Tumour Budding in TI and T2 Node-Negative (Dukes A) Colorectal Carcinoma
CA D'Arcy, D Kevans, AA Maguire, J O'Sullivan, H Mulcahy, D O'Donoghue, J Hyland, D Winter, PR O'Connell, K Sheahan. St Vincent's University Hospital and Centre for Colorectal Disease, Dublin, Ireland
Background: Tumour budding has been shown to be a strong, independent, and reproducible prognostic marker of patient outcome in T3N0 colorectal carcinoma by Wang et al (Am J Surg Path, in press). The significance of tumour budding in T1N0 and T2N0 tumours has not been widely studied.
Design: One hundred and twenty-one T1N0 and T2N0 colorectal carcinoma cases were identified from the database at this institution over a fifteen-year period (January 1993-July 2008). A subset of these included polypectomy specimens with carcinoma arising in polyps (n=28). Synchronous tumours, ulcerative colitis and familial adenomatous polyposis associated tumours were excluded. Haematoxylin & eosin stained sections from each case were examined for tumour budding using the rapid bud count method described previously (five areas examined at 200x per slide). Additional prognostic parameters studied included lymphovascular invasion, tumour margin, depth of invasion, and tumour differentiation. Patient outcome was evaluated and correlated with these findings.
Results: Eighty-five (70%) of carcinomas had low and thirty-six carcinomas (30%) had high bud scores. High budding was associated with an infiltrative margin (p=<0.0001). High budding was not seen in the six cases of intramucosal carcinoma. High budding was associated with a greater depth of invasion (p=0.03), present in 8/42 (19%) of T1 carcinomas and 28/72 (39%) of T2 carcinomas.The median follow-up of patients was 2.03 years (range 0-12.4 years). Cancer deaths were observed in 7% of patients in the low budding category versus 8.6% of patients in the high budding category (p=0.66).
Conclusions: High tumour budding was identified in 30% of Dukes A colorectal cancers. High tumour budding is associated with a greater depth of invasion and with an infiltrative margin. The presence of tumour budding did not predict a poorer prognosis in this series. A study of a larger number of cases with a longer duration of follow-up may be necessary to investigate the value of tumour budding as a predictor of survival in the T1N0 and T2N0 setting.
Monday, March 9, 2009 1:00 PM
Poster Session II # 86, Monday Afternoon