Abstract
BACKGROUND: CD5-positive diffuse large B-cell lymphoma (CD5+ DLBCL) is a rare subtype of DLBCL with invasive clinical features and poor prognosis. Current clinical variables based on prognostic systems for DLBCL are inadequate to accurately stratify the prognosis of CD5+ DLBCL. METHODS: A total of 195 CD5+ DLBCL patients were retrospectively recruited from nine centers in Huaihai Lymphoma Working Group. MaxStat analysis was used to identify optimal cutoff points for continuous variables; univariable and multivariable Cox analyses were used for variable selection; Kaplan-Meier curve was used to analyze the value of variables on prognosis; and C-index, Brier score, and decision curve analysis were measured for predicting model performance. RESULTS: The derivation and validation cohorts consisted of 131 and 64 patients. Of the whole cohort, median age at diagnosis was 61 years, of whom 100 (51.28%) were males and the 5-year overall survival rate was 42.1%. MYC, BCL-2, and the coexpression of MYC/BCL-2 could distinguish the survival of CD5+ DLBCL. Multivariable analysis showed that age, IPI, red blood cell count, neutrophil count, MYC expression, and hepatosplenomegaly were independent predictors, and the prognostic nomogram was developed. The C-index of the nomogram was 0.809 in the derivation and 0.770 in the validation cohort. Decision curve analysis proved that compared with IPI, the specific nomogram showed a better identification in CD5+ DLBCL. CONCLUSION: The proposed nomogram provided a valuable tool for prognosis prediction in patients with CD5+ DLBCL.