Molecules generated by mutational and epigenetic processes in tumors have been associated with recognition of tumors by the innate and adaptive immune system. For example, neoantigens have been implicated in response to checkpoint blockade therapies. Likewise, the display of pathogen-associated patterns by nucleic acids unsilenced by epigenetic alterations have been implicated in activation of the innate immune system. Here we determine molecular features which place a tumor at a selective advantage or disadvantage, and how these selective pressures depend on the tumor’s environment. We have proposed general frameworks to address these questions. In the case of tumor neoantigens, we present a fitness model of candidate immunogenic neoantigens distributed across a tumor’s subclonal structure in a given microenvironment. We show how our approach can be used to characterize response to checkpoint-blockade therapies and apply it to the general problem of immune-driven tumor evolution in a unique cohort of long-term survivors of pancreatic cancer. In the case of immunostimulatory RNA, we proposed a method of calculating entropic forces for determining the likelihood of tumoral RNA being recognized as pathogen-associated and characterizing classes of pathogen mimicry.

Citation Format: Marta Luksza, Alexander Solovyov, Nicolas Vabret, Vinod Balachandran, Nadeem Riaz, Vladimir Makarov, Matthew D. Hellmann , Alexandra Snyder, Samuel Funt, Romain Remark, Miriam Merad, Sacha Gnjatic, Dean F. Bajorin, Jonathan Rosenberg, Steven Leach, Arnold J. Levine, Timothy A. Chan, Nina Bhardwaj, Jedd Wolchock, Benjamin D. Greenbaum. Measuring the emergence of non-self in tumors [abstract]. In: Proceedings of the Fourth CRI-CIMT-EATI-AACR International Cancer Immunotherapy Conference: Translating Science into Survival; Sept 30-Oct 3, 2018; New York, NY. Philadelphia (PA): AACR; Cancer Immunol Res 2019;7(2 Suppl):Abstract nr IA31.