cancer patients, respectively.

Figure 2

Survival curves of EBV-negative MSS gastric adenocarcinoma patients (A and B) The survival analysis of irPFS and irOS of EBV-negative MSS gastric adenocarcinoma patients, respectively. These patients were grouped by their H. pylori infection status. Survival analyses were conducted by log rank test. p values are two sided.

Figure 3

Survival curves of dMMR/MSI-H colorectal adenocarcinoma patients (A and B) The survival analysis of irPFS and irOS of dMMR/MSI-H colorectal adenocarcinoma, respectively. These patients were grouped by their H. pylori infection status. Survival analyses were conducted by log rank test. p values are two sided.

Figure 4

Survival curves of ESCC patients (A and B) The survival analysis of irPFS and irOS of ESCC, respectively. These patients were grouped by their H. pylori infection status. Survival analyses were conducted by log rank test. p values are two sided.

Figure 5

The molecular characteristics of the TME of H. pylori–positive and H. pylori–negative gastric adenocarcinoma. (A–F) Comparison of the density of immune cells between H. pylori–positive GC and H. pylori–negative GC by Student’s t- test. (G and H) Comparison of PD-1 and PD-L1 expression level between H. pylori–positive and H. pylori–negative GC by limma package. (I) Comparison of gene signature scores between H. pylori–positive and H. pylori–negative GC by Student’s t test. p values are two sided. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ns, not significant.