Diagnosis of urinary bladder urothelial carcinoma by immunocytology with p53, MCM 5, MCM 2 and Ki‐67 antibodies using cell blocks derived from urine

2019 
OBJECTIVE: Immunocytochemistry has attained a marginal role in urology so far. Combining the morphological and immunophenotypical changes of the urothelial cells retrieved from urine is a logical approach. The study aimed to analyse the diagnostic potential of immunocytological staining in the detection of high-grade and low-grade urothelial carcinoma. METHODS: Freshly voided urine was collected from 152 consecutive individuals, cytology classes were determined and cell blocks produced. A total of 77 patients were diagnosed with urothelial carcinoma and 75 patients had various benign urological conditions. Immunocytochemistry was performed using four antibodies: p53, MCM2, MCM5 and Ki-67. A diagnostic power to detect low grade and high-grade urothelial carcinoma was analysed for each antibody and their combinations with cytology. RESULTS: There were no significant differences between patients with low-grade tumours and control group. Antibodies p53 and Ki-67 slightly improved the sensitivity of urinary cytology while maintaining its specificity. The best negative predictive value was demonstrated in combinations of cytology and MCM5 (88.9%) and cytology, p53 and MCM5 (90.6%). In the diagnosis of high-grade tumours, all antibodies apart from MCM2 yielded better sensitivity and specificity than cytology alone (receiver operating characteristic curves: p53 = 0.853, MCM5 = 0.931, and Ki-67 = 0.895). Combined with cytology, the sensitivities went even higher for the cost of lower specificity. The best diagnostic performance was observed in the combination of MCM5 and Ki-67 (sensitivity = 96.2%; specificity = 80%). CONCLUSIONS: Immunocytochemistry with p53, MCM5 and Ki-67 antibodies can improve the diagnostic power of urinary cytology in the detection and follow-up of urinary bladder urothelial carcinoma.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    29
    References
    1
    Citations
    NaN
    KQI
    []