Functional omics in high-risk T-cell acute lymphoblastic leukemia.

Introduction. Acute lymphoblastic leukemia (ALL) has seen improvements in prognosis over the last decades. However certain subgroups like T-ALL, still present an unmet clinical need due to their poor prognosis and the absence of alternative therapies (tx) for relapsed/refractory (RR) patients (pts). A promising strategy involves functional precision medicine (FPM), which relies on information obtained through ex-vivo chemogenomic profiling. Our study aims to develop a chemogenomic FPM platform for identifying new actionable targets and guiding tx. Methods. We enrolled pts with new diagnoses or RR T-ALL. Leukemic cells were isolated from either bone marrow or peripheral blood samples, and pleural effusion was used for pts with mediastinal bulky disease. We used a library of 77 molecules for drug response profiling (DRP). The compounds were dispensed in duplicate at 4 different concentrations, and cell viability was assessed after 72 hours of incubation. To identify resistance and sensitivity profiles, we used the drug sensitivity score (DSS), which is calculated based on the half-maximal inhibitory concentration and the area under the curve. The inter-individual analysis was based on the difference between each pt’s DSS and the median DSS of the entire cohort (dDSS) for each drug. For each pt we collected data on their genomic profile through a Combined Interphase FISH and a customized panel of recurrently mutated genes used for Next Generation Sequencing. Results. We collected a total of 79 samples from 76 pts. 27 pts participated in the GIMEMA trial ALL2720 (NCT04582487), and the DRP was assessable in 94% of cases. Genomic and functional characterization was conducted for 45 pts, including 3 individuals who experienced 2 different relapses, resulting in a total of 48 fully characterized samples (Table 1). Based on the DRP, we identified 3 response patterns according to dDSS. Pts exhibiting the highest sensitivity were subjected to an agglomerative clustering approach. Cluster1 (C1) mainly consisted of immature T-ALL with chemoresistance and sensitivity to antiapoptotic and kinase inhibitors, whereas C2 and C3 primarily featured a mature phenotype, with medium to high chemosensitivity and increasing resistance to kinase inhibitors (Figure 3). A significant difference in the expression of gene mutations and cytogenetic aberrations was observed when comparing C1 and C2 (Figure 4). From a translational perspective, 13 pts underwent tx guided by DRP. Of these, 76% achieved a response, including 9 CR, 5 of which achieved MRD negativity, and 1 PR. 6 responders successfully proceeded to allogenic hematopoietic transplantation (Figure 4). Conclusions Our data indicate that FMP is an innovative strategy for identifying new response patterns and druggable targets in highrisk T-ALL. Additionally, the chemogenomic approach represents a promising tool for guiding tx in these pts, particularly for achieving a deep response in transplant-eligible individuals.