MP18-15: Optical coherence tomography for grading of uppe ... coefficient cut-off for low and high grade UTUC

Optical coherence tomography for grading of upper tract urothelial carcinoma: a first validation of the attenuation coefficient cut-off for low and high grade UTUC

View Poster

INTRODUCTION

Optical coherence tomography (OCT) is an optical imaging technique that may enable intraoperative grading and staging of upper tract urothelial carcinoma (UTUC) during ureteroscopy. Quantitative analysis of the change of OCT signal with imaging depth, called the attenuation coefficient (mOCT), enables UTUC grading. In 2016, Bus et al proposed a mOCT cut-off of 2.4mm^-1 to differentiate between low and high grade UTUC. The current study is the first validation of the proposed mOCT cut-off for UTUC grading.

METHODS

UTUC was diagnosed with ureteroscopic biopsies in 40 renal units of 38 consecutive patients, who underwent ureteroscopic OCT (C7-XR OCT system with 2.7Fr Dragonfly probe) prior to biopsies. mOCT analysis was performed offline by two raters, who were blinded for clinical data. For each renal unit, a manually selected region was analyzed in 5 OCT images of the evaluated tumor. UTUC grade, based on the proposed mOCT cut-off, was compared to the histologic grade. Histologic grading of the biopsies was performed by a specialized uropathologist, who was blinded for OCT images.

RESULTS

Patient characteristics are presented in table 1. Based on 2x2 tables (table 2), sensitivity ranged from 82.7 to 86.2% and specificity from 9.1 to 36.4%. The intraclass correlation coefficient for observer agreement between mOCT values of both raters was 0.4.

CONCLUSION

In this first validation of the proposed mOCT to differentiate between low and high grade UTUC, the sensitivity is in concordance with the initial study. There is, however, a discrepancy in specificity. This discrepancy may arise from the limited number of high grade UTUC cases of the validation study, the limited sample size of the initial study or their difference in histologic reference standard. Therefore, we propose a reevaluation of the mOCT cut-off in a larger cohort to fully explore its potential for UTUC grading.

Funding: Stichting Cure for Cancer (Cure for Cancer foundation)