The predictive value of preoperative lymphocyte-to-monocyte ratio combined with platelet-to-lymphocyte ratio scoring model for prognosis of pancreatic ductal adenocarcinoma after radical resection / 中华消化外科杂志

ABSTRACT

Objective:To investigate the predictive value of preoperative lymphocyte-to-monocyte ratio (LMR) combined with platelet-to-lymphocyte ratio (PLR) (LMR-PLR) scoring model for prognosis of pancreatic ductal adenocarcinoma (PDAC) after radical resection.Methods:The retrospective cohort study was conducted. The clinicopathological data of 116 patients with PDAC who were admitted to the Second Hospital of Lanzhou University from January 2015 to December 2019 were collected. There were 73 males and 43 females, aged 61.5(range, 29.0-75.0)years. All patients underwent radical resection for PDAC. Observation indicators: (1) optimal cut-off value of LMR and PLR; (2) clinicopathological features of patients with different scores of preoperative LMR-PLR scoring model; (3) follow-up and survival; (4) influencing factors for prognosis of PDAC patients; (5) construction and verification of nomogram prediction model. Measurement data with skewed distribution were represented as M(range). Count data were described as absolute numbers, and comparison between groups was conducted using the chi-square test. Comparison of ordinal data was conducted using the Mann-Whitney U test. The Graphpad prism 8 was used to draw survival curve, the Kaplan-Meier method was used to calculate survival rate, and the Log-Rank test was used for survival analysis. The COX proportional hazard regression model was used for univariate and multivariate analyses. The X-tile software was used to determine the optimal cut-off values of LMR and PLR. The nomogram prediction model was conducted based on the results of multivariate analysis, and the receiver operating characteristic (ROC) curve was drawn. The area under curve (AUC) was used to evaluate the discrimination of nomogram prediction model. The calibration curve was used to evaluate the consistency of nomogram prediction model and the decision curve was used to evaluate the clinical benefits. Results:(1) Optimal cut-off value of LMR and PLR. The optimal cut-off values of LMR and PLR were 1.9 and 156.3. (2) Clinicopathological features of patients with different scores of preoperative LMR-PLR scoring model. Cases with LMR-PLR scoring as 0, 1, 2 were 11, 42, 63. Cases with CA125 <12.4 U/mL, cases postoperative with vascular invasion, cases with postoperative chemotherapy in patients with 0, 1, 2 of LMR-PLR scoring were 1, 8, 24, 9, 27, 27, 3, 26, 43, showing significant differences among them ( χ2=6.73, 8.37, 6.68, P<0.05). (3) Follow-up and survival. All 116 patients were followed up for 39(range, 2-86)months. The 1-, 2-, 3-year survival rate of 116 PDAC patients was 50.9%, 37.9%, 19.3%, respectively, with a survival time of 13(range, 1-85)months. The survival time of patients with LMR-PLR scoring as 0, 1, 2 was 3(range, 1-9)months, 7(range, 2-56)months, 26(range, 2-85)months, respectively, showing a significant difference among them ( χ2=48.78, P<0.05). (4) Influencing factors for prognosis of PDAC patients. Results of multivariate analysis showed that carcinoembryonic antigen (CEA), CA19-9, LMR-PLR score, tumor diameter were independent factors affecting prognosis of patients ( hazard ratio=1.61, 1.88, 0.27, 1.87, 95% confidence interval as 1.02-2.54, 1.18-3.00, 0.19-0.39, 1.13-3.09, P<0.05). (5) Construction and verification of nomogram prediction model. The nomogram prediction model was constructed based on CEA, CA19-9, LMR-PLR score and tumor diameter. The AUC of ROC curve in predicting 1-, 2-, 3-year survival rate of patients was 0.86 (95% confidence interval as 0.79-0.93, P<0.05), 0.86 (95% confidence interval as 0.79-0.92, P<0.05), 0.87 (95% confidence interval as 0.78-0.95, P<0.05), respectively. Results of calibration curve showed that the predicted survival rate of nomogram prediction model was consistent with the actual survival rate, with the consistency index as 0.74. Results of decision curve showed that the predictive performance of nomogram prediction model was superior to that of a single factor at a risk threshold of 0.12-0.85. Conclusions:CEA, CA19-9, LMR-PLR score, tumor diameter are independent factors affecting prognosis of patients undergoing radical resection for PDAC, and the nomogram prediction model can predict postoperative survival rate. The predicted survival rate of nomogram prediction model is consistent with the actual survival rate, and the predictive performance of nomogram prediction model is superior to that of a single factor at a risk threshold of 0.12-0.85.