Background <p>Although preoperative carcinoembryonic antigen (CEA) is a variable used in most prognostic scores assessing the outcome of patients with colorectal liver metastases (CRLM) undergoing resection, it is unclear what the optimal cut-off is or in which patient subgroups CEA is most relevant. The purpose of this study is to evaluate the prognostic effect of CEA in resected CRLM patients and to explore in which subgroups CEA is most closely associated with overall survival (OS).</p> Methods <p>A systematic literature search was performed, selecting studies that evaluated a relationship between preoperative CEA and OS in patients undergoing CRLM radical surgery. A meta-analysis assessed the overall effect size of the relationship on the selected study cohorts, based on CEA cut-off. An evaluation of 21 baseline variables was performed to explore their possible effect on the relationship between CEA and OS.</p> Results <p>The study confirms a significant negative prognostic effect of increased CEA on OS (HR 1.46, CI 1.30–1.65), but heterogeneity among studies is significant. The effect is consistent for all CEA cut-offs, although the relationship tends to weaken in more recent studies for cut-offs &lt; 10 ng/mL. Meta-regressions also suggest that the prognostic effect may be more pronounced in the elderly. In addition, the effect of CEA ≥ 20 ng/mL on OS appears significantly reduced in the subgroup with mutated RAS carcinoma.</p> Conclusions <p>For patients with resectable CRLM, the CEA cut-off should be increased to 20 ng/mL, and evaluation in prospective studies of the more pronounced negative prognostic effect of preoperative CEA in the elderly and wild-type RAS CRLM patients is recommended.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Prognostic Effect of CEA Cut-Off in Patients with Resectable Colorectal Liver Metastases: A Meta-Analysis and Meta-Regression

  • Antonella Venturino,
  • Giuseppe A. Colloca

摘要

Background

Although preoperative carcinoembryonic antigen (CEA) is a variable used in most prognostic scores assessing the outcome of patients with colorectal liver metastases (CRLM) undergoing resection, it is unclear what the optimal cut-off is or in which patient subgroups CEA is most relevant. The purpose of this study is to evaluate the prognostic effect of CEA in resected CRLM patients and to explore in which subgroups CEA is most closely associated with overall survival (OS).

Methods

A systematic literature search was performed, selecting studies that evaluated a relationship between preoperative CEA and OS in patients undergoing CRLM radical surgery. A meta-analysis assessed the overall effect size of the relationship on the selected study cohorts, based on CEA cut-off. An evaluation of 21 baseline variables was performed to explore their possible effect on the relationship between CEA and OS.

Results

The study confirms a significant negative prognostic effect of increased CEA on OS (HR 1.46, CI 1.30–1.65), but heterogeneity among studies is significant. The effect is consistent for all CEA cut-offs, although the relationship tends to weaken in more recent studies for cut-offs < 10 ng/mL. Meta-regressions also suggest that the prognostic effect may be more pronounced in the elderly. In addition, the effect of CEA ≥ 20 ng/mL on OS appears significantly reduced in the subgroup with mutated RAS carcinoma.

Conclusions

For patients with resectable CRLM, the CEA cut-off should be increased to 20 ng/mL, and evaluation in prospective studies of the more pronounced negative prognostic effect of preoperative CEA in the elderly and wild-type RAS CRLM patients is recommended.