Abstract Background. The goal of this study was to examine the human gut virome in advanced lung cancer patients being treated with immune checkpoint inhibitor (ICI) therapy to determine whether patterns of viral detection could predict clinical benefit. Methods. This cohort of 66 patients with stage III-IV lung cancer patients was recruited between June and February 2019. Eligibility was limited to treatment-naïve individuals who were scheduled to receive ICI therapy at Moffitt Cancer Center. DNA was extracted from stool collected within 24-72 hours prior to each patient’s first ICI cycle. To generate the mammalian virus and bacterial virus (phage) taxonomic profiles, reads were aligned to publicly available databases using program Bowtie2 version 2.3.5.1. Multivariable logistic regression was used to estimate the association between the relative abundance of viral taxonomic units (OTUs) and ICI clinical benefit. Random forest machine learning analysis was used to select a set of viral OTUs whose relative abundance most accurately classified patients as yes/no for clinical benefit of ICI therapy. Results. We observed clinical benefit in 34 lung cancer patients treated with ICI therapy. We observed differences in the relative abundance of multiple components of the human gut virome between lung cancer patients according to ICI clinical benefit. Most marked were the differences observed in the relative abundance of phages (bacterial viruses). We observed that 239 phages were abundant at different frequencies in lung cancer patients with versus without clinical benefit of ICI therapy at the q0.05 level (multiple testing corrected p-value). In addition, an area under the curve (AUC) for discrimination of ICI clinical benefit of 91% could be achieved using a 15-phage combination. This included 10 OTUs from the family Siphoviridae, 2 from the family Myoviridae, 2 from Podoviridae, and 1 from family Caudoviridales. For all but one of these 15 phages, we were able to identify a consensus bacterial genus as the phage target with high confidence. Two of these bacteria - Ruminococcus and Bacteroides - have consistently been identified in microbiome studies as being associated with ICI response. Conclusion. Implications of our research include not just the elucidation of a further component of the human gut environment - phages - that impacts human health and disease, but also the identification of its specific role in potentially altering response to cancer-directed immune checkpoint inhibitor therapy. Citation Format: Anna E. Coghill, Jin Xu, Aleksandr Lazaryan, Doratha Armenthus Byrd, YOUNGCHUL KIM, Richard Pollenz, Lary A. Robinson. The human virome and its association with clinical benefit of immune checkpoint inhibitor therapy among patients with advanced non-small cell lung cancer abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 3620.
Coghill et al. (Fri,) studied this question.