Construction and validation of a survival prediction model for KRAS-mutant advanced non-small cell lung cancer patients treated with immunosuppressants
10.3760/cma.j.cn115355-20230915-00108
- VernacularTitle:免疫抑制剂治疗KRAS突变晚期非小细胞肺癌患者生存预测模型的构建及验证
- Author:
Yanfeng XUE
1
;
Kang ZHENG
Author Information
1. 山西省肿瘤医院 中国医学科学院肿瘤医院山西医院 山西医科大学附属肿瘤医院特需医疗部，太原 030001
- Keywords:
Carcinoma, non-small-cell lung;
Genes, ras;
Mutation;
Immunosuppressive agents;
Prognosis
- From:
Cancer Research and Clinic
2024;36(10):747-751
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To construct and preliminary evaluate a survival prediction model for immunosuppressant treatment of advanced non-small cell lung cancer (NSCLC) patients with KRAS mutations.Methods:A retrospective cohort study was conducted. Eighty-seven KRAS-mutant advanced NSCLC patients treated with immunosuppressants who were admitted to Shanxi Province Cancer Hospital from May 2017 to May 2020 were selected and followed up until May 2023. Kaplan-Meier overall survival curves were plotted. The patients were categorized into survival and death groups based on survival status at the final follow-up, with 31 and 56 cases in each group respectively, and the clinical data were compared between the two groups. Cox proportional hazards model was used to screen the risk factors affecting the death of patients, and a Cox regression prediction model for the survival of such patients was constructed based on the above risk factors. Using survival status at the final follow-up as the gold standard, receiver operating characteristic (ROC) curves for this predictive model to predict survival in immunosuppressant-treated KRAS-mutant advanced NSCLC patients were plotted.Results:Of the 87 patients, 57 (65.5%) were male and 30 (34.5%) were female, aged (62±8) years. Patients had a median overall survival time of 18.5 months (95% CI: 7.5-37.5 months) and a 3-year cumulative survival rate of 35.63% (31/87). Compared with the survival group, the Eastern Cooperative Oncology Group physical status (ECOG-PS) score, carcinoembryonic antigen (CEA), red blood cell distribution width (RDW), fibrinogen (FIB), D-dimer (D-D), lactate dehydrogenase (LDH), and neutrophil-to-lymphocyte ratio (NLR) in the death group were higher, and the differences were statistically significant (all P < 0.05). Cox regression analysis showed elevated ECOG-PS score ( HR = 1.925, 95% CI: 1.745-2.515, P < 0.001), elevated RDW ( HR = 2.012, 95% CI: 1.820-2.619, P < 0.001), elevated FIB ( HR = 2.060, 95% CI: 1.908-2.678, P = 0.009), elevated D-D ( HR = 2.112, 95% CI: 1.885-2.791, P = 0.012), elevated LDH ( HR = 2.104, 95% CI: 1.901-2.643, P < 0.001), elevated NLR ( HR = 1.998, 95% CI: 1.764-2.580, P < 0.001) were all independent risk factors for death in immunosuppressant-treated advanced NSCLC patients with KRAS mutations. A Cox regression prediction model was constructed based on the above risk factors: prognostic index (PI)=-10.342+0.582×ECOG-PS score+0.605×RDW+0.610×FIB+0.599×D-D+0.612×LDH+0.618×NLR. ROC curve analysis showed that the area under the curve of the model for predicting survival of the patients was 0.885, and the optimal cut-off value was 1.252, as well as the sensitivity, accuracy and positive predictive value were 90.50%, 87.09% and 84.32%, respectively. Conclusions:ECOG-PS score, RDW, FIB, D-D, LDH, and NLR are associated with the survival of KRAS-mutated advanced NSCLC patients treated with immunosuppressants, and the Cox regression model constructed on the basis of the above factors has a good efficacy in predicting the survival of patients.
