Yan Leyfman, Medical Oncologist, Co-Founder and Executive Director of MedNews Week, shared a post on LinkedIn about a recent article by Michelle Wang et al, published in Blood Advances:
“New development in DLBCL care! A recently published real-world multicenter study reported in Blood Advances demonstrates that an AI/ML decision tree model can stratify patients by risk of early relapse within 6 months after Axicabtagene ciloleucel (axi-cel) therapy.
Using only age plus six routine lab values (LDH, CRP, ferritin, hematocrit, platelet count, prothrombin time), the model achieved an area under the ROC of 0.82 — a strong discriminator for early relapse.
Why it matters: identifying high-risk patients early may help tailor post-CAR-T interventions, optimize monitoring, or prompt early consolidation — ultimately aiming to improve survival outcomes in real-world DLBCL.
As we move toward precision-guided cellular therapies, predictive tools like this could become part of routine practice.”
Title: An AI model classifies risks of early relapse post–CAR T-cell therapy in a multicenter real-world population with DLBCL
Authors: Michelle Wang, Krishna V. Komanduri, Debajyoti Datta, Ayan Patel, Barbee Whitaker, Artur Belov, Benjamin Rubin, Rohit Vashist, Rosa Rodriguez-Monguio, Pelin Cinar, Steven Anderson, Atul J. Butte
Read the Full Article on Blood Advances
More posts featuring Yan Leyfman