Treatment-induced senescence in metastatic colorectal adenocarcinomas: implications for single-cell biology and therapeutic intervention

In cancers, the harmful effect of senescence is now well recognized and mediated by the local production of pro-inflammatory cytokines, soluble immunosuppressive factors and metalloproteinases, called ‘SASP' phenotype. Many chemotherapy and radiation regimens can induce senescence of tumor cells and a deleterious microenvironment that promote tumor growth, invasion and neoangiogenesis leading to treatment resistance, relapse and poor prognosis, especially at advanced stages. Senescence biomarkers, such as dipeptidyl-peptidase 4 (DPP4/CD26), a membrane glycoprotein involved in chemotactic and inflammatory responses, is frequently constitutively expressed and/or overexpressed on primary tumors and metastases, including in colorectal adenocarcinoma (mCRC). A study aimed at dissecting the pathophysiology of senescence induced by standard of care (SoC) treatments is currently underway on a large series of mCRC, as well as other digestive cancers and hepatic carcinomas. The pre- and post-SOC scoring of DPP4 expression is assessed by immunohistochemistry and will be correlated to clinical annotations and other senescence features on primary tumors and liver metastases (n=100 samples). A comprehensive bioinformatic study is conducted on a large pan-tumor bulk RNA sequencing database (n=1155 samples) assessing the expression of a broad panel of candidate genes and signatures relevant to senescence pathways, including DPP4, and signaling signatures of interest from MSigD, CellMarker2.0, panglaoDB and SenNet. The same evaluations are conducted on a large set of single cell RNA sequencing (scRNASeq) data from different mCRC cohorts (n=105 patients) in which pre- and post-SoC treatment samples were collected on treatment-naïve primary tumor or liver metastatic samples (n=72) and post-SoC biopsies (n=54), including MSI-high tumors (n=16). Bioinformatics pipelines were designed to address correlations of candidate genes and molecular signatures to cellular heterogenicity, context, biology and clinical outcomes. To our knowledge, this study will be the first to highlight at the protein and scRNASeq levels the deleterious impact of senescence in a large series of patients with mCRC during the progression and treatments of cancer. This could have important implications for the development of innovative therapies targeting senescence biomarkers and related signaling pathways for cancer cell elimination in refractory tumors.