Prognostic nutritional index as a predictor of prognosis in patients with diffuse large B cell lymphoma

The prognostic nutritional index (PNI), an indicator of nutritional status and systemic inflammation, is associated with short-term and long-term outcomes of various malignancies. The prognostic value of PNI in diffuse large B cell lymphoma (DLBCL) remains unknown. The aim of the present study was to determine the prognostic value of baseline PNI in DLBCL patients.

We retrospectively analyzed data from 103 DLBCL patients treated with R‑CHOP or R‑CHOP-like regimens. We evaluated the significance of PNI as a predictor of response to treatment, overall survival (OS) and event-free survival (EFS).

Patients with a PNI ≤ 44.55, where the cut-off was calculated by receiver operating characteristics (Youden index) and the same was obtained for response to treatment with 76.2 % sensitivity and a specificity of 85.4 %, for OS with 72.4 % sensitivity and a specificity of 90.5 % and for EFS with 65.6 % sensitivity and a specificity of 90.1 %, had significantly worse 5‑year OS (18.3 % vs 86.4 %, P < 0.001, log rank test) and 5‑year EFS (15.1 % vs 82.3 %, P < 0.001, log rank test). Regression analysis showed that PNI ≤ 44.55 was an independent prognostic factor for response to treatment with an odds ratio (OR) of 4.88 for treatment failure, 95 % confidence interval (CI) 1.077-22.105, OS hazard ratio (HR) 4.24, 95 % CI 1.451-12.392 and EFS HR 4.007, 95 % CI 1.48-10.852. Lower PNI levels were found in patients with advanced Ann Arbor clinical stage (46.6 ± 7.77 vs. 52.7 ± 5.43) and in those with poor response to therapy (40.58 ± 7.26 vs. 50.67 ± 6.26).

The PNI is a simple and useful marker to predict long-term survival outcome in DLBCL patients. Low PNI predicted poor outcome. A limitation of the study is its retrospective design in which the prognostic value was tested in the derivation cohort only. Notwithstanding, this is the first study suggesting that PNI is an important prognostic factor in DLBCL.