[672] Proposal for a More Applicable and Clinically Relevant Staging Evaluation of Ampullary Carcinomas

T Tajiri, I Coban, A Krasinskas, S Khayyata, O Basturk, E Levi, D Altinel, CK Chu, D Kooby, JM Sarmiento, C Staley, NV Adsay. Emory U., Atlanta; U. of Pittsburgh, Pittsburgh; WSU, Detroit; NYU, NY

Background: The staging of ampullary (AMP) carcinoma (CA) is challenging due to the anatomic complexity of this region and rarity of these tumors.
Design: 98 resected invasive AMP-CAs were evaluated for T-staging parameters. In addition to depth of invasion into different structures, size of invasion (largest diameter of non-mucosal component) was also analyzed.
Results: An attempt was made to classify the cases by the current TNM; however, it was readily evident, especially from the 30 cases grossed using the open-ampulla approach, that this system is inapplicable and irreproducible due to problematic definition, and presumed significance of duodenal wall, invasion to pancreas and extrapancreatic soft tissue. Thus, a modified staging approach was devised according to 3 dimensional (3-D) spread of the tumor, possible biologic significance of extension to the structures readily associated with the ampulla, and practicality and reproducibility. The ampulla was regarded as a tube in continuum with the duodenum and the following levels of tumor involvement were assigned: T1-submucosa (of any component); T2-muscles (regardless of whether it is Oddi, duodenal or CBD); T3-crossing the muscles into adjacent tissues; i.e duodenal subserosa (T3s) or 0.5 cm into the pancreas (T3p). 0.5 was chosen because it was noted that cases with subserosal involvement also had a peri-muscular spread of up to 0.5 cm. T4-serosal surface, or invasion into pancreas >0.5 cm. This proposed staging had significant prognostic value: Median overall survival (OS) for T1-T4 were 122, 65, 24, 25 mos, respectively (p=0.026). OS of T1/2 was markedly better than T3/4 (122 vs 24 mos; p=0.004). Separately, size of invasive CA was found to be highly predictive of clinical outcome, as invasion 1.0 cm was associated with OS of 122 mos, whereas >1.0 cm was 26 mos.(p=0.004).
Conclusions: The proposed staging system incorporates various facets of the 3-D spread of AMP-CA not properly captured in the current TNM. It is practical and reproducible, and it has strong prognostic value. Size of invasive CA provides additional important survival information and should be included in pathologic evaluation of AMP-CA.
Category: Gastrointestinal

Monday, March 9, 2009 11:45 AM

Platform Session: Section C, Monday Morning