Low Pre-Chemoradiotherapy Pan-Immune-Inflammation Value (PIV) Measures Predict Better Survival Outcomes in Locally Advanced Pancreatic Adenocarcinomas

Abstract

Objective: This study sought to determine whether pretreatment pan-immune-inflammation value (PIV) could be used to predict prognosis in patients with locally advanced pancreatic adenocarcinoma (LA-PAC) following definitive concurrent chemoradiotherapy (C-CRT). Methods: The outcomes of 178 LA-PAC patients who received definitive C-CRT were analyzed retrospectively. For all patients, the PIV was calculated using the peripheral blood platelet (P), monocyte (M), neutrophil (N), and lymphocyte (L) counts obtained on the first day of C-CRT: PIV=PxMxN divided by L. The optimum cutoff values for PIV connected to progression-free (PFS) and overall survival (OS) results were sought using receiver operating characteristic (ROC) curve analysis. The OS and PFS differences between the PIV groups constituted the primary and secondary endpoints, respectively. Results: ROC curve analysis indicated that the ideal PIV cutoff was 464 (AUC: 75.9%, sensitivity: 74.1%, specificity: 71.9%), which categorized patients into two groups based on PFS and OS results: low PIV (L-PIV; N = 69) and high PIV (H-PIV; N = 109). According to comparative survival analyses, patients in the L-PIV group had significantly longer median PFS (14.3 vs 7.3 months; HR: 3.04; P < 0.001) and OS (25.9 vs 13.3 months; HR: 2.86; P < 0.001) than those in the H-PIV group. Although none of the H-PIV patients could survive beyond 5 years, the estimated 5-year OS rate was 29.7% in the L-PIV cohort. In multivariate analyses, besides the L-PIV, N0 nodal stage, and CA 19-9 <= 90 U/mL appeared to be the independent predictors of better PFS (P < 0.05 for each) and OS (P < 0.05 for each) results. Conclusion: The present results indicated that pre-C-CRT L-PIV measures were associated with favorable median and long-term PFS and OS results in LA-PAC patients, suggesting that the PIV is a potent and independent novel prognostic biomarker.