Distinct tumor microbiome and metabolomic profiles to inform responses of patients with esophageal cancer to neoadjuvant chemoradiotherapy and immunotherapy.

Seoho Lee,James R White,Ronan Joseph Kelly,Ali Hussainy Zaidi,Josephine Louella Feliciano,Russell K. Hales,K. Ranh Voong,Richard J. James Battafarano,Blair A Jobe,Stephen C. Yang,Stephen Broderick,Jinny Suk Ha,Elizabeth Thompson,Eun Ji Shin,Ali Imran Amjad,Patrizia Guerrieri,Benny Weksler,Vincent K. Lam,Cynthia Sears,Fyza Shaikh
DOI: https://doi.org/10.1200/jco.2024.42.16_suppl.4084
IF: 45.3
2024-06-01
Journal of Clinical Oncology
Abstract:4084 Background: Preclinical and clinical data indicate that neoadjuvant chemoradiotherapy (CRT) may prime an anti-tumor immunological response in esophageal cancer driven by intratumoral CD8+ T cells. However, responses to neoadjuvant therapy can vary widely across patients. The microbiome, a novel and potentially modifiable biomarker of immunotherapy response, may inform neoadjuvant response. Methods: Fecal and tumor samples were collected from patients with stage II-III resectable distal esophageal or gastroesophageal junction carcinoma treated with neoadjuvant immunotherapy followed by chemoradiotherapy (CRT) and immunotherapy (IO). Patients on arm A (n=15) received 2 cycles of induction with nivolumab plus 3 additional cycles with CRT. Patients on arm B (n=10) received nivolumab and relatlimab on a similar induction schedule, but concurrent IO and CRT resulted in immune related adverse events and a protocol amendment to CRT (NCT03044613). We examined fecal (n=19) and tumor (n=12) patient samples across both arms using 16S rRNA amplicon sequencing and high-resolution taxonomic assignment. Additionally, fecal samples were analyzed using an untargeted panel (General Metabolics) and pathway analysis was performed using Metaboanalyst (v6.0). Results: The fecal microbiome of patients with pathological complete response (pCR) grouped in distinct clusters compared to patients without pathological complete response (non-pCR) using Bray-Curtis (p 10% and <10% residual tumor grouped similarly but were significantly different from patients with 0% residual tumor (p < 0.001). On a linear discriminant axis, the top 10 abundant bacteria were significantly increased in pCR compared to non-pCR (p < 0.001). After FDR correction, six bacteria had significantly differing abundances. Bacteroides finegoldii, Paraprevotella clara, Ruminococcus callidus, and Roseburis inulinivorans were significantly enriched in pCR patients. Clostridiales and Bacteroides were significantly enriched in non-pCR patients. Fecal metabolomic analysis demonstrated six significant metabolites: in non-pCR, ropinirole, valine, phenylalanine, and isoleucine were enriched; in pCR, C16cer and D-urobilinogen were enriched (p < 0.05). Patients with improved neoadjuvant response had significantly higher abundances of Bacteroides and corresponding C16cer levels. Pathway analysis revealed significant involvement in aminoacyl-tRNA biosynthesis and sphingolipid metabolism. Conclusions: Patients with complete responses to combined CRT and IO for operable esophageal cancer had distinct fecal microbiome profiles as well as metabolomic pathways. These results may be useful to further characterize and predict patients who have improved responses to neoadjuvant therapy and serve as a basis for therapeutic intervention.
oncology
What problem does this paper attempt to address?