Back

Deep single-cell immune and signaling profiles predict long term therapy response in chronic myeloid leukemia within hours

Gullaksen, S.-E.; Brevik, M.; Sefland, O.; Craig, E.; Hellesoy, M.; Olsson-Strömberg, U.; Stentoft, J.; Richter, J.; Gedde-Dahl, T.; Tibshirani, R.; Plevritis, S.; Wolf, D.; Mustjoki, S.; Hjorth-Hansen, H.; Gjertsen, B. T.

2025-11-11 cancer biology
10.1101/2025.11.10.687542 bioRxiv
Show abstract

Chronic myeloid leukemia (CML) is effectively treated with small molecule BCR::ABL1 tyrosine kinase inhibitors (TKIs) but like in all cancers, early identification of suboptimal- and non-responders is a challenge. Through mass cytometry analysis of peripheral blood leukocytes, we collected high-dimensional data from de novo chronic phase CML patients enrolled in two multicenter clinical trials (clinicaltrials.gov NCT01725204 and NCT01061177). In leukocytes, dasatinib and nilotinib inhibited intracellular signaling within one or three hours after first per oral dose, and each TKI had a unique signaling signature reflecting its kinase specificity profile beyond BCR::ABL1. An immune and signaling profile was constructed for each patient, predicting the treatment response (BCR::ABL1IS, %) the first 12 months of treatment. These results show that single cell immune and signaling profiles within hours of first dose can predict 12 months treatment response, anticipating future optimalization of kinase inhibitor treatment within days rather than months.

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.