Quantitative analysis of keratin 18 in the urine of patients with bladder cancer

Diagnostic value of cytokeratin-18 as a tumor marker in bladder cancer

OBJECTIVES

The aim of the study was to compare serum levels of cytokeratin-18 of patients with bladder cancer with those of the healthy controls, and to investigate the relation between cytokeratin level and the tumor stage.

DESIGN AND METHODS

Serum cytokeratin-18 levels of 38 patients with bladder cancer and of 25 healthy people were determined. Tumor stage was T(1) in 12 patients, T(2) in 9 patients, T(3) in 10 patients and T(4) in 7 patients. The serum cytokeratin-18 levels in these cases were analyzed with respect to the stage of the tumor.

RESULTS

Cytokeratin-18 level in the patient group was found to be significantly higher than that of the control group (p < 0.010) when the groups were totally compared. However, when the levels in patients with different tumor stages were compared with that of the controls, the difference was not significant in patients with stage 1 and 2 tumors (p > 0.05). Regarding the cut off value as 4.0 ng/mL, sensitivity and specificity for serum cytokeratin-18 were found to be 53% and 72% respectively. When sensitivity was calculated with respect to tumor stages, it was 8% for T(1,) 33% for T(2,) 90% for T(3) and 100% for T(4.) On the other hand, considering higher stage (T(3) and T(4)) tumors only, the sensitivity was calculated as 94%, but the sensitivity for lower stage (T(1) and T(2)) tumors was 19%.

CONCLUSIONS

It is clear that serum cytokeratin-18 level increases in patients with bladder cancer. However, it can only be useful as a tumor marker in the diagnosis of T(3) and higher staged tumors. This study indicated that cytokeratin-18 does not have any diagnostic value in lower stage bladder cancers.

Urinary bladder cancer test: a new urinary tumor marker in the follow-up of superficial bladder cancer

OBJECTIVES

To study the diagnostic performance of the Urinary Bladder Cancer (UBC) test in patients with superficial bladder carcinoma.

METHODS

One hundred one patients in follow-up for superficial bladder cancer (pTa, pT1, carcinoma in situ) were recruited for this study. Each patient underwent cystoscopy and transurethral resection or biopsy, with subsequent histologic confirmation in the case of abnormalities. In addition, specimens were assessed with an immunoenzymometric assay for cytokeratin expression (the UBC test), and the urinary creatinine concentration was determined to correct for different degrees of urinary dilution. Different methods were applied to calculate the diagnostic value of the UBC test.

RESULTS

Both noncorrected and corrected median values of the UBC test were comparable between patients with and without a recurrent bladder tumor. The overall sensitivity, specificity, and positive and negative predictive values of the noncorrected UBC test was 20.7%, 84.7%, 35.3%, and 72.6%, respectively. For the corrected UBC test, the corresponding values were 20.7%, 79.2%, 28.6%, and 71.3%. The area under the receiver operating characteristic curve was not significantly different from 0.50, indicating no diagnostic value of the UBC test in this study.

CONCLUSIONS

The diagnostic value of this new urinary marker appears insufficient for the follow-up of patients with superficial bladder cancer.

P53 and ploidy assessed by flow cytometry in bladder washings

OBJECTIVE

To determine p53 expression in cells in bladder washings and to relate this to DNA content and clinical outcome.

PATIENTS AND METHODS

Washings from 102 patients (41 with newly diagnosed superficial tumours [pTa and pT1], 49 with recurrent superficial tumours and 12 with carcinoma invading bladder muscle) were studied. In 39 cases, the primary bladder tumour was also analysed. The rates of tumour recurrence and progression were determined for the new superficial tumours and related to both p53 expression and DNA content.

RESULTS

Cells positive for p53 were detected in 22 of 90 (24%) washings from patients with superficial bladder cancer. P53 expression correlated with tumour stage (P < 0.05), grade (P < 0.05) and abnormal DNA content (P < 0.05). The analysis of pure urothelial (cyto-keratin-positive) cells improved the detection of DNA abnormalities (P < 0.001). In 74% of cases where both washings and tumour were analysed, the results for DNA content agreed. Of 41 new superficial tumours, 27 (66%) recurred (11 were p53-positive, 16 were p53-negative, P = 0.221; 17 had abnormal DNA content, 10 were diploid, P = 0.069). Four patients progressed (one was p53-positive, P = 0.315 and all had abnormal DNA content, P = 0.072).

CONCLUSION

P53-positive cells can be detected in washings using flow cytometry and were more commonly detected in association with aneuploid tumours. At short-term follow-up, flow cytometric analysis of DNA content in washings had greater predictive value than had p53 expression. Few washings contained aneuploid cells when the primary tumour contained diploid cells, although the collection of washings is a convenient way of sampling tumour cells.

Cytokeratin shedding in urine as a biological marker for bladder cancer: monoclonal antibody-based evaluation

Cytokeratin shedding into urine was measured using a double-antibody sandwich enzyme-linked immunosorbent assay (ELISA) in 282 individuals. Samples included urine from normal controls, patients with urogenital conditions and bladder cancer patients. A monoclonal antibody prepared against cytokeratins extracted from a hyperkeratotic low grade squamous cell carcinoma (UNME/K1) was used in the assay. The results indicated reasonable levels of sensitivity (83%), specificity (67%) and overall accuracy (70%) in the detection of bladder cancer. The levels of sensitivity in detecting squamous and transitional cell carcinoma patients were 87 and 73% respectively. The low level of specificity was due to a high frequency of false positive results (55%) within the urogenital controls; this suggests that further immunochemical and immunohistopathological analyses of associated urothelial cytokeratins are required.

Bladder cancer

Bladder cancer is largely a preventable disease; epidemiologic studies indicate that the majority of cases occur as a result of cigarette smoking or occupational exposures. The impact of screening high-risk populations is uncertain, but prompt and early diagnosis is essential for optimal therapeutic results. The management of different stages of disease varies greatly and is currently in a state of evolution. The majority of cancers are superficial, of low malignant potential, and can generally be treated cystoscopically. Few studies have addressed whether intravesical therapy will prevent high-risk patients with superficial disease from developing muscle invasion or distant metastases. Controversy exists as to optimal management of patients with invasive cancers. Improvements in technique and methodologies of urinary diversion have made cystectomy more tolerable for patients. Although cystectomy remains the "gold standard," probably not all patients require it. The careful selection of those patients whose bladders can be preserved is currently being evaluated. Combination chemotherapy for patients with metastatic bladder cancer is very active, appears to prolong survival, and may offer durable remissions to some patients. Whether chemotherapy will permit greater numbers of patients with invasive bladder cancer to be cured and bladders preserved remains to be determined.

Antigen common to several species, recognized by a rat monoclonal antibody raised against syngeneic rat bladder tumor

A rat monoclonal antibody (MAb) termed RS-II (Ig M) was obtained by syngeneic immunization with rat bladder tumor cells induced by N-butyl, N-hydroxybutylnitrosamine (BBN). Immunocytochemical analysis showed that RS-II is also reactive with mouse, dog and human bladder tumor-cell lines and some other human tumor-cell lines but not myeloma or leukemia cells. Immunohistochemical examination of paraffin-embedded tissues has shown that RS-II is reactive with mouse, rat, dog and human bladder tumors (5/5, 5/5, 1/1, 31/49) and some other tumors, but not with normal human urothelium or normal rat tissues. The antigen is expressed on the majority of low-grade or well-differentiated tumors, but less on advanced, invasive tumors. The immunofluorescence staining pattern of cultured cells was cytoplasmic and filamentous. Immunoblotting revealed that the antigen is a Mr 54,000 to Mr 56,000 protein which was extracted with difficulty from the cultured cells with Triton X-100. The antigen was also detected in the culture supernatant by means of ELISA. Our results suggest that the epitope is expressed on cytoskeletons common to a wide variety of mammalian cells at a certain stage of differentiation.

Characterization of urinary keratin number 18 using a new assay

The levels of the low-molecular weight keratin 18 appearing in the voided urine of 130 healthy volunteers was studied using a new immunoradiometric assay for keratin 18 (IRMAK-18). Keratin 18 was detected in 79 urines with an average level of 2.5 +/- 4.7 ng./ml. The assay does not detect complex epidermal keratins or non-keratin intermediate filaments. Only 50% of the purified keratin added to urine could be recovered. The recovered keratin had the same molecular weight as the input molecules, implying that substances in the urine physically inhibited the assay rather than destroyed the keratin. The independent variables of cell count, dead cells, and protein concentration were not found to be important in evaluating the results. However, the urine concentration, as measured by creatinine, might be an important consideration for assessing the levels of keratin detected.