ALL-268 genetic classification of B-Cell precursor adult acute lymphoblastic leukemia patients enrolled in LAL19 trial from the pethema group: response to treatment and survival

Abstract

Context: B-cell precursor acute lymphoblastic leukemia (BCP ALL) is a genetically heterogeneous neoplasm with >20 biologic subtypes. Each subtype shows specific genetic traits that determine relapse risk and patients' survival. Objectives: To establish the genetic subtype (primary alteration) of adult BCP ALL patients enrolled in the PETHEMA LAL19 trial (NCT 04179929) and to correlate them with measurable residual disease (MRD) level and survival. Patients and Methods: In the LAL19 trial (NCT04179929), Ph-negative patients (18–65 y) with MRD≥0.01% at day+35 or high-risk genetics receive alloHSCT and MRD<0.01% patients with standard-risk genetics receive maintenance chemotherapy. The genetic analyses are centralized: FISH and NGS DNA panel (Hospital de Salamanca), RNAseq panel (Hospital 12 de Octubre), FISH panel (Hospital La Fe), and SNP array (Josep Carreras Institute/ICO-Hospital Germans Trias i Pujol). MRD determinations are centrally done by next-generation flow cytometry in the Cytometry Service, NUCLEUS, University of Salamanca. Results: The genetic subtype was identified in 54% (82/152) of patients. The most recurrent subtypes were KMT2Ar (11%), Ph-like (mostly CRLF2::IGH, 11%), low-hypodiploid (7%), PAX5 P80R (7%), high-hyperdiploid (6%), and t(1;19)/TCF3::PBX1 (6%). In addition, t(12;21)/ETV6::RUNX1, ZNF384r, and iAMP21 subtypes (1.5% each) and MEF2Dr, MYCr, IDH1 R132 subtypes (<1% each) were found. Regarding secondary alterations, NRAS (15%), TP53 (13%), PAX5 (13%), and KRAS (10%) mutations were the most frequently observed. Twelve patients were refractory (mainly low-hypodiploid, Ph-like, MYCr, and B-other/unclassified patients). Statistically significant differences were observed for day+35 MRD levels between genetic subtypes. Ph-like, low-hypodiploid, and KMT2Ar showed lower frequencies of MRD<0.01% (17%, 33%, and 57%, respectively) than patients with PAX5P80R (100%), t(1;19)/TCF3::PBX1 (83%), and high-hyperdiploid (75%) (P=0.006). Despite the short median follow-up (11 months), differences in response to treatment were reflected in patients' survival. Significant differences in survival were observed between poor-response subtypes (Ph-like, KMT2Ar, and low-hypodiploid) and good-response subtypes (PAX5 P80R, t(1;19)/TCF3::PBX1, and high-hyperdiploid). Conclusions: Knowing the genetic subtype of each ALL is crucial to better predict relapse risk and offer the best (personalized) treatment for each patient.