The Microbiome and Clinical Outcome Among Patients with Multiple Myeloma (MM) Following CD38-Directed Antibody Treatment

Background:

The gut microbiota produce a variety of metabolites, which signal metabolic pathways influencing metabolism and immunity. There are growing data to support interactions between host-gut microbes and treatment responses in various malignancies. A prospective study of serial changes in gut microbiomes in patients with MM after autologous stem cell transplantation (ASCT) showed that reduced alpha diversity at engraftment was associated with a lesser response to ASCT. Another study in patients with MM after an induction therapy showed that minimal residual disease (MRD) negativity was associated with a higher abundance of Eubacterium hallii in stool samples. Our study was designed to assess changes in the gut microbiome associated with daratumumab (Dara) based therapy.

Methods:

Patients with relapsed MM and prior autologous transplantation who had received 1-4 prior lines of therapy were eligible. Two stool samples were collected, one within 1 week prior to Dara (pre-Dara) and one immediately after 4 doses of Dara (post-Dara). Stool samples were uniformly collected using the Zymo DNA/RNA Shield Fecal Collection Tubes (Zymo Research, Irvine) and stored at -80C until processing by TGen North Clinical Microbiome Services Center. Metagenomics sequencing was conducted by Illumina NovaSeq X. Microbial functional pathways were analyzed using HUMAnN3 and QIIME2.

Results:

Forty patients with MM were enrolled, with a median prior lines of therapy of 2 (range: 1-4). In this study, 60% of the patients were female, the median age was 63 (range: 43-78), and 35% of patients had high risk disease. There were no statistical differences between overall Pre- and Post-Dara gut microbiomes in all patients. Five patients did not respond (non-responders), and 35 patients achieved partial response (PR) or better (responders). ANCOM-BC analysis showed statistically significant (p ≤ 0.05) overgrowth of Alistipes finegoldii and Acidaminococcus intestini species in responders and Clostridium symbiosum, Enterocloster clostridioformis and Blautia producta in non-responders. No differentiating functional pathways were detected between Pre- and Post-Dara samples and between responders and non-responders. Adjustment for additional therapy to Dara (pomalidomide [12], lenalidomide [11] and others [5]) was analyzed, after which we found no statistically significant difference between single agent Dara and combination groups. For prophylaxis during Dara therapy, all patients received acyclovir, and 7 patients received sulfamethoxazole and trimethoprim for prophylaxis. Three patients took oral antibacterial antibiotics (levofloxacin and azithromycin) within 4 weeks before the second collection. Levofloxacin, azithromycin and sulfamethoxazole/trimethoprim did not significantly affect microbiome diversity.

Conclusion:

To our knowledge, this is among the first studies to evaluate the role of microbiota in clinical outcomes among relapsed MM treated with Dara. Despite our somewhat limited sample size, our preliminary results suggest statistically significant differences in specific species between clinical responders and non-responders. Future trials with an expanded number of patients are needed to confirm our findings and to further explore potential differences by pathways. If our results are confirmed, it could inform potential future trials to evaluate the use of probiotics to encourage growth of predominant species in responders to promote better responses to CD-38 directed antibody treatment. Alternatively, probiotic trials among MM patients to inhibit growth of dominant microbes in non-responders can potentially be considered.