Back

Mechanophenotype of synovial fibroblasts correlates with patients' pathotypes and treatment response in Rheumatoid Arthritis

Diaz, F.; Debatin, T.; Messner, L.; Mouchel dit Grainville, C.; Andres, J.; Riethmuller, F.; Vaishampayan Vikrant, R.; Fronza, R.; Taheri, F.; Wang, Z.; Selhuber-Unkel, C.; Colombo, F.

2026-10-06 biophysics
10.64898/2026.10.01.755901 bioRxiv
Show abstract

The mechanical features of synovial fibroblasts may encode clinically relevant information for the management of rheumatoid arthritis. Indeed, it remains poorly understood the links between cellular and microenvironment mechanics with patient stratification and therapeutic response. Here we show that combining organotypic synovial cultures with multimodal mechanophenotyping and transcriptomics identifies distinct ''soft'' and ''stiff'' fibroblast states. Healthy and rheumatoid arthritis fibroblasts were cultured in two or three dimensions (organotypic cell cultures), alone or with endothelial cells, and characterized by high-throughput nanoindentation, picobalance measurements, digital holographic microscopy, Brillouin microscopy and RNA sequencing. Three-dimensional culture reduced single-cell Young's modulus and total mass while increasing dry-mass surface density; endothelial co-culture increased fibroblast stiffness and the longitudinal modulus of intact organotypic cultures. Stiffness-based grouping revealed transcriptional programmes related to extracellular-matrix remodelling, cytoskeletal contractility, and TGF-beta, WNT, and Hippo signalling. Mapping these signatures to publicly available rheumatoid arthritis clinical transcriptomic datasets associated the ''stiff'' state with fibrotic and myeloid-fibrotic pathotypes and treatment-refractory programmes, whereas the ''soft'' state aligned with lymphoid pathotypes and rituximab- or tocilizumab-responsive programmes. These findings support mechanophenotype-associated transcriptional signatures as potential biomarkers for rheumatoid arthritis stratification.

Matching journals

The top 1 journal accounts 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.