Researchers highlight stromal control of early malignancy and identify HSP47 as a reversible driver of tissue stiffening

Keynote speaker Dr. Tilstery told attendees that evidence accumulated over decades points to stromal and extracellular‑matrix signals as required partners in many cancers, not merely reactive bystanders. She reviewed classic and recent experiments showing that tissues bearing oncogenic mutations often remain phenotypically normal until a change in the microenvironment—injury, chronic inflammation or altered fibroblast state—provokes progression to metaplasia and cancer.

The speaker summarized three lines of evidence. First, deep sequencing and lineage studies show that normal‑appearing tissues often harbor cells with oncogenic mutations without frank cancer. Second, classical animal experiments (viral carcinogenesis and injury models) demonstrate that tissue injury and inflammation convert otherwise tolerant tissues into loci of tumor formation. Third, reconstitution experiments replacing normal stroma with carcinoma‑associated fibroblasts (CAFs) convert indolent epithelial cells into proliferative, disordered structures resembling premalignant lesions.

The talk focused on a candidate stromal regulator, HSP47, a collagen chaperone that the presenter said is upregulated in a subset of tumor‑associated fibroblasts across multiple cancer types. She described proteomic and single‑cell transcriptional evidence showing higher HSP47 expression in fibroblast subpopulations that induce squamous metaplasia in bronchial models. In functional tests she said that knocking down HSP47 in fibroblasts eliminated their ability to induce metaplasia, whereas overexpressing HSP47 converted non‑inducing fibroblasts into potent inducers. “The expression of HSP 47 is sufficient and the expression of HSP 47 is necessary in order to induce the beginning, the initiation and progression of these types of cancers,” the speaker stated during the presentation.

She also described a small‑molecule HSP47 inhibitor that, in the presenter’s models, blocked induction of metaplasia and—critically—reversed already established pre‑malignant epithelial structures back toward a differentiated state when applied after the abnormal tissue architecture had formed. The speaker argued these data implicate mechanical transduction (via increased collagen deposition and tissue stiffening) and YAP/TAZ nuclear signaling as central mechanisms linking stromal rewiring to epithelial reprogramming.

The speaker framed translational implications cautiously: if stromal programs that promote malignancy are activated early and are reversible, they could become targets for interception. In audience questions she noted that chronic inflammation is a time‑dependent risk factor—acute inflammation does not have the same effect—and suggested that combining epithelial‑targeted therapies with approaches that normalize the matrix or blunt mechanotransduction could change outcomes. She also noted that some stress‑based diagnostic tests (discussed in a later Q&A) are already in clinical use for particular lesions.

The presenter repeatedly emphasized that the stromal influence is context‑dependent: some ECM compositions and stromal signals repress malignancy and can re‑normalize genetically aberrant epithelial cells, while others promote progression. She concluded by urging the field to consider both top‑down observational studies and bottom‑up synthetic perturbations to identify which stromal components are necessary and sufficient for malignant progression.