Impact of intratumoral microbiome on tumor immunity and prognosis in human pancreatic ductal adenocarcinoma

Abstract

Background: Recent evidence suggests that the presence of microbiome within human pancreatic ductal adenocarcinoma (PDAC) tissue potentially influences cancer progression and prognosis. However, the significance of tumor-resident microbiome remains unclear. We aimed to elucidate the impact of intratumoral bacteria on the pathophysiology and prognosis of human PDAC.

Methods: The presence of intratumoral bacteria was assessed in 162 surgically resected PDACs using quantitative polymerase chain reaction (qPCR) and in situ hybridization (ISH) targeting 16S rRNA. The intratumoral microbiome was explored by 16S metagenome sequencing using DNA extracted from formalin-fixed paraffin-embedded tissues. The profile of intratumoral bacteria was compared with clinical information, pathological findings including tumor-infiltrating T cells, tumor-associated macrophage, fibrosis, and alterations in four main driver genes (KRAS, TP53, CDKN2A/p16, SMAD4) in tumor genomes.

Results: The presence of intratumoral bacteria was confirmed in 52 tumors (32%) using both qPCR and ISH. The 16S metagenome sequencing revealed characteristic bacterial profiles within these tumors, including phyla such as Proteobacteria and Firmicutes. Comparison of bacterial profiles between cases with good and poor prognosis revealed a significant positive correlation between a shorter survival time and the presence of anaerobic bacteria such as Bacteroides, Lactobacillus, and Peptoniphilus. The abundance of these bacteria was correlated with a decrease in the number of tumor-infiltrating T cells positive for CD4, CD8, and CD45RO.

Conclusions: Intratumoral infection of anaerobic bacteria such as Bacteroides, Lactobacillus, and Peptoniphilus is correlated with the suppressed anti-PDAC immunity and poor prognosis.

Keywords: Major driver gene alteration; Pancreatic ductal adenocarcinoma; Prognosis; Tumor microbiome; Tumor-infiltrating lymphocytes.

Fig. 1

Study outline. A Flow chart to determine the presence of bacteria in human PDAC tissues. Of the 162 cases, 55 cases in which 16SrRNA could not be extracted by qPCR were excluded. In addition, of the remaining 107 cases, 55 cases in which 16SrRNA could not be identified in the tumor by ISH were excluded. Finally, 52 cases were included in this study. B Representative images of in situ hybridization targeting bacterial 16S rRNA gene. Red arrow heads indicate positive signals. Scale bars, 50 μm. PDAC pancreatic ductal adenocarcinoma, PCR polymerase chain reaction, ISH in situ hybridization

Fig. 2

Microbiome composition in PDAC and non-tumoral tissues. A Taxonomic profiles of predominant bacterial genera by mean relative abundance (%) in PDAC tissue and non-tumoral tissues. B Comparative analysis of the alpha diversity of the microbiome communities between PDAC and non-tumoral tissues using Shannon index. C Principal coordinate analysis (PCoA) using weighted-UniFrac distance of beta diversity among PDAC and non-tumoral tissues. PDAC pancreatic ductal adenocarcinoma

Fig. 3

Identification of intratumoral bacteria associated with prognosis in pancreatic cancer patients. A Kaplan–Meier curve of overall survival of patients included in this study. B Overall survival of patients with PDAC in short-term survival group and long-term survival group. The patients were divided into two groups based on the median follow-up period (23.6 months). The survival curves were compared between them by the log-rank test. C LEfSe calculations between the short-term survival group and the long-term survival group were performed using a threshold of 2.0. D Forest plots showing the hazard ratio of prognosis by the presence of each genus detected by LEfSe. PDAC pancreatic ductal adenocarcinoma, LEfSe linear discriminant analysis effect size

Fig. 4

Correlations among prognostic bacteria and tumor prognosis. A Spearman correlation between Bacteroides, Lactobacillus, and Peptoniphilus. B Kaplan–Meier curve of overall survival of patients with and without prognostic bacteria. PDAC pancreatic ductal adenocarcinoma

Fig. 5

Correlation of the number of TILs and the presence of prognostic bacteria. The abundance of any of the genera Bacteroides, Lactobacillus, or Peptoniphilus was analyzed by Spearman correlation. TILs tumor-infiltrating lymphocytes