As planned, exploiting our extensive collection of cell models, we studied cancer vulnerabilities weakening the efficiency of DDR pathways. To this end, we developed a loss of function CRISPR CAS9 murine library encompassing the mammalian complement of 500 genes involved in almost all DDR pathways. To study how the loss of function of DDR genes triggers an anticancer immune response, we exploited murine CRC cell lines and syngeneic mouse models.
Since the majority of CRC patients are immune refractory, we assessed whether disabling DDR genes could make their tumors eligible for immune-based therapies. Recent findings show the coexistence of MSS/MMRp and MSI/MMRd cancers cells in the same tumor lesion in almost 1% of CRC patients (Fig. 3). Intrigued by these evidences, we generated a preclinical model of MMR heterogeneity, mixing Mlh1+/+ and Mlh1-/- colorectal cancer murine cells at different ratio. Notably, tumor growth delay and tumor rejections occurred in Mlh1+/+ and Mlh1-/- mixed tumors and increased when the percentages of Mlh1-/- cells augmented in the mixed population. The observation that a MMRp tumor harboring a small fraction of MMRd cells can trigger an effective antitumor immune response has implications for the rational design of clinical trials for tumors recalcitrant to immune checkpoint blockade (ICB).
Tumors can evade immune surveillance through alterations in the antigen presenting machinery (APM). Several studies reported that molecular defects in the major complex of histocompatibility I (MHC I) and in the protein Beta 2 Microglobulin (B2M) represent mechanisms of acquired resistance to ICB in melanoma and lung tumors. We evaluated the impact of B2M loss in immune evasion and resistance to ICB in colorectal, pancreatic and breast cancer murine cell lines. Although the antigen presentation was compromised, the growth of MMRd murine cell lines was still severely impaired by the administration of ICB. The analysis of tumor microenvironment revealed that CD4+ T cells were pivotal in establishing an effective cancer immune response but only in the context of MMRd tumors (Fig. 4).
The impact of the non-coding DNA (98% of the entire genome) on the immunogenicity of MMRd tumors is largely unknown. As part of TARGET we recently found that non-canonical transcripts, can be sources of immunogenic peptides in murine CRC cells (Rospo et al., Genome Medicine, 2024). We initially identified the non-coding derived neoantigens that bound the MHC class I complex and then identified those that were lost after immune pressure in immune-competent mice (the same neoantigens were preserved in immune-compromised mice). Finally, we validated the immunogenicity of the peptides through the measurement of IFN-γ release from splenocytes co-cultured with synthetic peptides. Overall, we provided proof-of-concept that in MMRd tumors, the non-coding part of the genome can effectively contribute to defining the immunogenic properties of these tumor types.
As conceived, TARGET relies on the support of NeoPhore (
https://www.neophore.com/(s’ouvre dans une nouvelle fenêtre)) that aims to discover whether the DNA MMR machinery could be modulated with a small molecule inhibitor, thereby enabling studies on the time-dependent sequelae of MMR deficiency. In collaboration with NeoPhore we found that resorcinol fragment-derived inhibitor 1 (NP1867) binds irreversibly with high affinity and inactivation rate to PMS2 through adduction of Cys73 in the ATP binding site, as demonstrated by protein-ligand X-ray crystallography. In vitro and cell-based selectivity profiling of NP1867 revealed minimal off-target activity. Importantly, in cell-based mechanistic and functional assays, the inhibition of PMS2 by NP1867 was observed.
