We previously reported a better outcome in adult and pediatric T-cell acute lymphoblastic leukemia (T-ALL) harboring NOTCH1 and/or FBXW7 mutations without alterations of K-N-RAS and PTEN genes. Availability of high-throughput next-generation sequencing strategies (NGS) led us to refine the outcome prediction in T-ALL. Targeted whole-exome sequencing of 72 T-ALL related oncogenes was performed in 198 adult T-ALLs in first remission (CR1) from the GRAALL-2003/2005 protocols (ClinicalTrial.gov, NCT00222027, NCT00327678) and 242 pediatric T-ALLs from the FRALLE2000T. This approach enabled the identification of the first NGS-based classifier in T-ALL categorizing low-risk patients as those with N/F, PHF6, or EP300 mutations, excluding N-K-RAS, PI3K pathway (PTEN, PIK3CA, and PIK3R1), TP53, DNMT3A, IDH1/2, and IKZF1 alterations, with a 5-year cumulative incidence of relapse (CIR) estimated at 21%. Conversely, the remaining patients were classified as high-risk, exhibiting a 5-year CIR estimated at 47%. We externally validated this stratification in the pediatric cohort. NGS-based classifier was highly prognostic, independently of minimal residual disease (MRD) and white blood cells counts (WBC), in both adult and pediatric cohorts. Integration of the NGS-based classifier into a comprehensive risk stratification model, including WBC count at diagnosis and MRD at the end of induction, enabled the identification of an adverse risk subgroup (25%) with a 5-year CIR estimated at 51%, and a favorable risk group (32%) with a 5-year CIR estimated at 12%. NGS-based stratification combined with WBC and MRD sharpens the prognostic classification in T-ALL and identifies a new subgroup of patients who may benefit from innovative therapeutic approaches.