Immune checkpoint blockade (ICB) has demonstrated unprecedented efficacy across multiple cancers, but many patients develop resistance to therapy. Emerging evidence suggests that a durable ICB response requires that the initial anti-tumor inflammatory response be temporally restricted, in part because tumors can exploit homeostatic feedback mechanisms to drive chronically inflamed, immunosuppressed microenvironments. To fully harness the promise of ICB, it is critical to gain mechanistic understanding of how cancer cells adapt to participate in these processes. In this dissertation, we use mouse models of ICB resistance, multi-omic sequencing technologies, and analysis of human clinical data to shed light on how cancer cell epigenomes evolve in response to inflammatory stimuli. In particular, we focus on the enduring consequences of interferon (IFN) signaling, which mediates essential immune functions, but is also capable of inducing profound chromatin remodeling. We first establish that ICB-relapsed cancer cells retain epigenetic memories of immune attack, characterized by persistent chromatin accessibility and H3K4me1 deposition at inflammatory memory domains linked to elevated expression of a subset of interferon-stimulated genes. These genes include OAS1, an RNA pattern recognition receptor that amplifies type I IFN signaling and immune inhibitory genes to promote sustained immune dysfunction. Next, we show that subclonal de-repression of endogenous retroelements (EREs) harboring IFN signaling transcription factor binding sites are sensed by the RNA pattern recognition receptor IFIH1 in a feed-forward manner, driving an ICB-resistant chromatin state. Finally, we demonstrate that breaking this loop pharmacologically with JAK and TBK1 inhibitors in combination erases acquired inflammatory memory domains, ERE expression, and reverses ICB resistance. Inflammatory memory has previously been characterized as a facilitator of trained immunity or enhanced wound healing; this work suggests an additional maladaptive function in which it enables a shift from acute immune cell-derived IFN signaling to chronic cancer cell-derived IFN signaling. Epigenetic reversal experiments also suggest a potential intervention strategy for preventing or reversing acquired ICB resistance, for which there are no currently approved therapies. Collectively, these studies highlight how cancer cell epigenomes evolve to modulate the nature of inflammatory signaling in tumors, with implications for the tailored design of immune-stimulatory therapeutic strategies with optimal timing.