Matrix stiffness shapes transcriptional profiles and drug responses of pancreatic cancer cells

Pancreatic cancer tissues are made of different cell populations surrounded by a dense extracellular matrix. The stiff cancerous matrix impairs the diffusion of cytotoxic drugs, contributing to poor outcomes for patients. Matrix-targeting therapies normalise the extracellular matrix and, therefore, cell-matrix interactions. However, our knowledge of how matrix stiffness influences cancer cell and transcriptional dynamics and responses to anti-cancer compounds is incomplete. Here we developed a 3D cancer model to replicate the stiffness of patient-derived tissues and evaluated the effects of matrix-targeting compounds. Transcriptomic analyses showed that matrix stiffness regulated matrisome-related genes, cytokines and chemokines. The inclusion of stromal cells further increased the mechanical properties of our 3D cancer model. Treatment with the ROCK inhibitor fasudil induced matrix softening and improved sensitivity of cancer and stromal cells to cytotoxic treatment, whereas inhibition of matrix metalloproteinases disrupted cancer cell invasion. Our results indicate that matrix stiffness impacts cancer cell profiles, and targeting the cancerous matrix may lead to improved combination therapies for pancreatic cancer. STATEMENT OF SIGNIFICANCE: Pancreatic cancers have a dense extracellular matrix that drives resistance to therapy and disease progression, yet the mechanistic links between matrix stiffness and cancer cell behaviour are unclear. We engineered a 3D cancer model based on mechanical profiling of patient tissues to investigate how matrix stiffness influences gene expression and drug response. We found that increased stiffness regulates matrisome and inflammatory genes, and that stromal-induced stiffening affects sensitivity to matrix-targeting and cytotoxic drugs. The inhibition of Rho-kinase signalling enhanced the efficacy of cytotoxic drugs. These findings establish matrix stiffness as a key factor in tumour biology and therapy response, supporting the use of our preclinical model to guide new combination treatments for pancreatic cancer, aimed at improving outcomes for patients.