Interconnection of Gut Microbiome and Efficacy of Immune Checkpoint Inhibitors in Inoperable Non-Small-Cell Lung Cancer

Abstract

The efficacy of immune checkpoint inhibitors (ICIs) in non-small-cell lung cancer (NSCLC) varies widely across patients. Growing evidence indicates that the gut microbiome, through its interaction with the tumor microenvironment, may influence the response to immunotherapy. To investigate this, we analyzed fecal and tumor samples from 63 patients with inoperable NSCLC undergoing ICI therapy. Based on microbiome profiling using 16S rRNA sequencing, patients were grouped according to treatment benefit, defined as progression-free survival (PFS) of six months or longer. Associations between α-diversity indices, microbial composition at the genus and phylum levels, and a composite Sum Index of Binary Abundance (SIBA) were examined in relation to clinical outcomes. Higher microbial α-diversity was linked to improved response to ICIs (p-value = 0.0078 for the Chao1 index). Multiple specific taxa, such as Ruminococcus gauvreauii (p-value = 2 × 10^-4), Ruminiclostridium 9 (p-value = 8 × 10^-4), and [Eubacterium] ventriosum (p-value = 9 × 10^-4), were enriched in patients with favorable outcomes, whereas Oscillibacter and the Eubacterium hallii group were associated with disease progression (p-value = 2 × 10^-3 and 9 × 10^-3, respectively). The SIBA index, which reflects the absence of multiple beneficial bacterial taxa, proved to be a stronger predictor of treatment response than individual taxa alone. Median SIBA values were 18 vs. 24 in patients benefiting from IO therapy compared to non-responders (p-value = 9 × 10^-7). These findings suggest that gut microbiome diversity and composition are closely tied to immunotherapy outcomes in NSCLC. Composite microbial metrics like SIBA may enhance predictive accuracy and inform personalized treatment approaches.

Keywords: immune checkpoint inhibitors; immunotherapy; metagenome; microbiome; non-small cell lung cancer; tumor microenvironment; α-diversity.

Figure 1

Association of clinical characteristics with abundance of genus and phylum taxa. A higher number of distinct taxa were significantly associated with benefit from IO therapy than with other characteristics. Mann–Whitney’s (upper left triangles) and Fisher’s exact test (lower right triangles) p-values for different genus (black names) and phylum (pink names) taxa (horizontal axis), in which abundance values significantly differed between patient groups subdivided by clinical characteristics (vertical axis). Benefit_6—PFS ≥ 6; Benefit_10—PFS ≥ 10; PFS—progression-free survival; OS—overall survival; NLR—ratio of neutrophil to lymphocyte numbers; BMI—body mass index; NLRh_BMIh—high value of NLR (more than T3 of the population studied) and high value of BMI (≥29).

Figure 2

Hierarchical clustering of abundance values for genus and phylum taxa for patients with benefit (green left blocks) or without benefit (red left blocks) from IO. The abundance values for phylum taxa were divided by 10 to normalize with other values. SIBA values were divided by 200.