Abstract PD1-11: Associations between the gut microbiome and aromatase inhibitor (AI)-associated musculoskeletal symptoms (AIMSS)

Abstract Background: AIMSS commonly affect AI- treated patients, and can lead to treatment discontinuation. The etiology of AIMSS is poorly understood. Alterations in the bacterial composition of the gut microbiome have been associated with inflammation. In particular, the presence of Bifidobacteria, members of the phylum Actinomycetota, in the stool has been associated with decreased systemic inflammation. Studies have shown that the gut microbiome is both influenced by estrogen and influences estrogen levels. We examined whether estrogen deprivation with AI therapy alters the composition of the gut microbiome and whether there are associations between the microbial composition and development of AIMSS. Methods: We enrolled females who were starting AI therapy (Clinicaltrials.gov NCT05700006). 53 patients with breast cancer starting AI therapy collected a total of 151 stool samples, before initiation of treatment (n=53) and after 4 and/or 12 weeks (n=98). Stool samples were also collected and analyzed from 10 patients with breast cancer who were not initiating endocrine therapy. DNA was extracted from stool samples, and 16S rRNA gene sequences were used to characterize and compare microbial communities. Participants were followed prospectively to assess persistence with AI therapy and reasons for discontinuation. Wilcoxon signed rank tests were used to compare microbial communities from patients who did and did not develop AIMSS. Results: Of the 53 enrolled participants who were starting AI therapy, mean age was 63.4 years (SD 9.0), mean body mass index was 29.2 kg/m2 (SD 6.1), and 12 (22.6%) had received chemotherapy. Fifty-two (98%) initiated anastrozole, and 1 received exemestane. During the follow-up period, 12 (22.6%) discontinued the initially-prescribed AI therapy because of AIMSS. Stool samples from baseline compared to 4 and/or 12 weeks after initiation of AI therapy revealed no change over time in bacterial composition of the stool, as assessed using measures of both alpha and beta diversity. Comparing baseline samples from AI-treated patients who did and did not discontinue AI therapy because of AIMSS, there was a trend toward increased relative abundance of Actinomycetota in individuals that did not discontinue due to AIMSS (p=0.057). Differences were not identified for 6 other phyla examined. When species within Actinomycetota were examined, patients who did not discontinue AI therapy because of AIMSS had a trend toward higher relative levels of Bifidobacteria (p=0.058). Conclusions: This study demonstrated a potential association between increased relative levels of Bifidobacteria before AI therapy initiation and decreased likelihood of developing AIMSS. Future studies can investigate whether dietary manipulations that have been shown to increase levels of Bifidobacteria in the gut could decrease the incidence of AIMSS, thereby enabling increased persistence with AI therapy and improved quality of life. Citation Format: N. L. Henry, A. Stickland, A. W. Schmidt, K. M. Kidwell, E. V. Joyce, S. Kozar, A. F. Schott, C. Van Poznak, K. Kemmer, T. M. Schmidt. Associations between the gut microbiome and aromatase inhibitor (AI)-associated musculoskeletal symptoms (AIMSS) abstract. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PD1-11.