[766] Ueno Method Substaging pT1 Colorectal Adenocarcinoma by Depth & Width Measurement: An Interobserver Study

Lai M Wang, Eve P Fryer, Pelvender Gill, Caroline Hughes, Aron Szuts, Ranmith Perera, Runjan Chetty. John Radcliffe Hospital Oxford University Hospital, Oxford, United Kingdom

Background: Early pT1 polyp colorectal cancers (CRC) present challenges for accurate pathology substaging. The Haggitt and Kikuchi stages for pedunculated and sessile polyps respectively depend on polyp morphology and are often difficult to apply due to suboptimal specimen orientation or fragmentation. The Kikuchi method requires division of the submucosal region into thirds, requiring the presence of the muscularis propria, usually not included in endoscopic polypectomy or submucosal resection specimens. The European guidelines for quality assurance in CRC diagnosis suggest using the more objective approach by Ueno using depth and width beyond muscularis mucosae. We investigate the interobserver variability in the Ueno approach to early polyp CRC.
Design: The previous ten pT1 polyp CRC were identified by a search through the cellular pathology database. The section with the deepest advancing tumour edge for each polyp was selected. The slides were assessed by six GI pathologists and the depth and width of the invasive tumour in the submucosa were submitted independently. Results were recorded and agreement was assessed by Kendall's coefficient of concordance (KCC) and κ statistics. Polyp CRC is classified as high risk (HR) if depth ≥2000μm or width ≥4000μm and low risk (LR) if depth <2000μm or width <4000μm.
Results: Data for the depth and width are shown in table 1.

TABLE 1 Depth & Width of invasion: Mean ± [SD] & range. 6 pathologists´ collective data.
POLYPSMean ± [SD]RangeMean ± [SD]Range
A3883 ± 112031007070 ± 26005900
B1033 ± 2076003217 ± 14154000
C900 ± 96326003050 ± 16944200
D3367 ± 42310003383 ± 4831000
E667 ± 2075001333 ± 3781000
F2233 ± 39310004833 ± 15714500
G2533 ± 90326004400 ± 25107200
H6783 ± 677180014333 ± 18625000
I3483 ± 108928005250 ± 31008100
J5917 ± 183951008180 ± 24706300

The KCC for depth is 0.89 and width is 0.79. When polyp CRC are stratified into high and low risk categories based on depth, there is total agreement in 8 of ten cases, κ = 0.780. When based on width, only 4 of ten have complete agreement, κ = 0.403. Disagreement cases are listed in table 2.

TABLE 2 Disagreement cases

Conclusions: Ueno's method has the advantage of being independent of polyp morphology. Our study shows that there is great concordance amongst pathologists by depth measurement and allows good reproducibility in risk categorization of pT1 CRC. There is poor agreement when based on width, which should therefore be discouraged.
Category: Gastrointestinal

Wednesday, March 6, 2013 9:30 AM

Poster Session V # 81, Wednesday Morning


Close Window