Predictable regulation of survival by intratumoral microbe-immune crosstalk in patients with lung adenocarcinoma

Abstract

Intratumoral microbiota can regulate the tumor immune microenvironment (TIME) and mediate tumor prognosis by promoting inflammatory response or inhibiting anti-tumor effects. Recent studies have elucidated the potential role of local tumor microbiota in the development and progression of lung adenocarcinoma (LUAD). However, whether intratumoral microbes are involved in the TIME that mediates the prognosis of LUAD remains unknown. Here, we obtained the matched tumor microbiome and host transcriptome and survival data of 478 patients with LUAD in The Cancer Genome Atlas (TCGA). Machine learning models based on immune cell marker genes can predict 1- to 5-year survival with relative accuracy. Patients were stratified into high- and low-survival-risk groups based on immune cell marker genes, with significant differences in intratumoral microbial communities. Specifically, patients in the high-risk group had significantly higher alpha diversity (p < 0.05) and were characterized by an enrichment of lung cancer-related genera such as Streptococcus. However, network analysis highlighted a more active pattern of dominant bacteria and immune cell crosstalk in TIME in the low-risk group compared to the high-risk group. Our study demonstrated that intratumoral microbiota-immune crosstalk was strongly associated with prognosis in LUAD patients, which would provide new targets for the development of precise therapeutic strategies.

Keywords: immune cell; intratumoral microbiota; lung adenocarcinoma; prognosis; tumor microenvironment.

Figure 1. FIGURE 1: Overview of the analysis pipeline.

The tumor microbiome abundance of LUAD was annotated by Poore et al. from RNA-Seq data and the matched host gene expression was downloaded from The Cancer Genome Atlas. Immune cell marker genes were used to build machine learning models to predict patient survival. COX regression analysis based on immune cell marker genes stratified patients into high- and low-risk. The intratumoral microbiota, the tumor immune microenvironment, and their crosstalk between the high-and low-risk groups were further explored.

Figure 2. FIGURE 2: Survival prediction model based on immune cell marker genes using machine learning.

(A) The number of marker genes corresponding to specific immune cell types. (B) ROC curves of 1- to 5-year survival prediction by five-fold cross validation of six machine learning algorithms.

Figure 3. FIGURE 3: Tumor immune microenvironment and related functions are associated with survival in LUAD patients.

(A) Survival curve in different risk score groups obtained by COX regression analysis. (B) Sankey plot showing the correlation between immune cells and the ten genes most associated with survival. (C) Relative abundance of the immune cell components in each patient. (D) Boxplot showing the differences in the abundance of immune cells between the high- and low-risk group. Wilcoxon test was used to perform the statistical test. (E) Five GO terms with the largest number of genes in each class. (F-H) The most significantly enriched GO terms in each class, along with the corresponding genes.

Figure 4. FIGURE 4: The intratumoral microbial profile was significantly different between the high- and low-risk group.

(A) Relative abundance of the intratumoral microbes at the genus level in each patient. Boxplot showing the difference in (B) microbial richness, (C) the Shannon and (D) Simpson indices between the high- and low-risk group. Wilcoxon test was used to perform the statistical test. (E) PCoA based on the Bray-Curtis dissimilarity matrix showing the difference in intratumoral microbial community composition between the high- and low-risk group. (F) Boxplot showing the difference in Bray-Curtis dissimilarity index between the high- and low-risk group. Significantly different microbes in abundance between the high- and low-risk group at the (G) phylum and (H) genus level.

Figure 5. FIGURE 5: Intratumoral microbe-immune crosstalk was associated with survival in LUAD patients.

Microbe-immune cell interaction networks in (A) high- and (B) low-risk groups. Only edges with p < 0.05 were shown in the figure. The size of the node indicates the number of nodes connected to it in the network. The solid yellow line and the dotted gray line indicate positive and negative correlations, respectively. (C) Comparison of parameters of microbe-immune interaction network in high- and low-risk group. On the basis of Figure 5a and b, microbial-immune cell relationship pairs with an absolute value of correlation coefficient greater than 0.2 were screened, further resulting in network plots of (D) high- and (E) low-risk groups.