Construction and application value of nomogram predictive model for the prognosis of rectal cancer liver metastases based on SEER database
10.3760/cma.j.cn115610-20230910-00081
- VernacularTitle:基于SEER数据库直肠癌肝转移预后列线图预测模型的构建及其应用价值
- Author:
Jun YING
1
;
Yahuang SUN
;
Anqi WANG
;
Ce BIAN
;
Guoliang CHEN
;
Yu TAO
;
Junnan CHEN
;
Hao LU
;
Qing YOU
;
Haiyang ZHOU
;
Zhiguo WANG
;
Canping RUAN
;
Jian ZHANG
Author Information
1. 海军军医大学第二附属医院（上海长征医院）肛肠外科，上海 200003
- Keywords:
Rectal neoplasms;
Liver metastasis;
SEER database;
Prognosis;
Nomo-gram predictive model
- From:
Chinese Journal of Digestive Surgery
2023;22(S1):51-57
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To investigate the construction and application value of a nomogram predictive model for the prognosis of rectal cancer liver metastases based on Surveillance, Epidemio-logy, and End Results (SEER) database.Methods:The retrospective cohort study was conducted. The clinicopathological data of 6 192 patients with rectal cancer liver metastases in the SEER database ( http://seer.cancer.gov/) and 312 patients who were admitted to The Second Affiliated Hospital of Naval Medical University January 2010 to December 2016 were collected. Of 6 192 patients, there were 3 592 males and 2 600 cases. There were 1 076 cases with age lower than 50 years, 2 862 cases with age as 50-69 years, 2 254 cases with age equal to or more than 70 years, respectively. Of 312 pati-ents, there were 177 males and 135 cases. There were 51 cases with age lower than 50 years, 155 cases with age as 50-69 years, 109 cases with age equal to or more than 70 years, respectively. Patients of the SEER database were set as the training set, and patients in The Second Affiliated Hospital of Naval Medical University were set as the validation set. Univariate and multivariate COX proportional hazards regression models were used to analyze risk factors associated with prognosis, and construct and verify the accuracy of nomogram predictive model for the prognosis of rectal cancer liver metas-tasis. The training set were used to construct the nomogram prediction model, and the validation set were used to verify its performance. Observation indicators: (1) prognostic factors analysis in patients with rectal cancer liver metastases; (2) construction and verificative of the predictive model for the prognosis of rectal cancer liver metastasis. Measurement data with normal distribution were represented as Mean± SD, and comparison between groups was conducted using the t test. Count data were described as absolute numbers or percentages, and comparison between groups was conducted using the chi-square test. Comparison of ordinal data was analyzed using the rank sum test. The COX regression model was used for univariate and multivariate analyses. Kaplan-Meier method was used to calculate survival rates, and Log-Rank test was used for survival analysis. Results:(1) Prognostic factors analysis in patients with rectal cancer liver metastases. Results of multivariate analysis showed that age >50 years, TNM Ⅱ-Ⅳ stage, stage T3-T4, stage N1-N2, the number of lymph nodes dissected <12, tumor diameter >5.1 cm, positive carcinoembryonic antigen, peripheral nerve infiltration, radiotherapy and adjuvant chemotherapy, poorly differentiated or undifferented tumor were independent prognostic factors of patients ( P<0.05). (2) Construction and verification of the predictive model for the prognosis of rectal cancer liver metastasis. A nomogram predictive model for the prognosis of rectal cancer liver metastasis was constructed based in the multivariate analysis. The C-index of the nomogram predictive model was 0.91, with area under the curve as 0.726, indicating a good discriminant ability. Results of the calibration curve in validation dataset showed that the colorectal cancer survival rate predicted by the nomogram predictive model was consistent with the actual survival rate. Conclusion:The nomogram predictive model can accurately predict the survival probability of patients with rectal cancer liver metastases.
