Recurrence Following Hepatectomy for Metastatic Colorectal Cancer: Development of a Model that Predicts Patterns of Recurrence and Survival
Recurrence Following Hepatectomy for Metastatic Colorectal Cancer: Development of a Model that Predicts Patterns of Recurrence and Survival
复制标题
DOI：
10.1245/s10434-011-1921-y
复制
发表时间：
2012-01-01
影响因子：
3.7
通讯作者：
Scoggins, Charles R.
中科院分区：
文献类型：
--
作者：
St Hill, Charles R.;Chagpar, Ryaz B.;Scoggins, Charles R.
虽然已经开发了几种预后模型来预测接受肝切除术治疗转移性结直肠癌（mCRC）的患者的生存率，但很少有数据可以预测复发后的生存率。我们试图建立一个模型来预测mCRC肝切除术后复发患者的生存率。我们对连续的mCRC肝切除术患者的数据进行了回顾性分析。分析临床病理资料、复发模式和预后。采用Kaplan-Meier生存分析、单因素和多因素分析。建立了一个基于整数的模型来预测复发模式和复发后的生存率，该分析包括280例患者，中位随访时间为50.1个月。其中，53%接受了肝切除术，87%切缘阴性。63%的患者复发。与肝切除术后无病期（DFI）和总生存期（OS）相关的因素包括CEA > 200 ng/ml（P < 0.0005）、> 1个转移（P < 0.0005）和高Fong评分（P < 0.0005）。复发后，复发模式是OS的强预测因子（P < 0.0005）。多变量分析中复发模式的独立预测因子包括CEA > 200 ng/ml、肿瘤大小> 5 cm和> 1个肝转移。根据与复发模式相关的β系数（P < 0.0005）建立了一个简单的预测评分系统，复发模式可以很好地预测肝切除术后复发mCRC患者的生存率，并且复发模式可以用一个简单的模型预测。这也可以扩展到创建一个估计预期生存率的评分系统。
While several prognostic models have been developed to predict survival of patients who undergo hepatectomy for metastatic colorectal cancer (mCRC), few data exist to predict survival after recurrence. We sought to develop a model that predicts survival for patients who have developed recurrence following hepatectomy for mCRC.A retrospective analysis was performed on data from consecutive patients that underwent hepatectomy for mCRC. Clinicopathologic data, recurrence patterns, and outcomes were analyzed. Kaplan-Meier survival analysis and univariate and multivariate analyses were performed. An integer-based model was created to predict the patterns of recurrence and survival after recurrence.This analysis included 280 patients with a median follow-up of 50.1 months. Of these, 53% underwent major hepatectomy and 87% had negative margins. Recurrent disease developed in 63% of patients. After hepatectomy, factors associated with short disease-free interval (DFI) and overall survival (OS) included CEA > 200 ng/ml (P < 0.0005), > 1 metastasis (P < 0.0005), and a high Fong score (P < 0.0005). After recurrence, the pattern of recurrence was a strong predictor of OS (P < 0.0005). Independent predictors of the pattern of recurrence on multivariate analysis include CEA > 200 ng/ml, tumor size > 5 cm, and > 1 liver metastasis. A simple predictive scoring system was developed from the beta coefficients of this analysis that correlated with recurrence pattern (P < 0.0005).After hepatectomy, survival of patients with recurrent mCRC is strongly predicted by the patterns of recurrence, and the recurrence pattern can be predicted with a simple model. This can also be extended to create a scoring system that estimates expected survival.