LAUSR

Immune checkpoint inhibitors (ICIs) have made a profound impact on the treatment of a variety of cancers. However, as with any systemic treatment, toxicities are inevitable. With most classes of cancer therapies, toxicities are relatively predictable based on clinical trial safety data and therefore can be handled with prophylactic or supportive care measures. However, ICIs are unique in their ability to cause rare but severe auto-immune toxicities. The molecular underpinnings of these toxicities, as well as unique features of the patient, tumor, or affected tissue, have not been extensively explored. We recently reported a small case series of two patients with myocarditis resulting in death arising following combination ICI therapy (Johnson et al, N Engl J Med, 2016). High lymphocytic infiltration, coupled with PD-L1 expression was present in the affected myocardium and skeletal muscle. Common T cell clones were identified between the affected tissue and tumor, and abnormal expression of muscle-specific transcripts was identified in the associated tumor, suggesting release of peripheral tolerance to tumor-expressed self-antigens.To expand upon our reported study, we collected healthy and afflicted tissue from a series of cancer patients with immune-related colitis, myocarditis (MC), and encephalopathy following ICI treatment. We hypothesize that molecular analysis of these tissues will identify causal factors in the etiology of these toxicities, and how to better predict, prevent, and treat them. Thus, we performed molecular characterization of the immune infiltrate and diseased tissue microenvironment. A total of 20 affected (colon, cardiac, brain) and non-diseased control specimens were examined by spatial digital profiling (nanoString). This process generates a spatial heatmap of digital counts of 20 selected immunology and cellular markers and proteins across each specimen. Using this technology, the landscape of inflammation in ICI-affected organs can be resolved for insights into the mechanism whereby ICI-mediated auto-immunity occurs. Targeted RNAseq for selected immuno-oncology mRNA targets was also performed. In initial RNA sequencing analyses of MC cases, affected myocardium, skeletal muscle, and patient-matched tumors all demonstrated expression of immune activation markers (e.g. interferon-gamma and granzyme B), expression of PD-L1, and muscle-specific genes. In the expanded population, including colitis, digital spatial profiling analyses and targeted NGS (RNAseq) are underway. Although data analyses are incomplete at the time of this abstract, this work will be the largest and most comprehensive analysis of the molecular underpinnings of ICI-mediated auto-immune toxicity reported to date. These data should offer clarity in the mechanisms and features of these adverse events, how to prevent or predict them with precision medicine, and how to treat them when they do occur. Citation Format: Justin M. Balko, Daniel Y. Wang, Yu Wang, Rami Al-Rohil, Margaret Compton, Jeffery A. Sosman, Igor Puzanov, Bret Mobley, Robert D. Hoffman, Yaomin Xu, Javid J. Moslehi, Chanjuan Shi, Douglas B. Johnson. Advanced molecular characterization of severe autoimmune toxicities associated with checkpoint inhibitor therapies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 588. doi:10.1158/1538-7445.AM2017-588