He M, Jia Z, Hu L, Wu H. Development and validation of a nomogram to predict which patients with colorectal cancer liver metastases would benefit from primary tumor resection.
Int J Colorectal Dis 2023;
38:144. [PMID:
37237238 DOI:
10.1007/s00384-023-04426-5]
[Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Accepted: 05/01/2023] [Indexed: 05/28/2023]
Abstract
PURPOSE
The use of primary tumor resection (PTR) in the treatment of colorectal cancer liver metastases (CRLM) patients has become increasingly controversial. Our goal is to establish a nomogram to screen for the candidates that would benefit from PTR in CRLM patients.
METHODS
The Surveillance, Epidemiology, and End Results (SEER) database was searched for 8366 patients with colorectal liver cancer metastases (CRLM) from 2010 to 2015. Overall survival (OS) rates were calculated using the Kaplan-Meier curve. After propensity score matching (PSM), predictors were analyzed by logistic regression analysis, and a nomogram was created to predict for survival benefit of PTR using R software.
RESULTS
After PSM, there were 814 patients in both PTR group and non-PTR group, respectively. The median OS time in the PTR group was 26 months (95%CI = 23.33 ~ 28.67) and the median OS time in the non-PTR group was 15 months (95%CI = 13.36 ~ 16.64). The Cox regression analysis found that PTR was an independent predictive factor (HR = 0.46, 0.41 ~ 0.52) for OS. Additionally, logistic regression was used to study the factors impacting PTR benefit, and the results showed that CEA (P = 0.016), chemotherapy (P < 0.001), N stage (P < 0.001), histological grade (P < 0.001), and lung metastasis (P = 0.001) are independent predictive factors affecting the therapeutic outcome of PTR in patients with CRLM. The developed nomogram displayed good discriminative ability in predicting the beneficial probability of PTR surgery, with the area under the curve (AUC) values of 0.801 in training set and 0.739 in validation set respectively.
CONCLUSION
We developed a nomogram that predicts the survival benefits of PTR in CRLM patients with relatively high accuracy, and quantifies the predictive factors for PTR-related benefits.
Collapse