Diagnostic Accuracy of Ureteroscopic Biopsies from Upper Urinary Tract Carcinomas: Does Size Really Matter?
CP Rojas, S Rodriguez, V Bird, I Reis, W Zhao, J Santos, R Leveillee, C Gomez-Fernandez, M Jorda. University of Miami, Jackson Memorial Hospital and Sylvester Comprehensive Cancer Center, Miami; University of Miami, Jackson Memorial Hospital and Sylvester Comprehensive Cancer Center, Miami
Background: Urothelial carcinomas from the upper urinary tract system represent 7-10% of all kidney malignancies. With the advent of new uteroscopic equipment, small tissue samples are available representing a diagnostic challenge. Preciseness in diagnosis is essential for treatment decision. This study correlated tumor grade and pT stage between biopsy and resected specimens and examined if sample volume impacts accuracy of diagnosis.
Design: We reviewed histologic slides from ureteroscopic biopsy samples and corresponding nephroureterectomy specimens from 33 patients with urothelial carcinoma of upper urinary tract. Two urologic pathologists independently evaluated the following disease characteristics: histologic grade (ISUP/WHO 2004), pleomorphism, mitosis, lymphovascular invasion (LVI), and pT stage. We reported the number, percentage of concordant pairs and corresponding 95% confidence interval (95%CI) for the true proportion of concordant pairs between biopsy and resected specimens. A repeated measures model analysis was used to examine whether agreement on a particular disease characteristic was related to specimen size. The outcome for our analysis was log-transformed data on specimen volume ([/6]×a×b×c, where a, b, and c are orthogonal measurements in mm).
Results: The majority of patients were males (21, 63.6%). Mean age was 73 years and 79% of patients were 65 years or older. Good agreement was observed for grade (concordance 87.9%, 95%CI: 71.8 to 96.6%), pleomorphism (75.8%, 95%CI: 57.5 to 88.9%), mitosis (84.8%, 95%CI: 68.1 to 94.9%), and LVI (87.9%, 95%CI: 71.8 to 96.6%), but agreement was lower for pT stage (45.5%, 95%CI: 28.1 to 63.6%). Biopsy volume yield a geometric mean of 0.022 mm3, range 0.001 to 4.514 mm3; tumor volume yield geometric mean of 1.351 mm3, range 0.002 to 65.0 mm3 (p<0.05). The volume of biopsy was not related to agreement for any of the disease characteristics (p>0.05).
Conclusions: 1- Histologic grade assigned on the biopsy sample accurately predicts histologic grade in the resected specimen (87.9%), even when biopsy volume was small. 2- Ureteroscopic biopsies provided sufficient information for clinical decision-making. 3- The agreement between biopsy and resected specimen was lower for the characteristic of pT stage.
Category: Genitourinary (including renal tumors)
Monday, March 22, 2010 9:30 AM
Poster Session I Stowell-Orbison/Surgical Pathology/Autopsy Awards Poster Session # 135, Monday Morning