Abstract
The outcome of pancreatic ductal adenocarcinoma (PDAC) remains poor due to few therapeutic options available and challenges with precision therapy to target each tumour's specific characteristics. In this study, a biologically meaningful patient stratification-prognostic model with therapeutic suggestion value based on tumor senescence was developed and validated in multiple independent cohorts. Further mechanistic investigation based on single-cell transcriptomic data and in vitro experiments revealed that complement derived from non-senescent tumor cells stimulates M1 differentiation and antigen presentation, while senescent tumor cells secrete CCL20 to favor immunosuppressive M2 polarization. Also, senescent phenotype depends on proteasome function, suggesting that high-risk, high-senescence patients may benefit from proteasome inhibitors, which reverse senescence-mediated resistance to conventional chemotherapy and improve outcome. In conclusion, the current study identified senescence as a tumor-specific, hazardous factor associated with immunosuppression in PDAC. Mechanistically, senescence abrogates complement-induced M1 activation and antigen presentation, and upregulates CCL20 to favor M2 polarization. The senescence-related risk model is prognostic and therapeutic-suggestive. In light of the reliance of senescent cells on proteasomal functions, proteasome inhibitors are promising agents for high-risk patients with senescent PDAC.