The effects of the current World Health Organization/International Society of Urologic Pathologists bladder neoplasm classification system on urine cytology results

Abstract

Background: Historically, the diagnostic accuracy of urine cytology for the detection of urothelial carcinoma has been low, particularly for low-grade lesions. In 1998, the World Health Organization and International Society of Urologic Pathologists (WHO/ISUP) established a new classification system for urothelial neoplasms. It has been postulated that the accuracy of urine cytology for the detection of low-grade carcinomas should improve, because the current WHO/ISUP system classifies bladder lesions that lack any significant cytologic atypia as papillary urothelial neoplasms of low malignant potential (PUNLMP). Also, the accuracy of urinary cytology for the detection of high-grade lesions is expected to decrease, because the criteria for high-grade lesions is lowered, placing most of the previous WHO Grade 2 tumors into the current high-grade carcinoma category.

Methods: One hundred bladder biopsies that were classified according to the previous WHO classification system were examined by one pathologist and were reclassified according to the new WHO/ISUP scheme. All biopsies had corresponding urine specimens that had been diagnosed previously by a different cytopathologists. The patients' previous WHO and current WHO/ISUP surgical diagnoses were compared with the corresponding urine cytology. In addition, cytospins obtained from 40 patients with a histologic diagnosis of low-grade urothelial carcinoma (n = 20 patients) and high-grade urothelial carcinoma (n = 20 patients) according to the WHO/ISUP classification system were reviewed to identify any outstanding cytologic features.

Results: According to the original WHO classification system, there were 26 patients with Grade 1 transitional cell carcinoma (TCC), 61 patients with Grade 2 TCC, and 13 patients with Grade 3 TCC. The corresponding cytology was positive in 11 of 26 Grade 1 tumors, 28 of 61 Grade 2 tumors, and 12 of 13 Grade 3 tumors. After the reclassification, there was 1 papilloma, 12 PUNLMP lesions, 50 low-grade urothelial carcinomas, and 37 high-grade carcinomas. The cytology was positive in 0 of 1 papillomas, 5 of 7 PUNLMP lesions, 18 of 50 low-grade carcinomas, and 28 of 37 high-grade carcinomas. In addition, morphologic uniformity and cytoplasmic homogeneity (50.0% and 45.0%, respectively) were seen more commonly in low-grade bladder tumors.

Conclusions: The diagnostic accuracy of urine cytology for high-grade lesions decreased, as expected; however, cytologic detection of low-grade urothelial carcinomas was remarkably lower than expected. In addition, no outstanding cytologic features could be identified to make a definitive diagnosis.