Nomogram for predicting survival of patients with diffuse large B-cell lymphoma.

ABSTRACT

The international prognostic index (IPI) system has been widely used to predict prognosis in diffuse large B-cell lymphoma (DLBCL). However, this system categorizes DLBCL patients into four risk groups, and cannot optimize individualized prognosis. In addition, other clinicopathological factors, such as molecular aberrations, are not incorporated into the system. To partly overcome these weak points, we developed nomograms to predict individual patient survival. We also incorporated MYD88L265P and CD79BY196 mutations into the nomograms since these mutations are associated with a worse prognosis and their signaling pathways have been highlighted as a therapeutic target. We analyzed 302 DLBCL cases for which multivariate analysis by Cox proportional hazard regression was performed. Nomograms for progression-free survival (PFS) and overall survival (OS) were constructed and assessed by a concordance index (C-index). The nomograms were also evaluated using an open external dataset (n = 187). The MYD88L265P and/or CD79BY196 (MYD88/CD79B) mutation was detected in 62/302 patients. The nomograms incorporating IPI factors exhibited a C-index of 0.738 for PFS and a C-index of 0.765 for OS. The nomograms incorporating IPI factors and the MYD88/CD79B mutation showed a C-index of 0.745 for PFS and a C-index of 0.769 for OS. The nomograms we created were evaluated using an external dataset and were well validated. The present nomograms incorporating IPI factors and the MYD88/CD79B mutation have sufficient discrimination ability, and may effectively predict prognosis in DLBCL patients. The prognostic models we have presented here may help clinicians personalize prognostic assessments and clinical decisions.