PWE-004 Detection of Venous Invasion to Stage Colorectal Cancer. (4th June 2013)

• Record Type: Journal Article
• Authors: Foulis, A K; Roxburgh, C S D; McMillan, D C; Richards, C H; Atwan, M; Anderson, J H; Harvey, T; Horgan, P G
• Is Part Of: Gut. Volume 62(2013)Supplement 1
• Page Start: A131
• Page End: A132
• Publication Date: 2013-06-04
• Subjects: Gastroenterology -- Periodicals; 616.33
• DOI: 10.1136/gutjnl-2013-304907.293
• Languages: English
• ISSNs: 0017-5749

Abstract

Abstract : Introduction: Elastica staining of tumour sections increases the sensitivity of detection of venous invasion in colorectal cancer. We compared the prognostic value of elastica detected venous invasion with that of other pathological features in colorectal tumours. Methods: This was a single-centre clinical outcome study of pathological variables, including elastica detected venous invasion, in stage I-III electively resected colorectal cancer specimens. Results: 631 resection specimens, excised between 1997–2009, were analysed (176 retrospectivley and 455 prospectively). The median follow up was 73 months (24–178) and during this time there were 238 deaths (134 from cancer). Venous invasion was detected in 56% of cases and was a stronger predictor of poor long term cancer-specific survival than other pathological features on univariate and multivariate analyses. On multivariate analysis of all cases the hazard ratio (HR) for failure to survive 5 years for venous invasion = 3.94 (95%CI 2.33–6.65, P < 0.001); HR for lymph node involvement = 1.81, (95% CI 1.43–2.30, P < 0.001) and HR for T stage = 1.64 (95%CI 1.16–2.30, P = 0.005). In node negative cases the HR for failure to survive 5 years for venous invasion on multivariate analysis = 3.55 (95%CI 1.81–6.97, P < 0.001) and for T stage was 2.03 (95%CI 1.26–3.28, P = 0.004). Venous invasion strongly related to other high-risk pathological variables. In cases with no venous invasion, no pathological characteristic related to survival other than T stage. When T stage and venous invasion were considered together, patients could be stratified by risk of 5-year cancer mortality from 100–54% in node negative disease and 100–33% in node positive disease. The importance of elastica detected venous invasion can be appreciated from the development of a novel staging system based only on T stage and venous invasion (TVI). This simple TVI system was at least as predictive as the gold standard TNM system when considering all cases, and provided increased prognostic value in both T1 and T2 tumours, as well as in node negative disease. Conclusion: Sensitive, accurate detection of venous invasion on elastica stained sections improves its prognostic importance such that it becomes a key pathological characteristic, arguably of more importance than nodal status, in determining outcome in patients with colorectal cancer. TVI staging provides a novel and simple method by which venous invasion coupled with T stage can be utilised to predict survival. Disclosure of Interest: None Declared