Melanoma tumor microbiome as a predictor of anti-PD-1 and anti-CTLA-4 immunotherapy response.

Matthew Uzelac,Wei Tse Li,Yuxiang Li,Weg M. Ongkeko

DOI: https://doi.org/10.1200/jco.2024.42.16_suppl.e14666

IF: 45.3

2024-06-01

Journal of Clinical Oncology

Abstract:e14666 Background: In 2023, it is estimated that 97,610 melanoma diagnoses were made in the United States, resulting in 7,990 deaths. For high-risk melanomas, immune checkpoint inhibitors are often administered as an adjuvant to surgery, though they yield variable response rates. The gut microbiome’s association to these treatments’ success has been demonstrated, largely through fecal transplant and probiotic trials. In this study, we seek to determine whether tissue-resident microbes in melanoma tumors could be used to predict immunotherapy response. Methods: Sequencing data of 289 melanoma tumor tissue samples was downloaded across three distinct studies. Intratumor microbial species abundances were derived from RNA or DNA sequencing data through direct alignment to microbial reference databases. Gene expression profiling was also performed. Microbial abundance values were used to construct models predicting patients’ responses to check-point blockade immunotherapies, which were then validated on an external dataset. Results: 12 and 14 microbial species were differentially abundant in anti-PD-1 responders and anti-CTLA-4 responders, respectively. We identified significant correlations of these species to genes of the PD-1 and CTLA-4 signaling pathways. Among others, greater abundance of Corynebacterium kroppenstedtii and Streptococcus gordonii correlated to decreased expression of PD1 and CTLA-4. With the abundances of these taxa and the expressions of 37 genes known to be implicated in immunotherapy response, we developed two classifiers to predict a patient’s response to anti-PD-1 and anti-CTLA-4 therapies. When tested externally, these models successfully predicted a majority of responders and non-responders with sensitivities of 80.0% and 60.0% and specificities of 95.2% and 95.8%, respectively. Conclusions: Our study revealed that checkpoint blockade response for melanomas could be predicted with microbe abundances. Our analysis demonstrates the potentially promising clinical utility of intratumor microbiota in cancer treatment. Given their location in the tumor microenvironment and their ability to influence immune processes, intratumor microbes could play a critical role in modulating the therapeutic effects of immune checkpoint blockades. We hope that our results could also be applied to other solid tissue cancers, where a significantly smaller fraction of patients respond to immunotherapy. [Table: see text]

oncology