Abstract LB065: Genetically heterogeneous mouse models identify IFNg signaling as a shared signature of acquired resistance to immune checkpoint inhibitors

Abstract

Abstract Immune checkpoint inhibitors (ICI) for cancer therapy have improved outcomes for a variety of malignancies, however many patients fail to benefit. To date, most pre-clinical studies investigating the tumor microenvironment (TME) have utilized different spontaneous or implanted tumor lines to separately investigate immunologically cold or hot TMEs. These models are often in different mouse strains, leading to confounding variables and a disconnect as to whether the tumor or the host is contributing to the observed immunological phenotype. To account for host diversity and reduce variation on the part of the tumor, we have developed immunotherapy models using genetically heterogeneous Diversity Outbred (DO) and recombinant inbred Collaborative Cross (CC) mice. We crossed DO mice with C57BL/6 (B6) to generate DOB6F1 mice that reliably accept B6-syngeneic B16F0 tumors after subcutaneous inoculation. DOB6F1 mice (n=142) treated with combined anti-PD1/anti-CTLA-4 ICI on days 3, 6, and 10 after inoculation exhibited a wide variation in tumor latency, up to a maximum of 65 days, with 19 mice never developing tumor. Quantitative Trait Locus analysis revealed multiple loci influencing response to ICI. We utilized this data to challenge 12 CCB6F1 strains selected based on predicted response where ICI outcomes range from non-responsive to near complete response. In addition, we show evidence that the DOB6F1 model recreates acquired resistance to ICI, with 9 mice having an extremely delayed tumor latency (>40 days). Melanin-free regions were observed in 3/9 of these tumors, suggesting tumor editing. Acquired resistance was also noted 3/10 ICI-treated CC051B6F1 mice. Whole transcriptomic analysis compared tumors from non-responder versus delayed latency DOB6F1 mice. Despite implantation within genetically heterogeneous mice, transcriptomic profiles from late-onset tumors cluster together. Gene Set Enrichment Analysis identified immune processes, with antigen processing and presentation as the most significantly dysregulated gene set. Many of the upregulated genes in late-onset tumors are driven by IFNγ, suggesting IFNγ signaling may contribute to immune escape. We tested this directly by culturing B16 cells with IFNγ prior to inoculation. Short exposure of B16 to IFNγ results in aggressive growth regardless of treatment, but long exposure increases tumor immunogenicity and responsiveness to ICI in inbred B6 mice. Experiments in ICI-responsive CT26-bearing BALB/c mice indicate IFNγ signaling contributes to ICI response, as demonstrated by increased tumor burden after a single 250 µg dose of neutralizing anti-IFNγ antibody concurrent with ICI treatment. Collectively, our DO and CC F1 models allow for reduced tumor variation with a focus on the host and associated TME, and we show differential roles of IFNγ in response to ICI based on the timing of IFNγ exposure. Citation Format: Justin B. Hackett, James Glassbrook, Nasrin Movahhedin, Madeline Bross, Alicia Kevelin, Maria Múniz, Heather Gibson. Genetically heterogeneous mouse models identify IFNg signaling as a shared signature of acquired resistance to immune checkpoint inhibitors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr LB065.