Worth of pan-immune-inflammation value in trismus prediction after concurrent chemoradiotherapy for nasopharyngeal carcinomas.

Abstract

Radiation-induced trismus (RIT), one of the rare but serious side effects of concurrent chemoradiotherapy (C-CRT), is difficult to predict with high accuracy. We aimed to examine whether the pretreatment pan-immune-inflammation value (PIV) measures predict RIT in patients with locally advanced nasopharyngeal carcinoma (LA-NPC) receiving C-CRT. Data of patients with LA-NPC who underwent C-CRT and had maximum mouth openings (MMO) > 35 mm were reviewed. Any MMO of 35 mm or less after C-CRT was considered RIT. All PIV values were computed using the complete blood count test results: PIV = (Platelets × Monocytes × Neutrophils) ÷ Lymphocytes. The receiver operating characteristic analysis was employed to dissect a possible association between pre-treatment PIV readings and RIT status. Confounding variables were tested for their independent relationship with the RIT rates using logistic regression analysis. The research comprised 223 participants, and RIT was diagnosed in 46 (20.6%) at a median time from C-CRT to RIT of 10 months (range: 5-18 months). Pre-C-CRT PIV levels and RIT rates were analyzed using receiver operating characteristic curve analysis, with 830 being the optimal cutoff (area under the curve: 92.1%; sensitivity: 87.5%; specificity: 85.5%; Youden index: 0.730). RIT was significantly more prevalent in the PIV > 830 cohort than its PIV ≤ 830 counterpart (60.3% vs. 5%; hazard ratio 5.79; P < 0.001). Multivariate logistic regression analysis revealed that advanced T-stage (P = 0.004), masticatory apparatus dose V58Gy≥%32 (P = 0.003), and PIV > 830 (P < 0.001) were independently linked with significantly elevated rates of RIT. The presence of elevated pre-C-CRT PIV is a unique biological marker that independently predicts increased RIT rates in LA-NPC undergoing C-CRT.