Tumour-stroma signaling drives a reversible program with neural-like features in cancer-associated fibroblasts

Pancreatic ductal adenocarcinoma (PDAC) is a particularly aggressive cancer, largely due to a dense and adaptable tumor microenvironment primarily composed of cancer-associated fibroblasts (CAFs). Previous research identified a population of CAFs expressing a neural-associated program (NES+SOX4+NOTCH1+) that intensified with disease progression, but the signaling mechanisms behind this state were not well understood.

This study aimed to investigate how tumor-derived signals activate this program and determine if it represents a stable fibroblast identity or a reversible one.

By analyzing spatial transcriptomic data from normal pancreas to invasive carcinoma, researchers identified EGF, FGF2, and BDNF as potential tumor-to-fibroblast signals correlated with disease progression. Functional experiments with patient-derived CAFs and human fibroblast cell lines confirmed that EGF and FGF2 alone are sufficient to induce the NES+SOX4+NOTCH1+ program.

Upon withdrawal of these growth factors, the program was reversed at both the protein and transcriptional levels, indicating that this state is dynamic and contingent upon the environment.

Further single-cell transcriptomic analysis uncovered a multi-phase response, starting with an early loss of fibroblast-associated features, followed by a transient activation of neural-associated genes, and concluding with a later stress-associated state. Regulatory analyses pinpointed ETS-family transcription factors as key regulators of the program, while phosphoproteomic profiling and inhibitor studies revealed that MAPK/ERK signaling plays a central role in its transcriptional activation.

Most intriguingly, pan-cancer single-cell analysis demonstrated that this reversible program is enriched in CAFs and becomes more prevalent from normal to tumor tissue across various cancer types. This study not only defines a dynamic, growth factor-driven fibroblast program but also identifies signaling mechanisms that could potentially allow for the selective modulation of fibroblast states that support tumor growth.