Digital image analysis supports a nuclear‐to‐cytoplasmic ratio cutoff value of 0.5 for atypical urothelial cells
- 5 June 2017
- journal article
- research article
- Published by Wiley in Cancer Cytopathology
- Vol. 125 (9) , 710-716
Abstract
BACKGROUND: An elevated nuclear‐to‐cytoplasmic (N:C) ratio of ≥0.5 is a required criterion for the diagnosis of atypical urothelial cells (AUC) in The Paris System for Reporting Urinary Cytology.METHODS: To validate the N:C ratio cutoff value and its predictive power for high‐grade urothelial carcinoma (HGUC), the authors retrospectively reviewed the urinary tract cytology specimens of 15 cases of AUC with HGUC on follow‐up (AUC‐HGUC) and 33 cases of AUC without HGUC on follow‐up (AUC‐N‐HGUC). The number of atypical cells in each case was recorded, and each atypical cell was photographed and digitally examined to calculate the nuclear size and N:C ratio.RESULTS: On average, the maximum N:C ratios of atypical cells were significantly different between the AUC‐HGUC and AUC‐N‐HGUC cohorts (0.53 vs 0.43; P =.00009), whereas the maximum nuclear sizes of atypical cells (153.43 μM2 vs 201.47 μM2; P = .69) and the number of atypical cells per case (10.13 vs 7.88; P = .12) were not found to be significantly different. Receiver operating characteristic analysis demonstrated that the maximum N:C ratio alone had high discriminatory capacity (area under the curve, 79.19%; 95% confidence interval, 64.19%‐94.19%). The optimal maximum N:C ratio threshold was 0.486, giving a sensitivity of 73.3% and a specificity of 84.8% for predicting HGUC on follow‐up.CONCLUSIONS: The identification of AUC with an N:C ratio >0.486 has a high predictive power for HGUC on follow‐up in AUC specimens. This justifies using the N:C ratio as a required criterion for the AUC category. Individual laboratories using different cytopreparation methods may require independent validation of the N:C ratio cutoff value. Cancer Cytopathol 2017;125:710‐6. © 2017 American Cancer Society.Keywords
This publication has 20 references indexed in Scilit:
- Morphologists overestimate the nuclear‐to‐cytoplasmic ratioCancer Cytopathology, 2016
- Should the BK polyomavirus cytopathic effect be best classified as atypical or benign in urine cytology specimens?Cancer Cytopathology, 2016
- Atypical urothelial tissue fragments in noninstrumented voided urine specimens are associated with low but significantly higher rates of urothelial neoplasia than benign‐appearing urothelial tissue fragmentsCancer Cytopathology, 2015
- The Johns Hopkins Hospital template for urologic cytology samplesCancer Cytopathology, 2012
- The Johns Hopkins Hospital template for urologic cytology samplesCancer Cytopathology, 2012
- pROC: an open-source package for R and S+ to analyze and compare ROC curvesBMC Bioinformatics, 2011
- Accuracy of Urine Cytology and the Significance of an Atypical CategoryAmerican Journal of Clinical Pathology, 2009
- Analysis of atypical urine cytology in a tertiary care centerCancer, 2005
- Subclassifying atypical urinary cytology specimensCancer, 2000
- Current status of urinary cytology in the evaluation of bladder neoplasmsHuman Pathology, 1990