Back

Targeting glucocorticoid-induced CD20 activation in preclinical models of B-ALL

Bigot, J.; Bouttier, M.; Fregona, V.; Rouzier, C.; Bayet, M.; Hebrard, S.; Prade, N.; Lagarde, S.; Didier, C.; Largeaud, L.; Vergez, F.; Quillet-Mary, A.; Ysebaert, L.; Broccardo, C.; Baruchel, A.; Strullu, M.; Clappier, E.; Pasquet, M.; Lainey, E.; Delabesse, E.; Gerby, B.

2025-12-18 cancer biology
10.64898/2025.12.16.694550 bioRxiv
Show abstract

Pediatric B-cell acute lymphoblastic leukemia (B-ALL) is effectively controlled with contemporary multi-agent chemotherapy, resulting to 5-year survival rates above 90%. However, relapse occurs in 15-20% of patients due to minimal residual disease (MRD), characterized by the presence of persisting and resistant leukemic cells, and associated with a poor clinical outcome. Despite its prognostic relevance, the molecular features driving MRD are poorly characterized. In this study, we developed patient-derived xenograft (PDX) models from matched diagnosis and relapse B-ALL samples combined to chemotherapy to mimic MRD in vivo. Drug-tolerant leukemic cells were profiled using single-cell RNA sequencing and we identified a transcriptionally distinct MRD-like population enriched for cell-quiescence, inflammatory stress, and B-cell receptor pathway signatures. Strikingly, the B-lymphocyte surface antigen CD20, encoding by MS4A1 gene, emerged as a consistent upregulated marker in MRD cells from PDXs and patients with diverse oncogenic subtypes. We further demonstrated that CD20 expression is induced by glucocorticoid exposure, creating a therapeutic opportunity where anti-CD20 monoclonal antibodies selectively eradicated MRD cells in vivo. Our data highlight CD20 not only as a biomarker but as an actionable vulnerability in B-ALL MRD, supporting clinical evaluation of anti-CD20 immunotherapy during induction treatment to kill drug-resistant cells and reduce relapse risk.

Matching journals

The top 6 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.