Back

Enhancing Patient Lymphocyte Response to Peritoneal Malignancies Using a Personalized Immunocompetent Microfluidic Co-Culture Platform

Schaaf, C. R.; Hutchins, D. R.; Liu, T.; Kooshki, M.; Wagner, C.; Edenhoffer, N.; Wajih, N.; Forsythe, S.; Greissinger, R.; Levine, E.; Shen, P.; Triozzi, P.; Miller, L. D.; Hall, A. R.; Soker, S.; Votanopoulos, K. I.

2026-02-20 cancer biology
10.64898/2026.02.19.706377 bioRxiv
Show abstract

Harnessing patient immune cells via adoptive cellular therapy is a promising cancer therapeutic strategy. However, major challenges remain for advanced solid malignancies, including difficulty isolating sufficient tumor infiltrating lymphocytes (TILs) and limited targeting of diverse neoantigens in heterogenous tumors. To address this, we have developed a tumor-on-a-chip platform with co-cultured patient-derived tumor cells, autologous peripheral blood mononuclear cells (PBMCs), and lymphoid tissue-derived antigen presenting cells. This approach generates organoid interacting lymphocytes (OILs) with enhanced anti-tumor activity. In peritoneal malignancies, OIL-induced cytotoxicity of patient-matched tumor cells surpasses both TILs and static-expanded PBMCs. We find that this improved performance was linked to increased CD8+ T and NK cells among OILs, and increased effector cytokine polyfunctionality, particularly Granzyme A. Our platform represents a versatile and scalable approach to generate patient-specific therapeutic lymphocytes even when TILs are insufficient, offering a promising avenue to treat diverse solid tumors associated with poor outcomes under current immunotherapies. TeaserA tumor-on-a-chip device primes patient immune cells with tumor recognition for personalized immunotherapy applications.

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.