Establishment and validation of a long-term prognosis prediction model for patients with non-small cell lung cancer

Journal of Thoracic Disease

article

Nianxi Tan¹  Cuicui Xu²  Yu Mao³ 

[1]Department of Cardiothoracic Vascular Surgery, Zhuzhou Central Hospital

[2]Department of Respiratory Medicine, Shanghai Fengxian District Central Hospital

[3]Department of Thoracic Surgery, The First Hospital of Hohhot

关键词: Non-small cell lung cancer;    mortality;    predictive models;    risk factors;   

DOI  :  10.21037/jtd-23-381

学科分类：呼吸医学

来源: Pioneer Bioscience Publishing Company

【 摘 要 】

Background: The incidence of non-small cell lung cancer ranks second among malignant tumors, while the mortality rate ranks first. We established a prediction model for the long-term prognosis of lung cancer patients to accurately identify patients with a high risk of postoperative death and provide a theoretical basis for improving the prognosis of patients with non-small cell lung cancer. Methods: The data of 277 non-small cell lung cancer patients who underwent radical lung cancer resection at Shanghai Fengxian District Central Hospital between January 2016 and December 2017 were retrospectively collected. The patients, who were followed up for 5 years, were divided into a deceased group (n=127) and survival group (n=150) according to whether the patients had died 5 years after surgery or not. The clinical characteristics of the two groups were observed, and the risk factors of death within 5 years of surgery in lung cancer patients were analyzed. A nomogram predictive model was then established to analyze the value of the model in predicting the death within 5 years of surgery in patients with non-small cell lung cancer. Results: 193.5 ng/mL, stage III lung cancer, peritumor invasion, and vascular tumor thrombus were independent risk factors of tumor-specific death after surgery in patients with non-small cell lung cancer (P<0.05). R 4.0.3 statistical software was used to randomly divide the dataset into a training set and validation set. The sample size of the training set was 194, and the sample size of the validation set was 83. The area under the receiver operating characteristic (ROC) curve was 0.850 [95% confidence interval (CI): 0.796–0.905] in the training set, and it was 0.779 (95% CI: 0.678–0.880) in the validation set. In the validation set, the model was assessed using the Hosmer-Lemeshow goodness-of-fit test, with a chi-square value of 9.270 and a P value of 0.320. Conclusions: Our model could accurately identify high risk of death within 5 years of surgery in non-small cell lung cancer patients. Strengthening the management of high-risk patients may help improve the prognosis of these patients.

【 授权许可】

Unknown