OBJECTIVES: Approximately 30% of DLBCL patients are older than 70 yr. This study evaluated the prognostic impact of a cachexia score (CS) including fat tissue loss (adipopenia) and sarcopenia as assessed by computed tomography (CT scan) in elderly DLBCL patients treated with chemotherapy and rituximab (R). METHODS: This retrospective analysis included 80 DLBCL patients older than 70 yr treated with R-CHOP or R-miniCHOP. Skeletal muscle (SM) and visceral (V) and subcutaneous (S) adipose (A) tissues were measured by analysing CT images at the third lumbar (L3) level. RESULTS: The median age of the patients was 78 yr. Forty-four and 46 patients were considered sarcopenic and adipopenic, respectively. The median progression-free survival (PFS) was 13.6 months in the adipopenic group and 49.4 months in the non-adipopenic group [hazard ratio (HR) = 2.27; 95% confidence interval (CI): 1.3-4; P = 0.0042]. The median overall survival (OS) was 25.7 months in the adipopenic group and 57.1 months in the non-adipopenic group (HR = 1.93; 95% CI: 1.05-3.55; P = 0.0342). A two-point CS including adipopenia and sarcopenia was created and defined two distinct risk groups with differences in outcomes that were highly significant. The CS was predictive of the prognosis in a multivariate analysis including body mass index (BMI) (< or ≥ 25 kg/m(2) ), age (< or ≥ 80 yr), international prognostic index (IPI) and albuminaemia (HR = 3.67; 95% CI = 1.93-6.97; P < 0.0001). CONCLUSION: A CS including sarcopenia and adipopenia, assessed by a single CT scan slice, predicts outcome independent of BMI and the IPI.