[CANCER RESEARCH 51, 2762-2767. June 1, 1991]Cellular F-Actin Levels as a Marker for Cellular Transformation: Correlation withBladder Cancer Risk1Jian Yu Rao, George P. Hemstreet III,2 Robert E. Hurst, Rebecca B. Bonner, Kyung Whan Min, and Philip L. JonesDepartments of Urology (J. Y. R., G. P. H., R. E. H.. R. B. B., P. L. J.], Pathology [K. W. M., G. P. H.J, Biochemistry and Molecular Biology ¡R.E. H., P. L. J.J. andEnvironmental Health Sciences [G. P. H., R. E. H.], University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma 73190ABSTRACTPrevious findings in cultured cells that differentiated cells had markedly higher F-actin levels than undifferentiated cells (Cancer Res., 50:2215-2220, 1990) suggested that quantitative F-actin measurements inurinary cells might provide diagnostic or prognostic information byidentifying those individuals with cells tending towards a lower degree ofdifferentiation. The feasibility of such an approach was investigated usinga risk stratification schema. Bladder wash samples were obtained from163 symptomatic patients being evaluated for bladder cancer and 41asymptomatic controls without hematuria or other symptoms consistentwith bladder cancer. F-actin levels were evaluated by flow cytometryusing a fluorescent phalloidin probe. The risk of bladder cancer wasstratified according to biopsy, either DNA ploidy by flow cytometry orquantitative fluorescence image analysis cytology, previous bladder cancer history, and hematuria. A strong correlation between the presence ofcells with abnormally low F-actin content in cells obtained by bladderwash from 38 patients and biopsy-proved bladder transitional cell carcinoma (/' < 0.001) was observed. A strong correlation was also observedbetween the presence of cells with low F-actin content and risk of bladdercancer assessed by either stratification schema (/' < 0.0001). The correlation was more consistent with the stratification by quantitative fluorescence image analysis cytology because of the 37% false-positive rateof ploidy analysis by flow cytometry among the control patients. Furtherevidence that low F-actin was correlated with cellular abnormality wasobtained from simultaneously labeling cells for F-actin and with M344antibody, a monoclonal antibody against a low-grade bladder tumor-related antigen. These studies showed that the F-actin content of theM344-positive cells was lower than that of the M344-negative cells.These results suggest that F-actin could be an early and sensitive markerfor bladder cancer detection and risk prognostication.INTRODUCTIONBladder cancer develops as multiple foci in both time andplace in a bladder that may contain wide areas of partiallyaltered cells existing along a continuum from normal to precan-cerous (1, 2). Carcinogenesis is a multistep process that mayproceed in parallel at different rates in many cells within thebladder, and long times may transpire between initiation andits attendant oncogenic activation and the appearance of macroscopic tumors (3) or the classic markers of abnormal cytology,histology, or altered DNA ploidy (4). Random bladder biopsiesto detect concurrent severe dysplasia or carcinoma in situ doesprovide a means to specifically sample the urothelium andReceived 8/28/90; accepted 3/13/91.The costs of publication of this article were defrayed in part by the paymentof page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.' This work is the second in a series. It was supported, in part, by a subcontractfrom the Cancer Institute of the Chinese Academy of Medical Sciences, Beijing,P. R. C.. for "Quantitative Fluorescence Image Analysis (QFIA) of EsophagealSamples to Detect Abnormal Levels of Nuclear Nucleic Acid," as a part of aNational Cancer Institute-sponsored "Study of the Effect of Nutritional Intervention on Intermediate Endpoints of Esophageal Cancer in Linxian, China," andby a Postdoctoral Research Fellowship awarded to J. Y. R. by the GraduateCollege of the University of Oklahoma Health Sciences Center, and in part bythe Veterans Administration and Centers for Disease Control/National Instituteof Occupational Safety and Health Grant ROI-OH02647.1To whom requests for reprints should be addressed, at Department ofUrology, Oklahoma University Health Sciences Center, P. O. Box 26901. Oklahoma City, OK 73190.identifies some patients with advanced premalignant lesionswho are likely to develop recurrences rapidly (5). Yet theobservation that hematuria can precede positive biopsies by atleast 8 years (6) suggests both a need for and the possibilitythat markers could detect elevated risk well before an abnormalhistology is detectable.Actin, one of the major components of cytoskeleton, is important in a variety of cell functions, including regulation ofcell shape, motility, secretion, intracellular transport, endocy-tosis and exocytosis, and cell division (7, 8). The phenotypicchanges resulting from transformation are reflected in distinctalterations in the cytoskeleton (9, 10), and in a previous studyusing model systems of differentiable transformed cells wedemonstrated that quantitative measurements of F-actin content provided a marker for transformation that reflected thelevel of differentiation of individual cells (11). This result indicated that F-actin might be a general marker that could reflectalterations in several oncogenic pathways and therefore mightwell be a useful biochemical marker for delineating the extentand progression of field disease well before morphologicalabnormalities become evident. A marker for bladder cancer riskwould be particularly useful in evaluating patients with unexplained hematuria and no previous history of bladder cancer.Our previous study stimulated us to investigate the possibilityof using F-actin as an early marker in clinical samples and tocompare it with ploidy and cytological markers. F-actin, DNAploidy, and proliferation index were measured by flow cytometry in 163 clinical bladder wash samples from symptomaticpatients being evaluated for bladder cancer and from 41 urolog-ical patients asymptomatic of bladder cancer. The bladder washspecimens also were analyzed by QFIA1 cytology for highlyabnormal hyperdiploid cells and abnormal visual cytology (12,13). The individuals were stratified by bladder cancer risk usingeither QFIA cytology or flow cytometry DNA results, in combination with a previous history of