Prognostic Value of Molecular Detection of Lymph Node Metastases After Curative Resection of Stage II Colon Cancer: A Systematic Pooled Data Analysis

Sharlene Gill, Jean Francois Haince, Qian D Shi, Emily S. Pavey, Guillaume Beaudry, Daniel J. Sargent, Yves Fradet

Research output: Contribution to journalArticle

2 Scopus citations

Abstract

Background: We aimed to clarify the prognostic value of guanylyl cyclase C (GCC) lymph node ratio (LNR) status as a predictor of recurrence in untreated stage IIA colon cancer on the basis of pooled individual data from previous studies. Methods: Patients were classified according to predefined GCC LNR risk groups (low, LNR≤ 0.1; intermediate, 0.1< LNR≤ 0.2; high, LNR > 0.2). Outcomes included time to recurrence, disease-free survival, and overall survival. Stratified log-rank tests and multivariate Cox models assessed the association between outcomes and GCC lymph node status. Results: The final data set contained 553 patients with stage IIA colon cancer with a median of 18 lymph nodes examined after resection; 65 patients (11.8%) had recurrence. Overall, 109 patients (19.7%) were classified high risk on the basis of GCC LNR. In multivariate analysis, high GCC LNR value (> 0.2) was a significant predictor of cancer recurrence (hazard ratio [HR], 3.18; 95% confidence interval [CI], 1.77-5.71; P< .001) and lower disease-free survival (HR, 2.40; 95% CI, 1.60-3.62; P< .001) and overall survival (HR, 2.12; 95% CI, 1.35-3.33; P=.001). Conclusion: Patients considered at high risk on the basis of their GCC LNR status have significantly inferior outcomes compared to those with low GCC LNR values, particularly among those traditionally considered to be at low risk for recurrence.

Original languageEnglish (US)
JournalClinical Colorectal Cancer
DOIs
StateAccepted/In press - Sep 25 2014

    Fingerprint

Keywords

  • Colon cancer
  • Guanylyl cyclase-C receptor
  • Lymph node ratio
  • Molecular staging
  • Prognosis

ASJC Scopus subject areas

  • Oncology
  • Gastroenterology

Cite this