Back

Fibroblast-derived Collagen VI shapes the structure and function of the tumor-immune microenvironment in clear cell renal cell carcinoma

Wess, M.; Andreev, G.; Feilen, T.; Diel, N.; Zirngibl, M.; Gueib-Picard, C.; Koessinger, A. L.; Hinrichs, C.; Naegel, A.; Essmann, C. L.; Roemer, W.; Helmstaedter, M.; Vollmer, T.; Werner, M.; Grabbert, M.; Schilling, O.; Rogg, M.; Schell, C.

2026-03-23 cancer biology
10.64898/2026.03.19.712351 bioRxiv
Show abstract

The importance of the extracellular matrix (ECM) influencing tumor biology in stroma-rich tumors is well established. However, the relevance of individual ECM proteins in rather stroma-poor cancers such as clear cell renal cell carcinoma (ccRCC) is ill-defined. Using bulk proteomics, spatial imaging, and single-cell transcriptomics, we identify collagen VI (COL6) as a predominant ECM component of the ccRCC interstitial stroma, synthesized primarily by fibroblasts and pericytes. Using cell-derived matrix (CDM) models, we demonstrate that COL6 is essential for maintaining an isotropic ECM network architecture and governs the broader matrisomal composition, with direct pro-proliferative consequences for tumor cells both in vitro and in situ. Granular spatial analysis reveals that COL6-rich stromal septa constrain tumor-infiltrating T cells to boundary zones, where CD8+PD1+ phenotypes predominate. Importantly, tyrosine kinase inhibition (TKI) with cabozantinib suppresses COL6 expression in fibroblasts in vitro and in ex vivo tumor models, mirroring COL6-depleted CDM phenotypes. Our findings establish COL6 as a central stromal regulator of ccRCC tumor biology and immune contexture, revealing ECM remodeling as an underappreciated mechanism of TKI action, with implications for combination immunotherapy strategies.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.