Clinical impact of prognostic nutritional index in diffuse large B cell lymphoma

Abstract

We evaluated the association between the prognostic nutritional index (PNI) and the clinical features of diffuse large B cell lymphoma (DLBCL) and developed a novel prognostic model using a nomogram including the PNI and other biomarkers for cancer cachexia. A total of 228 DLBCL patients treated with first-line R-CHOP (rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone) were retrospectively reviewed. PNI was calculated as 10xserum levels of albumin (g/dL)+0.005xabsolute lymphocyte count (/mm(3)). Patients were categorized into low- and high-PNI groups based on a cut-off value of 40. The nomogram for predicting overall survival (OS) was constructed using a Cox regression model. PNI was positively correlated with skeletal muscle index, body mass index, and serum levels of albumin. The low-PNI group had a lower complete response rate (60.3% vs. 87.6%), increased treatment-related toxicity, and more frequent treatment discontinuation (43.5% vs. 8.8%) than the high-PNI group. The median OS was shorter in the low-PNI group than the high-PNI group (15.6months vs. not reached; p<0.001). Multivariate Cox regression analyses showed that PNI, sarcopenia, and the international prognostic index (IPI) were independent prognostic factors for OS. The nomogram developed using this regression model showed excellent discriminatory ability for predicting OS (c-index, 0.80) compared to the IPI alone (c-index, 0.75). Low PNI was associated with adverse clinical features of DLBCL. The proposed nomogram supports the clinical impact of cachexia on survival and may contribute to individualized therapy in DLBCL.