Transitional cell carcinoma of the bladder: diagnosis and prognosis

Artificial neural network in diagnosis of urothelial cell carcinoma in urine cytology

AIMS AND OBJECTIVE

To build up an artificial neural network (ANN) model in the diagnosis of urothelial cell carcinoma (UCC) in urine cytology smears.

MATERIAL AND METHODS

We randomly selected a total of 115 urine cytology samples, out of which 59 were histopathology proven UCC cases and remaining 56 were benign cases from routine cytology samples. All the carcinoma cases were proven on histopathology. Image morphometric analysis was performed on Papanicolaou's stained smears to study nuclear area, diameter, perimeter, standard deviation of nuclear area, and integrated gray density. Detailed cytological features were also studied in each case by two independent observers and were semi-quantitatively graded. The back propagation ANN model was designed as 17-11-3 with the help of heuristic search. The cases were randomly partitioned as training, validation, and testing sets by the program. There were 79 cases for training set, 18 cases for validation set and 18 cases for test set.

RESULT

In the training set, ANN was able to diagnose all the malignant and benign cases. In the test set, all the benign and malignant cases were diagnosed correctly. However, one of the low grade cases was diagnosed as high grade UCC by ANN model.

CONCLUSIONS

We successfully built an ANN model in urine from the visual and morphometric data to identify the benign and malignant cases. In addition, the system can also identify the low grade and high grade UCC cases.

Expression of cytokeratin 20 in urine cytology smears: a potential marker for the detection of urothelial carcinoma

BACKGROUND

Urine cytomorphology is one of the oldest methods for screening and monitoring patients with transitional cell carcinoma (TCC). Sensitivity of urine cytology is relatively low. Ancillary techniques on urine sample may increase the sensitivity.

AIM

To explore the utility of cytokeratin 20 (CK20) immunostaining in identifying malignant cells in urine cytology smears.

MATERIALS AND METHODS

Fourteen cases each of confirmed TCC and benign urinary cytology along with five cases of atypical cells in urine were immunostained with a monoclonal CK20 antibody. Of 14 cases of TCC, 12 showed strong positive staining with the antibody. All benign cases were negative except for a few cases in which the umbrella cells were weakly to moderately positive. In all five cases of atypical urine cytology the atypical cells stained positive with the antibody. These cases were later confirmed as TCC on histopathology of bladder wall biopsy.

CONCLUSION

CK20 is an important biomarker that can be used to identify TCC in urine cytology smears. It is particularly useful in those cases where malignancy cannot be confirmed by morphology alone.

Prognostic value of keratin subtyping in transitional cell carcinoma of the upper urinary tract

To investigate the prognostic value of keratin subtyping in invasive transitional cell carcinomas (TCCs), we performed a systematic study applying 15 different monoclonal keratin antibodies on 53 upper urinary tract TCCs using a tissue microarray technique. Immunoreactivity was correlated with pT stages and tumour grades using the Fisher's exact test. Impact on disease-free survival was analysed using the Kaplan-Meier method and compared by the log-rank test. Immunoreactivity for keratins 5/6, 6, 7, 8, 13, 14, 17, 18, 19, 20, low molecular weight (LMW) keratins (8, 18) and high molecular weight (HMW) keratins (1, 5, 10, 14) was detected in varying quantities. Regarding semi-quantitative assessment, a prognostic impact was found for keratins 5/6, 7, 8, 13, 17, 20 and HMW keratins, with reduced expression or loss of immunoreactivity being significantly associated with disease progression. With respect to analysis of staining patterns, the retention of a basally accentuated labelling for keratin 5/6 and HMW keratin as well as a superficially accentuated labelling for keratin 20 was significantly associated with a favourable outcome. In conclusion, this investigation is the first to demonstrate a possible prognostic value for keratin subtyping in invasive (upper urinary tract) TCCs with respect to metastasis-free survival. Further studies, however, are needed to substantiate our results.

Identification of genes up-regulated in urothelial tumors: the 67-kd laminin receptor and tumor-associated trypsin inhibitor

Studies investigating changes in gene expression in urothelial carcinoma have generally compared tumors of different stages and grades but comparisons between low-grade, noninvasive tumors and normal urothelium are needed to identify genes involved in early tumor development. We isolated the urothelium from a low-grade tumor and corresponding normal mucosa by laser capture microdissection on frozen sections. The RNA extracted was amplified to generate suppressive subtractive cDNA libraries. Random sequencing of cDNA clones identified approximately 100 unique species. Of these 83% were known genes, 15% had homology to genes with an unknown function in humans, and 2% did not show homology to any published gene sequence. Two of the known genes, the 67-kd laminin receptor (67LR) and tumor-associated trypsin inhibitor (TATI), had previously been associated with metastatic progression in many tumor types, although 67LR has not been investigated in urothelial tumors. Immunolabeling of the original tissue with antibodies against these two genes confirmed overexpression, validating our strategy: 67LR was not expressed in the normal urothelium but was present in the tumor, whereas TATI expression was confined to umbrella cells in the normal urothelium, but extended to all cell layers in the tumor. We investigated both markers further in a separate series of tumors of different stages and grades. TATI was more consistently overexpressed than 67LR in all tumor grades and stages. Levels of secreted TATI were significantly higher in urine samples from patients with tumors compared to controls. Our strategy, combining laser capture microdissection and cDNA library construction, has identified genes that may be involved in the early phases of urothelial tumor development rather than with disease progression, highlighting the importance of comparing tumor with normal rather than just tumors of different stages and grades.

Urine: beyond cytology for detection of malignancy

In the present review we discuss various ancillary modalities for detection of malignancies in urine samples, with an emphasis on urothelial carcinomas. Flow cytometry, bladder tumor antigen (BTA), nuclear matrix protein (NMP), matrix metalloproteinase (MMP), human chorionic gonadotrophic (HCG), telomerase, and other techniques are discussed. DNA FCM is a relatively costly and sophisticated technique. It has a practical application in the diagnosis of bladder cancer among subjects at high risk and is of value in monitoring the course of the disease and anticipating recurrence following conservative treatment. The BTA test is a simple, rapid, and inexpensive adjunct to cystoscopy and the results of the test are equivalent or superior to those of voided urinary cytology. NMP-22 immunoassay is a useful diagnostic test for predicting recurrence of urothelial malignancy. It is also a cost-effective and sensitive screening test for detecting tumor in patients with urothelial carcinoma. Beta-HCG estimation in urine samples appears to be an efficient diagnostic marker for the assessment of distant metastasis in bladder carcinoma rather than a screening test. Other ancillary techniques such as detection of expression of cytokeratin 20 by RT-PCR, MMP-9 estimation, and fluorescent in situ hybridization and telomerase activity are rarely applied clinically in routine urinary samples and are not cost-effective.

Expression of cytokeratin 20 redefines urothelial papillomas of the bladder

BACKGROUND

Most non-invasive urothelial tumours of the bladder are diagnosed as papillary carcinomas in accordance with the WHO classification and because the identification of papillomas is difficult by routine histology; some patients are therefore misdiagnosed. This practice is associated with psychological morbidity for the patient and may also skew cancer statistics. Cytokeratin 20 (CK20) is a sensitive marker of urothelial differentiation. We investigated whether this marker could be used in the identification of urothelial papillomas and used the rate of recurrence as an indicator to assess the biological behaviour of such tumours.

METHODS

In a prospective study, immunocytochemistry for CK20 was done on tumours of all patients who presented for the first time with non-invasive papillary bladder tumours. We classified the expression pattern of CK20 as normal or abnormal at the time of initial diagnosis. We recorded time to first biopsy-proven recurrence or length of follow-up when no recurrence was observed.

FINDINGS

Of 58 consecutive patients, ten had tumours with a normal pattern of CK20 expression. No patients developed further tumours during the follow-up (median 18 [range 13-28] months). By contrast, 30 (73%) of the 41 evaluable patients with tumours that showed abnormal CK20 expression developed further tumours; the median time to a second tumour was 6 (2-24) months. The only factor that had a significant effect on the outcome of patients in terms of recurrence was expression of CK20 (p<0.0001).

INTERPRETATION

Normal urothelial differentiation, as evidenced by a normal pattern of CK20 expression, is retained in a proportion of non-invasive papillary urothelial tumours and thus justifies use of the term urothelial papilloma. A large-scale study is needed to investigate the outcome of patients with such tumours.

Uroplakin gene expression by normal and neoplastic human urothelium

cDNA sequences for human uroplakins UPIa, UPIb, UPII, and UPIII were cloned and used to investigate uroplakin transcription by normal and neoplastic urothelial cells. Normal urothelium expressed mRNA for all four uroplakins, although UPIII could be detected only by ribonuclease protection assay. By in situ hybridization, UPIa and UPII were confined to superficial cells and UPIb was also expressed by intermediate cells. Cultured normal human urothelial cells showed a proliferative basal/intermediate cell phenotype and constitutive expression of UPIb only. Uroplakin expression by transitional cell carcinoma cell lines was related to their differentiated phenotype in vitro. RT4 cells expressed all uroplakins, VM-CUB-3 expressed three uroplakins, RT112 and HT1376 cells expressed only UPIb in high abundance, and COLO232, KK47, and EJ cells had no detectable expression. These results correlated with patterns of uroplakin expression in tumors. UPIa and UPII were detected superficially only in well differentiated transitional cell carcinoma papillae. UPIb was positive in seven of nine and overexpressed in five of nine noninvasive transitional cell carcinomas and was also present in four of eight invasive transitional cell carcinomas. Lymph node metastases retained the same pattern of UPIb expression as the primary tumor. Unlike the three differentiation-regulated uroplakins, UPIb may have an alternative role in urothelial cell/tissue processes.

Production of trophoblastic hormones by transitional cell carcinoma of the bladder: association to tumor stage and grade

Cancer registration statistics of economically advanced countries indicate that bladder carcinoma incidence ranks fourth in men and eighth in women, but a reliable tumor marker for predicting the disease course is still lacking. We designed an immunohistochemical study to comprehensively assess the trophoblastic hormone production profile of transitional cell carcinoma (TCC) of the bladder. Moreover, we correlated histological differentiation and tumor stages with marker expression and, finally, evaluated a potential tumor origin of hCGbeta core-fragment (hCGbetacf). To this end, formalin-fixed, paraffin-embedded tumor tissues from 104 patients with urothelial neoplasms of various histological grades (23 GI, 24 GII, and 38 GIII) and stage (19pTis, 21pTa, 29pT1, and 35pT2-T4) were analyzed by the immunoperoxidase technique using our own well-characterized monoclonal antibodies against the glycoprotein hormones human chorionic gonadotropin (hCG) and its derivatives hCGalpha, hCGbeta, hCGbetacf, luteinizing hormone (LH, LHbeta), follicle-stimulating hormone (FSH, FSHbeta), and the protein hormones placental lactogen (hPL) and growth hormone (hGH-V/N). Overall, trophoblastic hormone immunoreactivity was found in 36% of TCC. Detailed analysis showed 35% hCGbeta, 17% hCGbetacf, 9% hCGalpha, 4% hCG, and 2% hPL-positive cases. The tumors produced neither GH-N, placental GH-V, nor the pituitary gonadotropins FSH/FSHbeta and LH/LHbeta. Marker positivity significantly increased with high-grade lesions (26% GI- v 55% GIII-TCC) and advanced tumor stages (24% pTa v 63% > or = pT2). Hormone immunoreactivity was frequently observed in highly proliferating areas. Our findings, together with recent structural and clinical studies, strongly suggest that these hormones, or derivates thereof, might act as local tumor growth factors. Normal urothelium, urothelial papillomas, and carcinoma in situ showed no positive reactions. All tumors producing hCG-derived molecules were negative for the concommitantly analyzed neuroendocrine markers chromogranin A, synaptophysin, and neuron-specific enolase (NSE). In summary, one third of TCC ectopically produce trophoblastic hormones, which is specifically correlated with stage and grade. Apart from hCGbeta (97% of the marker-positive cases), the intracellular occurrence of hCGbetacf, apparently the second most frequently produced marker, was surprising, and there was also a lesser degree free hCGalpha and intact holo-hormone expression. The placental protein hormones PL and GH-V are not appropriate tumor marker candidates. Finally, our histogenetic findings support a metaplastic origin of the hCG producing choriocarcinomatous phenotype of some TCC.