Abstract
Primary diffuse large B-cell lymphoma, leg type (PCDLBCL, LT) is a rare lymphoma with a poor prognosis. An accurate predictive model for the prognosis of these patients is lacking. Here, we retrospectively analyzed the clinical data of 968 adult patients who were diagnosed with PCDLBCL, LT between 2000 and 2020 obtained from the Surveillance, Epidemiology and End Results (SEER) database. Multivariate Cox regression analysis revealed that age, marital status and radiation were independent prognostic factors for overall survival (OS), while age, Ann Arbor stage and radiation were independent prognostic factors for disease-specific survival (DSS). Nomograms were constructed to predict 1-, 3- and 5-year OS and DSS. The concordance indices, receiver operating characteristic curves and calibration curves revealed that the established nomograms had good accuracy in predicting the outcomes of patients with PCDLBCL, LT. The web-based calculators were subsequently developed to provide individualized predictions for OS and DSS in patients with PCDLBCL, LT. Moreover, according to the risk scores of the nomograms, patients were divided into low-, median-, and high-risk groups. In conclusion, the nomograms and accompanying web-based survival calculators offer practical tools for individualized prediction of OS and DSS in patients with PCDLBCL, LT. Risk stratification based on these tools may assist clinicians in identifying high-risk patients and guiding optimal treatment decisions. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s00277-025-06692-5.