Relationship between intestinal flora structure and metabolite analysis and immunotherapy efficacy in Chinese NSCLC patients

BACKGROUND: Many immune checkpoint inhibitors (ICIs) have been approved in China to treat non‐small cell lung cancer (NSCLC). However, in the long term, less than 20% of patients benefit from ICIs. To maximize the benefit for NSCLC patients, it is necessary to guide the choice of immunotherapy through biomarkers. Recent studies have shown that gut microbiota can affect tumor response to immunotherapy and might be a potential predictive biomarker. This study analyzed the relationship between intestinal flora structure and metabolomic characteristics in NSCLC and the efficacy of ICIs. METHODS: Prospective analysis of samples from 63 patients with advanced NSCLC who attended the Department of Respiratory Medicine of the Peking Union Medical College Hospital from March 2018 to June 2019, and were prescribed programmed cell death 1 (PD‐1) inhibitors, was carried out. The follow‐up deadline was 31 December 2019. Stool samples were collected from all patients before the start of immunotherapy. DNA was extracted from all samples and libraries were constructed. This was followed by sequencing using the Illumina sequencing platform, and results were studied using a biological information data analysis process. We divided the data into two groups based on progression‐free survival (PFS) ≥ six months and PFS < six months. RESULTS: The median PFS was 7.0 months, not reaching the median overall survival (OS). We obtained 373.5 G of original sequencing data. The phyla Bacteroidetes, Firmicutes, Proteobacteria, and Actinobacteria accounted for most of the bacterial communities in the stool samples studied. Compared with the PFS < six‐month group, the patients in the PFS ≥ six‐month group had significantly higher β‐diversity in the intestinal microbiome at the baseline level. There were also differences in composition between the two groups. Samples in the PFS ≥ six‐month group were rich in Parabacteroides and Methanobrevibacter, while those in the PFS < six‐month group were rich in Veillonella, Selenomonadales, and Negativicutes. The KO, COG, and CAZy databases were used to study functional group protein families, yielding 390 (KO), 264 (COG), and 859 (CAZy) functional group abundances, with significant differences between the two groups. Bacterial metabolites analysis suggested significant differences in the metabolic potential of methanol and methane between the two groups. CONCLUSIONS: We found a close correlation between intestinal microbiome β‐diversity and anti‐PD‐1 immunotherapy response in Chinese patients with advanced NSCLC. The intestinal flora composition, functional group protein family, and KEGG metabolism also differed between the two groups. Differences in pathways and flora metabolites were also noted.