Background Peripheral blood inflammatory markers are considered to be indicators for predicting the effectivenessof immunotherapy. However, there are few studies on the dynamic changes and combined prediction effect of each marker.
Objective To investigate the predictive value of single and combined application of dynamic changes in peripheral blood inflammatory markers for the effectiveness of immunotherapy combined with anti-angiogenesis therapy for advanced hepatocellular carcinoma (HCC).
Methods This study included 189 patients with advanced HCC who received the combination therapy of camrelizumab and apatinib. From baseline to 30 days after the last treatment, peripheral blood was detected every 2 weeks, and the dynamic changes of peripheral blood inflammatory markers during the whole treatment were calculated, such as neutrophil to lymphocyte ratio (NLR), platelet to lymphocyte ratio (PLR), prognostic nutritional index (PNI) and systemic immune-inflammation index (SII). In the time-dependent Cox regression model, the time-varying NLR, PLR, PNI and SII during treatment were defined as time-dependent NLR, PLR, PNI and SII, respectively, and the time-dependent mean was the mean value of each marker during treatment. The patients were divided into three groups based on efficacy, including complete response (CR)/partial response (PR) group, stable disease (SD) group, and progressive disease (PD) group. The correlation between the baseline values and the time-dependent mean values of each marker and the treatment effects was analyzed. The survival curve was drawn by Kaplan-Meier method. Univariate and multivariate Cox regression models, as well as time-dependent Cox regression models, were utilized to explore the related influencing factors of overall survival (OS). The concordance index (C-index) was used to evaluate the predictive efficacy of each inflammatory marker alone and in combination on OS.
Results Among the 189 patients, there were 169 males and 20 females, and 44 patients were older than 60 years old. There were 51, 95 and 38 patients in the CR/PR group, SD group, and PD group, respectively. The differences in time-dependent mean values of NLR (MIQR: 1.9 1.3-2.3 vs 2.2 1.6-3.2 vs 2.8 1.8-3.9, P＜0.001), PLR (MIQR: 87 67-104 vs 99 84-127 vs 131 86-179, P＜0.001) and SII (MIQR: 233 166-266 vs 292 189-412 vs 376 233-695, P＜0.001) among the three groups were statistically significant. The median OS of all patients was 21.7 months (95% CI: 18.171-25.189). Multivariate analysis of the individual application of the four markers showed that time-dependent NLR, time-dependent PLR, time-dependent PNI and time-dependent SII were independent related factors affecting OS (P＜0.05). With an increase in the NLR, PLR and SII, the hazard ratio (HR) of OS increased. Conversely, as the PNI increased, the HR of OS decreased. Among all predictive models, the combined predictive of time-dependent PLR, time-dependent PNI and time-dependent SII demonstrated the best predictive effect, with a C-index of 0.750 (95% CI: 0.709-0.790).
Conclusion The study of peripheral blood inflammatory markers in predicting the efficacy of immunotherapy combined with anti-angiogenesis therapy for advanced HCC shows that patients with lower levels of NLR, PLR, SII and higher levels of SII during treatment have better effectiveness. The predictive effect of the time-dependent model during the whole observation period is superior to the baseline data. Clearly, the combined prediction of the time-dependent PLR, time-dependent PNI and time-dependent SII exhibits the most accurate predictive effect.
DownLoad: