Section 3 -

Urine Cytology – Beyond Morphology Eva M. Wojcik Loyola University Medical Center Chicago

Urinary bladder cancer is the second most common malignancy of the genitor-urinary tract. According to the American Cancer Society there will be over 60,000 new cases diagnosed in 2006 and over 13,000 people will die in the United States of this disease. The majorities of newly diagnosed urothelial carcinomas (UC) are low grade and low stage and usually treated by local resection. However, approximately 70% of them recur and approximately 10% progress to higher stage of higher grade. Therefore, UC is a challenging condition for both urologists and pathologists. For the urologist, the challenge is to predict at an early stage which patient will have further tumor occurrences or will develop invasive properties. For the pathologist, the challenge is mainly to detect low grade lesions. By definition, the nuclear differences between low grade urothelial carcinoma and normal urothelium are subtle and very subjective. Additionally, reactive processes due to treatment, lithiasis of the urinary tract or viral infections may be difficult to distinguish from neoplastic processes. A sensitive non-invasive test to detect bladder cancer remains an elusive but desirable goal. Although urine cytology is highly sensitive for detection of high grade UC, its sensitivity for low grade cancer remains unacceptably low. Consequently, invasive cystoscopy remains the mainstays for the diagnosis of bladder urothelial carcinoma and it is used to monitor patients with high risk for recurrence and progression. Therefore there is an obvious need for a development of additional more sensitive tests that could detect urothelial neoplasia. In this context, a number of techniques, including DNA ploidy, morphometry, immunohistochemistry, cytogenetics, urine chemical assays and lately, molecular techniques have been proposed to be used either in conjunction with cytologic examination or even to completely replaced cytology. DNA Ploidy by Flow Cytometry, Static Image Analysis and Laser Scanning Cytometry Evaluation of DNA ploidy either by flow cytometry, static image analysis or laser scanning cytometry has been one of the first adjuvant tests in urine cytology. These techniques measure a total amount of DNA and are only able to detect a significant and numerous chromosomal changes. This test has been used for both diagnostic and prognostic purposes. However, an abnormal DNA ploidy is not automatically associated with malignancy since low grade urothelial carcinomas are diploid and only high grade tumors are aneuploid. As a result, aneuploidy is a strong indicator of high grade malignancy as well as carcinoma in situ. Also aneuploidy in conjunction with suspicious cytology is highly predictive of tumor recurrence. In addition, it has been shown that DNA ploidy analysis provides independent prognostic information. In addition, there are numerous operator-dependent and independent factors that affect the results. It has been shown that the superficial cells, seminal vesicle cells and cells infected with Polyoma virus are causes of false positive results and these cells should be eliminated from any analyses evaluating DNA ploidy. DNA Ploidy by Fluorescence in Situ Hybridisation The understanding of the genetic changes that are associated with bladder cancer initiation and progression has increased tremendously in the last twenty years. There are two separate genetic pathways leading to the development of UC. The most common one is associated with a loss of part or all of chromosome 9 and it leads to noninvasive, papillary tumors that often recur and rarely progress. However, if this pathway converts to the other much less common one, the disease will progress to a high grade often invasive tumor. This progression is associated with an increased chromosomal instability and aneuploidy. Fluorescence in situ hybridization (FISH) has been demonstrated as a viable method for determination of chromosome specific anomalies in cells obtained from urine specimens for early tumor detection or recurrence. Recently, a multi-color FISH Probe Mixture designed for interphase cell analysis for detection and quantification of chromosome 3, 7, 17 and the 9p21 region has been made commercially available (UroVysion ä Multi-color FISH Probe Mixture, Vysis/Abot, Des Plains , IL ). In the initial study performed by Sokolova et al. (J Mol Diagn 2000;2:116), a number of potential probes was studied. The four probes with the greatest sensitivity for UC detection were selected. These include three Centromeric Enumeration Probes (CEPs) for chromosome 3 (red), 7 (green) and 17 (aqua), and one Locus-Specific Identifier (LSI) probe for 9p21 (yellow). A minimum of 25 morphologically abnormal cells is evaluated. The test is positive if > 4 cells show gains of > 2 chromosomes. In addition, loss of 9p21 signals in > 12 cells is also considered positive. In the last few years, since the FDA approval, the UroVision has gained a significant popularity and has been used by numerous laboratories. Also, by now there is an impressive body of literature validating this test. FISH outperformed cytology across all stages and grades of UC in all published comparative studies. Significantly, depends what chromosomal changes are present, we can predict either recurrence (deletion of 9p21) or progression (aneuploidy). Morphometry Morphometry has been defined as the "quantitative description of a structure". In practice, this term is usually applied to quantitative techniques that measure features of size, shape, and texture in two dimensions and/or spatial relationships from cells or other tissue structures. The need for measurement comes from the recognition that interobserver and intraobserver diagnostic decisions are poorly reproducible. Morphometry has several advantages over conventional visual assessment: objectivity, reproducibility and the ability to detect changes too subtle to be visually appreciated in individual cells. Therefore morphologic diagnostic accuracy and precision can be improved by applying this technique. A practical application of morphometry has been utilized by a group from the Netherlands which developed a quantitative karyometric cytology system – QUANTICYT that is based on evaluation of DNA content and nuclear shape features. The reported sensitivity of this system ranges between 59% - 69% and specificity between 72% to 93%. Biomarkers (FDA- Approved) BTA (Polymedco, Redmont , WA ) – Bladder tumor antigen detects the presence of basement membrane antigens that have been isolated and characterized in the urine of bladder cancer patients. The original BTA test that is no longer commercially available is a qualitative latex agglutination assay. This test has been replaced by a qualitative, point-of-care test BTA Stat and a quantitative BTA Trak assay that detects human complement factor H-related protein. The overall sensitivity ranges from 50% to 80% and specificity from 50% to 75%. NMP 22 (Matritech, Newton , MA ) - is an enzyme immunoassay for the quantification of nuclear matrix proteins that comprise the internal structural framework of the nucleus. The antibodies in this assay recognize two domains of the nuclear mitotic apparatus protein. There are two tests – laboratory-based quantitative assay and point-of-care qualitative test. There are wide ranges of reported sensitivities (68% - 88%) and specificities (34% - 76%). UCyt/ImmunoCyt (Diagnocure, Quebec City, Canada) is based on three monoclonal antibodies (19A211 antigen {HMW CEA, Texas red} and the M344 and LDQ10 {cytoplasmic mucin antigens expressed in low-grade bladder cancer cell, fluorescein}). The test is performed on voided urine. Results are verified with morphology. The sensitivity and specificity are reported to be 77% and 84%, respectively. Accu-Dx (Intracel Corp., Rockville , MD ) detects fibrinogen-fibrin degradation products in urine. Its sensitivity and specificity are 68% and 86%, respectively. This test is currently not commercially available. Other Markers Telomerase – Telomers are the terminal ends of eukaryotic chromosomes and their length decreases at each cell cycle. The gradual loss of telomers causes chromosomal instability that leads to cell senescence. Telomerase is a protein/RNA complex that prevents this process and make cell immortal. Cancer cells have a high level of telomerase. Telomerase activity is measured using a telomeric repeat amplification protocol (TRAP). This method requires a minimum 30 ml of urine. The reported sensitivity ranges from 56 to 90% and specificity ranges from 70 to 96%. HA-HAase – This test measures urinary hyaluronic acid and hyaluronidase by two similar ELISA-type assays. Combining the two tests into the HA-HAase test resulted in the detection of bladder cancer with overall 91% sensitivity and 93% specificity. Proteomics - The main proteomic technology platform, two-dimensional gel electrophoresis, is used to separate complex protein mixtures allowing individual protein spots on the gel to be identified by mass spectrometry. If two related samples are compared, for example urine from a normal healthy patient and urine from a patient with bladder cancer, differences in the abundances of particular proteins may be evident. Using this method five potential novel biomarkers and seven protein clusters have been identified. The reported sensitivity and specificity are 87% and 66%, respectively. EGF-R - is a member of the TKI (thyrosine kinase receptor) family and is over-expressed in 81% of primary bladder cancers and in 67% of metastases. Expression is correlated with stage, grade and survival. HER2/neu - its over-expression is reported in the muscle-invasive urothelial carcinoma of the bladder. HER2/neu is a member of the tyrosine kinase family and encodes a 185-kDa transmembrane protein. IL-8 - is a leukocyte chemo-attractant and induces angiogenesis. Sources of IL-8 include urothelial cells, endothelial cells, mast cells, neutrophils, T cells and macrophages. IL-8 levels are greater in patients with UC compared with those with successfully treated UC. Level increases with higher stage indicating that it is greater in more invasive tumors and it is reduced after treatment with Bacille Calmette Guerin or mitomycin. IL-4 – it induces the activation and differentiation of B cells; inhibits macrophage activation and may be involved in cancer formation; IL-4 gene intron-3 polymorphism is associated with UC. E-cadherin - Abnormal expression of E-cadherin has been associated with more rapid progression and reduced survival. The soluble form (E-cadherin, 80kDa) is elevated in UC (among other tumors) and is found in both serum and urine. It is up-regulated in papillary tumors and itsexpression gradually lost in high grade, invasive carcinomas. Uroplakin II is a membrane protein which is expressed in transitional cell carcinoma and associated with differentiation/grade and seems to be useful for detection of micro-metastases. Survivin - a unique member of the inhibitor of apoptosis (IAP) protein family. Survivin is highly expressed in urine but is undetectable in non-malignant tissues, suggesting a potential role in tumor genesis. References: Badelment RA, Kimmel M, Gay H, et al: The sensitivity of flow cytometry compared with conventional cytology in the detection of superficial bladder carcinoma. Cancer 59: 2078-2085; 1987. Melamed MR: Flow cytometry for detection and evaluation of urinary bladder carcinoma. Semin Surg Oncol 8: 300-307; 1992. Murphy WM, Emerson LD, Chandler RW, et al: Flow cytometry versus urinary cytology in the evaluation of patients with bladder cancer. J Urol 136: 815-819; 1986. Koss LG, Eppich EM, Melder KH, Wersto R: DNA cytophotometry of voided urine sediment . Anal Quant Cytol Histol 5: 398-403; 1987. Wojcik EM, Bassler TJ, Orozco R: DNA ploidy of seminal vesicle cells. A potential diagnostic pitfall in urine cytology. Analyt Quant Cytol Histol 21: 29-34, 1999 Wojcik EM, Brownlie RJ, Bassler TJ, Miller MC: Superficial urothelial cells (umbrella cells) - a potential cause of abnormal DNA ploidy results in urine specimens. Analyt Quant Cytol Histol 2000; 22:411-415 Wojcik EM, Bridges VJ, Miller MC, O'Dowd GJ: The influence of season on the incidence of DNA hypodiploidy in urinary cytology. Cytometry (Comm Clin Cytom) 42: 218-220, 2000 Hendricks JB, Wojcik EM: Quantitative cytology and histology by Laser Scanning Cytometry. Clin Immunol Newsletter 18: 21-26, 1998. Wojcik EM, Saraga S, Jin JK, Hendricks JB: Application of Laser Scanning Cytometry for the evaluation of DNA ploidy in routine cytologic specimens. Diagn Cytopathol in press. Wojcik EM, Miller MC, O'Dowd GJ, Veltri RW: The value of quantitative nuclear grading in differentiation of normal urothelial cells from low and high grade transitional cell carcinomas. Analyt Quant Cytol Histol. 20: 69-76, 1998 Wojcik EM: Diagnostic and prognostic application of computer-based morphometry in transitional cell carcinoma of the urinary bladder. UroOncol 2000; 1: 43-50. Skacel M, Fahmy M, Brainard JA, et al.: Multitarget fluorescence in situ hybridization assay detects transitional cell carcinoma in the majority of patients with bladder cancer and atypical or negative urine cytology. J Urol 2003 169(6), 2101-2105. Liou LS: Urothelial cancer biomarkers for detection and surveillance. Urol 2006; 67:25-34 Grossman HB, Messing E, Soloway M, et al.: Detection of bladder cancer using a point-of-care proteomic assay. JAMA 2005;293:810-815 Lokeshwar VB, Obek C, Pham HT, et al.: Urinary hyaluronic acid and hyaluronidase: markers for bladder cancer detection and evaluation of grade. J Urol 2000; 163:348-356 Halling KC, King W, Sokolova IA, et al.: A comparison of BTA stat, hemoglobin dipstick, telomerase and Vysis UroVision assays for detection of urothelial carcinoma in urine. J Urol 2002;167:2001-2006