Immune checkpoint blockade (ICB) therapeutics are a breakthrough class of immune-activating monoclonal antibodies for the treatment of advanced solid tumors. Limitations on clinical response to ICB include aspects of immunity: i) intrinsic to the tumor, ii) extrinsic to the tumor and derived from peripheral immune fitness and environmental immunomodulatory factors, and iii) involved in development of therapeutic toxicity. Here, I approach these limitations to ICB response through the measurement and functional characterization of endogenous and exogenous metabolites, effector molecules in the energetic and signaling determinants of immunity.
In studying the determinants of successful intratumoral immune activation, I describe a unique metabolic dysregulation which propagates infection-free induction of Type-I Interferons, key cytokines in cellular immunity. Cells depleted of ADA2, a purine metabolic enzyme, bear diminished reserved of s-adenosyl methionine, a key cofactor in DNA methylation and repression of genomic endogenous retroviral elements (ERVs). Under alleviated repression, ERVs are expressed and sensed by cell-intrinsic viral sensing machinery, producing Type-I Interferon induction. This mechanism of purine metabolic reprogramming can be used to significantly alter the inflammatory profile of a tumor towards ICB responsiveness.
In complement, I studied the systemic progression of ICB toxicity, a multi-organ collection of ICB-induced autoimmunity which causes therapeutic termination independent of tumor responsiveness. Loss of two circulating lipid metabolites, LPC (18:2) and LPC (16:0), was observed in patients who develop both ICB-induced and traditional autoimmunity. Supplementation of LPC (18:2) in mice ameliorated deleterious inflammation of chemical or ICB origin via negative regulation of circulating neutrophils, effector myeloid cells in the inflammatory response. This protective effect occurred without impacting anti-tumor immunity, suggesting other pro-resolution metabolites in circulation may similarly regulate toxicity without limiting ICB efficacy.
Finally, I sought to characterize mechanisms of reported influences by the commensal gut microbiome on tumor ICB responsiveness via measurement of microbially derived metabolites in circulation, and their value as response predictive features. I characterize a class of dihydroxy-bile acid glucosides which predict ICB response and are positively regulated by previously characterized response-associated gut microbes. Cumulatively, these observations represent three approaches to addressing the clinical limitations of ICB interventions through metabolic characterization and reprogramming.