Differential Expression Of Rictor Protein In The Gastric Adenocarcinoma And Its Prognostic Prediction Nomogram

Objective:To detect Rictor expression in gastric adenocarcinoma and construct a line graph model for predicting progression-free survival(PFS)in D2+ radical gastric adenocarcinoma resected patients with positive and negative test tissue for Rictor protein,and analyse its predictive value.Methods:Gastric adenocarcinoma patients who underwent D2+ radical resection at Shanxi Cancer Hospital from May 2005 to December 2020 were retrospectively collected in1366 cases,and the preserved tissue samples were assayed for Rictor protein using the immunohistochemical SP method.690 patients with Rictor-negative gastric adenocarcinoma were randomly grouped in a 7:3 ratio,including 486 in the training cohort and 204 in the validation cohort.Six hundred and seventy-six patients with Rictor-positive gastric adenocarcinoma were randomized in a 7:3 ratio,including 496 in the training cohort and 180 in the validation cohort.The relationship between Rictor protein expression status and other clinicopathological factors and progression-free survival(PFS)in Rictor-positive and negative patients was analysed using a multifactorial Cox proportional risk model to identify independent influences on PFS.Column plots predicting 3-and 5-year PFS in patients with gastric adenocarcinoma were constructed based on PFS-independent influencing factors.Internal cross-validation with10,000 replicate samples was performed to test the accuracy of the prediction model.The internal and external predictive power of the model was further assessed by calibration curves,area under the curve(AUC)of temporal subject operating characteristics(ROC)and decision curve analysis(DCA).The column line graph model was applied to the training cohort PFS scores,and the X-tile software was used to obtain the best critical values for the scores.The overall cohort,training cohort and validation cohort cases were divided into a low risk group(≤ best critical value)and a high risk group(> best critical value)based on the best critical values,and the Kaplan-Meier method was used to compare the differences in progression-free survival(PFS)between the two risk strata for each cohort.Results:Rictor protein was positively expressed in 49.5%(676/1366)of gastric adenocarcinomas,much higher than the 24.8%(123/496)of Rictor protein expression in normal paracancerous tissue.Rictor expression correlated significantly with lymph node-related metastases(2 = 8.341,P = 0.039)and surgical margins(2 = 10.915,P= 0.001).The difference in overall survival(OS)between Rictor-negative and positive patients was significant(P = 0.046),while the difference in progression-free survival(PFS)was not significant(P = 0.112).The C-index for the Rictor expression-negative PFS line chart was 0.760(95% CI: 0.720-0.799),which was better than the AJCC(American Joint Committee on Cancer)8th edition TNM staging C-index of 0.683(95%CI:0.646-0.721).Columnar plots were developed for predicting PFS with positive Rictor expression,including gender,age,p T stage,number of positive lymph nodes,nerve invasion,maximum tumour diameter,omental invasion,Clavien-Dindo postoperative complication grade,and CGA expression for nine variables.the C-index of the columnar plots for PFS with positive Rictor expression was 0.795(95% CI: 0.764-0.825),which was better than the AJCC(American Joint Committee on Cancer)8th edition TNM staging C-index of 0.693(95% CI: 0.662-0.723).The probabilistic calibration curves of the two sets of columnar graph models indicated that the columnar graph predictions were in good agreement with the actual observations.The area under the curve(AUC)of the time-dependent respondent operating characteristic(ROC)for both sets of columnar graph models showed good differentiation between the two sets of columnar graphs.Kaplan-Meier survival analysis showed statistically significant differences in PFS between the low-and high-risk groups for each cohort for both models(all p<0.001).Conclusion:Rictor protein expression in gastric adenocarcinoma has predictive value for its prognostic monitoring,and our proposed column line plot for predicting PFS in Rictor expression negative and Rictor expression positive gastric adenocarcinoma after resection of D2+ radical gastric adenocarcinoma shows good differentiation between the training and validation cohorts and may help guide clinicians’ decision making as well as help identify patients at high risk or estimate survival rates.Care is taken to control tumour metastases and postoperative complications for high-risk patients with negative Rictor expression,but is not necessary for low-risk patients.Also,no other post-operative treatment is required for low-risk patients with positive Rictor expression,but high-risk patients should be aware of targeted therapies for positive Rictor expression in addition to controlling metastases and post-operative complications of the tumour.