Abstract
INTRODUCTION: Inflammation and malnutrition adversely impact outcomes in patients with various malignancies. Composite indices such as the C-reactive protein/albumin ratio (CAR), the CRP × fibrinogen/albumin ratio (CFA), and the modified Glasgow Prognostic Score (mGPS) integrate these parameters, although their prognostic role in T-cell acute lymphoblastic leukaemia (T-ALL) remains underexplored. METHODS: In this single-centre retrospective study, 74 adults with T-ALL were included. CAR, CFA, and mGPS were calculated at diagnosis. Receiver operating characteristic curve analysis revealed the optimal cut-off values for the CAR (0.387) and CFA (0.396). Patients were stratified into low- and high-risk groups. Endpoints included rates of complete remission/complete remission with incomplete haematologic recovery (CR/CRi) at end-of-induction (EOI), minimal residual disease (MRD), overall survival (OS), and progression-free survival (PFS). RESULTS: Patients with low CAR, low CFA, or mGPS0 achieved significantly higher rates of CR/CRi (all p < 0.05) and MRD < 0.1% (all p < 0.05) at EOI. These low-risk groups also exhibited significantly fewer chemotherapy cycles to achieve the first CR/CRi (all p < 0.001) and shorter time to achieve MRD negativity (all p < 0.001). Survival analysis revealed significantly longer OS and PFS in the low-risk group (all p < 0.05). Multivariate analysis revealed high CAR (p = 0.004) and MRD positivity ≥0.1% at EOI (p = 0.043) as independent predictors of poor OS. Subgroup analysis indicated that allogeneic hematopoietic stem cell transplantation significantly improved survival only in high-risk patients. CONCLUSION: Pretreatment CAR, CFA, and mGPS are robust, accessible prognostic biomarkers in adults with T-ALL. Their integration into initial risk assessment could help guide personalized treatment strategies, including the identification of high-risk patients who may derive greater benefit from aggressive interventions.