The problem of distinction between reactive and dysplastic atypical changes in the urothelium has not been resolved, even after 2004 WHO/ISUP classifications [1, 3]. Much inter and intraobserver disagreement constitute a dilemma due to lack of definite morphological criteria to diagnose CIS, dysplasia and reactive atypia.
CIS is a high-grade intraurothelial neoplasm. Patients with CIS are at significant risk for the development of invasive urothelial carcinoma, cancer recurrence, progression, and even death from bladder cancer [3–5, 13, 14]. Therefore the recognition of CIS in bladder biopsy specimens is of great importance because it not only has important prognostic implications but may also alter subsequent therapy. The histopathology of urothelial CIS may overlap with RUA resulting in diagnostic difficulty when interpreting bladder biopsies . The difficulty and importance of distinguishing CIS from benign atypias should not be underrated as malpractice cases involving the failure to recognize CIS are growing in number . Dysplastic intraepithelial lesions are another problematic area that requires distinction from CIS and which pathologists frequently feel uncomfortable evaluating on the basis of histological criteria alone.
As the distinction is therapeutically and prognostically critical, objective ancillary markers to help in the histologic differentiation between CIS and dysplasia and reactive atypia are necessary. Importance of special and immunohistochemical stains for correct diagnosis and accurate subclassification of controversial cases in other lesions of urinary tract are also emphasized recently by Jin et al  and Pradhan et al . Previous studies have demonstrated the diagnostic utility of separate p53 and CK20 immunohistochemistry (IHC) in assessing neoplasia in bladder biopsies . These markers are known to be easy to manage and interpret. Distinctive immunoreactivity patterns with antibodies against p53 and CK20 were identified for reactive atypia and CIS . Abnormal CK20 expression in urothelial cells accompanied by overexpression of p53 are considered indicators of dysplastic change in urothelial mucosa and immunostaining with p53 and CK20 may help accurately diagnose CIS [11, 18].
To our knowledge, no previous study in the literature has used DIHC of p53+CK20 in distinction of non-neoplastic and neoplastic bladder lesions. In this study, we investigated the utility and advantages of p53+CK20 DIHC as a tool for detecting synchronous expression of both markers in bladder biopsies and for objectively distinguishing the cases with CIS and dysplastic urothelial changes from reactive nonneoplastic atypia.
The nuclear overexpression of p53 by immunohistochemistry has been shown to be representative of p53 tumor suppressor gene mutation which is a common event in neoplastic urothelium [19–26]. p53 is considered to be nonexpressed in the nonneoplastic urothelium and overexpression of the p53 gene product has been reported as a marker of progression in urothelial carcinoma [24, 25, 27–29]. Mutations of the p53 gene and immunohistochemical positivity for the p53 protein have been found in 40% to 60% of urothelial carcinomas in studies of Olumi, Sidransky and Wright et al. [21, 28, 30]. In nonneoplastic and reactive urothelium, expression of p53 has been described as varying from negative to weak and patchy, predominantly in basal and parabasal intermediate cells [9, 11, 31]. In parallel to these results, 60% (21/35) of the benign/reactive group cases of our study were found to be p53(-). In contrast, p53 was considered positive by Mallofre in 80% of the CIS cases, with 70% of those cases showing positivity in 50% of the cells . Similarly, McKenney found that p53 was positive in 57% of the CIS cases, with all of the cases exhibiting positivity in more than 50% of malignant cells . In our study, 10% (1/10) of the dysplasia, 67% (6/9) of CIS and 29% (2/7) of invasive carcinoma cases showed strong diffuse p53 positivity whereas another 29% (2/7) of invasive carcinoma cases exhibited moderate p53 positivity.
If found in the cytoplasm of cells different than superficial umbrella cells and occasional intermediate cells, CK20 is considered abnormally expressed as a marker of abnormal urothelial differentiation [31, 32]. Urothelial de-differentiation, as with neoplastic change, is accompanied by expression of CK20 in all cell layers as shown by Harnden et al. [8, 32]. This change in the extent of expression makes CK20 a useful and reliable marker of the neoplastic change of urothelial cells [8, 32–36]. Immunostaining for CK20 is therefore an important addition to morphology in the diagnosis of neoplasia, especially, in the differentiation from reactive states where diagnostic difficulties are greatest [8, 37]. Cases of atypia and dysplasia that display abnormal CK20 staining should raise the possibility of CIS and be followed up appropriately. In the studies of Mckenney, Mallofre and Kunju et al it has been reported that in nonneoplastic epithelium as well as the reactive urothelium CK20 showed patchy cytoplasmic immunoreactivity in only the superficial umbrella cell layer [9, 11, 37]. In favor of those studies, in our study, we found that 92% (35/38) of benign/reactive cases were either CK20(-) or showed CK20 positivity only in the upper 1/3 urothelium. While confined to superficial umbrella cells in normal urothelium, CK20 has been reported to show diffuse full thickness staining in 72-89% of cases of CIS and has also been reported to be expressed in 22-58% of invasive urothelial carcinomas [7, 9, 11, 37–39]. However, CK20 immunoexpression was found to be nondiscriminatory in 11-28% of CIS cases due to lack of CK20 expression [7, 9, 11, 37]. In this study, abnormal expression of CK 20 was found in 90% (9/10) of dysplasia, 89% (8/9) of CIS and 71% (5/7) of IC cases whereas the rest of the cases lacked abnormal CK20 expression.