Nivolumab plus ipilimumab with or without live bacterial supplementation in metastatic renal cell carcinoma: a randomized phase 1 trial

Abstract

Previous studies have suggested that the gut microbiome influences the response to checkpoint inhibitors (CPIs) in patients with cancer. CBM588 is a bifidogenic live bacterial product that we postulated could augment CPI response through modulation of the gut microbiome. In this open-label, single-center study (NCT03829111), 30 treatment-naive patients with metastatic renal cell carcinoma with clear cell and/or sarcomatoid histology and intermediate- or poor-risk disease were randomized 2:1 to receive nivolumab and ipilimumab with or without daily oral CBM588, respectively. Stool metagenomic sequencing was performed at multiple timepoints. The primary endpoint to compare the relative abundance of Bifidobacterium spp. at baseline and at 12 weeks was not met, and no significant differences in Bifidobacterium spp. or Shannon index associated with the addition of CBM588 to nivolumab-ipilimumab were detected. Secondary endpoints included response rate, progression-free survival (PFS) and toxicity. PFS was significantly longer in patients receiving nivolumab-ipilimumab with CBM588 than without (12.7 months versus 2.5 months, hazard ratio 0.15, 95% confidence interval 0.05-0.47, P = 0.001). Although not statistically significant, the response rate was also higher in patients receiving CBM588 (58% versus 20%, P = 0.06). No significant difference in toxicity was observed between the study arms. The data suggest that CBM588 appears to enhance the clinical outcome in patients with metastatic renal cell carcinoma treated with nivolumab-ipilimumab. Larger studies are warranted to confirm this clinical observation and elucidate the mechanism of action and the effects on microbiome and immune compartments.

Fig. 1. Efficacy outcomes in the treatment of patients with mRCC using nivolumab–ipilimumab with or without CBM588.

a,b, Progression-free response (a) and overall survival (b). c–e, Best response by treatment arm (c), best change in target lesions from baseline (d), and a swimmers plot showing the response and survival characteristics (e). The data are from n = 29 patients (19 patients in the nivolumab–ipilimumab with CBM588 arm and 10 patients in the nivolumab–ipilimumab arm). The Kaplan–Meier log-rank test was used to compare survival between the two arms. Source data

Fig. 2. Microbiome assessment in patients with mRCC treated with nivolumab–ipilimumab with or without CBM588.

a, Change in Bifidobacterium spp. from baseline to week 12 in patients by treatment arm, and by treatment arm and response. b,c, Decreases (b) and increases (c) in relative abundance of gut microbiome species associated with response to nivolumab–ipilimumab with CBM588. Analyses were performed using n = 52 stool samples from n = 26 patients (n = 18 patients in the nivolumab–ipilimumab with CBM588 arm and n = 8 patients in the nivolumab–ipilimumab arm). The Wilcoxon signed rank test was used to perform comparisons between two timepoints within the same treatment arm and the Mann–Whitney U test was used for comparisons between the two arms. Source data

Fig. 3. Changes in metabolic pathways in patients with mRCC treated with nivolumab–ipilimumab with or without CBM588.

a,b Metabolic pathways with significantly different counts between baseline and week 12 in the nivolumab–ipilimumab with CBM588 arm (a) and the nivolumab–ipilimumab arm (b). Gut microbiome analyses were performed using n = 52 stool samples from n = 26 patients (n = 18 patients in the nivolumab–ipilimumab with CBM588 arm (n = 11 responders and n = 7 non-responders); and n = 8 patients (n = 7 non-responders and n = 1 responder) in the nivolumab–ipilimumab arm). The Wilcoxon signed rank test was used to compare metabolic pathways between the two timepoints. Source data

Fig. 4. Changes in circulating cytokine levels from baseline to week 13 by treatment arm.

Cytokine analyses were performed using n = 54 blood samples from n = 27 patients (n = 19 patients in the nivolumab–ipilimumab with CBM588 arm and n = 8 patients in the nivolumab–ipilimumab arm). Wilcoxon signed rank test was used to compare cytokine levels between the two timepoints. Source data

Extended Data Fig. 1. CONSORT diagram.

Patient CONSORT diagram.

Extended Data Fig. 2. Assessment of Shannon diversity index by treatment arm and treatment arm and response.

Assessment of Shannon diversity index by treatment arm (a) and treatment arm and response (b). Gut microbiome diversity was assessed using n=52 stool samples from n=26 patients (n=18 patients in nivolumab/ipilimumab with CBM588 arm [n=11 responders and n=7 non-responders]; and n=8 [n=7 non-responders and n=1 responder] patients in nivolumab/ipilimumab arm). Two-sided Mann-Whitney U test between nivolumab/ ipilimumab with CBM588 and nivolumab/ipilimumab. Two-sided Wilcoxon signed rank test between baseline and week 12. Shannon diversity index was calculated using the species-level abundance data (a,b). Source data

Extended Data Fig. 3. Assessment of fungal microbiome characteristics at order level.

Fungal microbiome analyses were performed using n=52 stool samples from n=26 patients (n=18 patients in nivolumab/ipilimumab with CBM588 arm [n=11 responders and n=7 non- responders]; and n=8 [n=7 non-responders and n=1 responder] patients in nivolumab/ipilimumab arm). Taxa with a mean relative abundance less than 0.001 were summed as other. Source data

Extended Data Fig. 4. Differentially abundant taxa in patients who developed grade 3–4 adverse events (a) and those who did not develop grade 3–4 adverse events.

Differentially abundant taxa in patients who developed grade 3–4 adverse events (a) and those who did not develop grade 3-4 adverse events (b). Analyses were performed using n=52 stool samples from n=18 patients in nivolumab/ipilimumab with CBM588 arm and n=8 patients in nivolumab/ipilimumab arm). Two-sided Mann-Whitney U test used to compare species across two groups. The length of the box plots represents the interquartile range (IQR) and the inside lines of the boxes represent the median. Source data

Extended Data Fig. 5. Changes in circulating cytokine levels from baseline to week 12 by treatment arm.

Changes in circulating cytokine levels from baseline to week 12 by treatment arm. Cytokine analyses were performed using n=54 blood samples from n=27 patients (n=19 patients in nivolumab/ipilimumab with CBM588 arm and n=8 patients in nivolumab/ipilimumab arm). Wilcoxon signed rank test was used to compare cytokine levels across two timepoints. Source data

Extended Data Fig. 6. Changes in myeloid-derived suppressor cell and regulatory T cell populations by time in nivolumab/ipilimumab arm (a) and nivolumab/ipilimumab with CBM588 arm (b).

Changes in myeloid-derived suppressor cell and regulatory T cell populations by time in nivolumab/ipilimumab arm (a) and nivolumab/ipilimumab with CBM588 arm (b). Immune cell populations were assessed in n=54 blood samples from n=27 patients (n=19 patients in nivolumab/ipilimumab with CBM588 arm and n=8 patients in nivolumab/ipilimumab arm). Wilcoxon test was used to compare immune cell populations at two timepoints. The length of the box plots represents the interquartile range (IQR) and the inside lines of the boxes represent the median. Whiskers (the vertical lines extending below and above each box) are used to represent the minimum and the maximum observation. Source data