Differential expression of cancer-related genes supports prediction of poor response to first-line treatments in T-ALL pediatric patients with high minimal residual disease.

T-cell acute lymphoblastic leukemia (T-ALL) originates from the malignant transformation of immature lymphoblasts committed to the T-cell lineage. Relapsed or refractory T-ALL patients show a dismal outcome with limited therapeutic options and cure rates below 10%. Accurate risk stratification is essential for optimizing first-line treatments and maximizing initial complete response rates. Minimal residual disease (MRD) assessment is the most relevant clinical parameter in T-ALL and a direct measure of treatment response, but it requires an initial treatment course that reduces the time for decision-making. Consequently, there is an urgent need to identify novel biomarkers that can predict the response to first-line treatments at diagnosis. By integrating clinical and transcriptomic data from diagnostic T-ALL samples, we found that high MRD patients show a specific transcriptional profile. Moreover, we identified a transcriptional signature characterized by the differential expression of HSH2D, LAT2, BCL2, MAST4, METRN, and PITPNM2 genes that is tightly associated with an increased MRD, which could improve the prediction of poor treatment response in T-ALL patients, especially during early treatment phases.