Cancer-associated fibroblasts during and after therapy: disentangling state, function and persistence.

Cancer-associated fibroblasts (CAFs) influence extracellular matrix organisation, immune-cell positioning, tumour-cell survival and intratumoural drug distribution. Untreated-tumour studies alone cannot define CAF features during or after therapy; experimental models provide mechanistic evidence but cannot replace paired human tissue measurements. We conducted a focused narrative review of studies of treatment-exposed solid tumours, distinguishing paired human tissue measurements from unpaired associations, functional perturbations, post-withdrawal experiments and clinical studies. Paired specimens showed differences in stromal area, marker expression, mixed-tissue transcriptional signals and relative representation of study-defined CAF states; the observed patterns varied with the measured quantity, cancer and regimen. In defined models, fibroblast-restricted or CAF-side perturbations altered matched immune and treatment endpoints. Direct treatment of the extracellular matrix altered intratumoural drug access. Post-withdrawal evidence for persistent CAF function remained limited; residual or recurrent lesions represented later disease states without establishing persistence. Clinical CAF measures remained candidate response-associated markers or pharmacodynamic signals. Among the included clinical studies, we did not identify one that combined a stroma-directed intervention, CAF-resolved pharmacodynamics and a controlled treatment-effect estimate. Interpretation depends on tumour type, complete regimen, sampling time, specimen, assay, measured quantity and endpoint. Progress requires paired and serial human sampling, matched functional perturbation, independent assay validation, and trials linking stromal target engagement to CAF-resolved pharmacodynamics and patient outcomes.