Back

Memory CD4 T cells orchestrate neoadjuvant-responsive niches in colorectal cancer liver metastases

Mayoux, M.; Litscher, G.; Wiesmann, F.; Vermeer, M.; Jorba Adolff, P.; Guido Vinzoni, E.; Gonzalo Nunez, N.; Sparano, C.; Vetter, V. K.; Friebel, E.; Wertheimer, T.; Ficht, X.; Burr, S.; Karimi, A.; Dergun, S.; Schmid, J.; Krishnarajah, S.; Mair, F.; Moor, A.; Burkhard, B.; Sobottka, B.; Tugues, S.

2025-11-16 immunology
10.1101/2025.08.22.671704 bioRxiv
Show abstract

Colorectal cancer frequently progresses to liver metastases (CRLM), a stage with limited treatment options and poor prognosis. Neoadjuvant chemotherapy is used to control tumor growth and enable resection, yet many patients fail to respond, and the mechanisms underlying this variability remain unclear. To identify determinants of treatment response, we profiled T cell states and their spatial organization in CRLM. We found that spatial arrangement and polarization of CD4 memory T cell networks determine treatment outcome. In responders, Th1-like CD4 memory T cells organized with effector-memory CD8 T cells and antigen-presenting cells (APCs) into therapy-responsive immune niches (TRINs) that support CD4-mediated APC licensing and local immune engagement. Non-responders lacked such immune architecture, exhibiting myeloid-rich regions dominated by circulating-like CD4 memory and regulatory T cells. CD4-driven TRINs thus emerge as key determinants of chemotherapy efficacy and provide a rationale for developing biomarkers and strategies that enhance CD4-APC-CD8 crosstalk within organized immune niches. Statement of significanceTh1-polarized CD4 memory T cells form therapy-responsive immune niches (TRINs) that orchestrate CD4, CD8, and APC function in colorectal cancer liver metastases, a clinically challenging and immunologically cold tumor type. TRINs define chemotherapy response and provide a mechanistic foundation for biomarker development and immunotherapy strategies designed to restore anti-tumor immunity.

Matching journals

The top 5 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.