Mechanics shape cell and tissue plasticity and maintain their homeostasis. In cancers, mechanical signals regulate cancer hallmarks via mechanotransduction pathways, such as proliferation, metastasis and metabolic reprogramming. However, comprehensive characterization of mechanotransduction pathway genes and their clinical relevance across different cancer types remains untouched. Herein, we systematically portrayed the alterations of mechanotransduction pathway genes across 31 cancer types using The Cancer Genome Atlas (TCGA) databases. All the cancer types could be categorized into 6 subtypes based upon the transcriptional pattern of mechanics pathway genes. Each subtype has its own unique molecular expression pattern, mutation landscapes, immune infiltrates, and patient clinical outcome. We further found that the responses of two subtypes of cancers, one with the optimal outcome and the other with the worst prognosis, to a classical mechanotherapeutic agent (Fasudil, RhoA/ROCK inhibitor) were totally different, indicating that our cancer stratification system based upon mechanotransduction pathway genes could inform clinical responses of patients to mechanotherapeutic agents. Collectively, our study provides a novel pan-cancer landscape of the mechanotransduction pathways and underscores its potential clinical significance in the prediction of clinical prognosis and therapeutic responses to mechanotherapy among cancer patients.