Back

Clinical Relevance and Applicability of the 2022 World Health Organization Classification of Childhood B Lymphoblastic Leukemia in the Context of MRD-Directed Therapy

Rajpal, S.; Chatterjee, G.; Bhanshe, P.; Terse, V.; Joshi, S.; Chaudhary, S.; Shetty, D.; Mohanty, P.; Dhamne, C.; Tembhare, P.; Srinivasan, S.; Chichra, A.; Moulik, N. R.; Banavali, S.; Gujral, S.; Narula, G.; Subramanian, P. G.; Patkar, N.

2024-10-17 hematology
10.1101/2024.10.16.24315431 medRxiv
Show abstract

WHO5-2022 classification of B-lymphoblastic leukemia (B-ALL) incorporates several novel entities requiring high-throughput sequencing for their accurate characterization. The clinical relevance of this classification in the context of contemporary MRD-directed therapy is unclear. We analyzed 533 pediatric B-ALL uniformly treated with ICiCLe-ALL-14 protocol as defined by WHO2016 and reclassified them as per WHO5-2022 using targeted sequencing, FISH, and cytogenetics. Subtype-defining genetic abnormalities were identified in 81.2% of the cohort as per the WHO5 classification. Among the new subtypes, PAX5alt, MEF2D-r, and BCR::ABL1-like(ABL-class) were associated with an inferior 2-year event-free survival (EFS) of 39.1% (p<0.0001), 53.8% (p=0.024) and 60.6% (p=0.043), respectively. We developed a 3-tier genetic risk stratification model incorporating 15 genetic subtypes and the IKZF1 deletion. Children with standard, intermediate, and high genetic risk demonstrated 2-year EFS of 92.6%, 71.0%, and 50.7% (p<0.0001), and 2-year overall survival of 94.3%, 81.9%, and 71.6% (p<0.0001), respectively. Genetic risk further identified heterogeneous outcomes among ICiCLe risk groups (p<0.0001). Standard genetic risk was associated with superior OS and EFS irrespective of MRD status. We demonstrate the applicability of the WHO5 classification in routine practice and create a general framework for incorporating the WHO5 classification in risk-adapted therapy for childhood B-ALL.

Matching journals

The top 4 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.