Cancer Causes and Control, 1, 133 - 141 Comparison of the .descriptive epidemiology of urinary tract cancers Susan S. Devesa, Debra T. Silverman, Joseph K. McLaughlin, Charles C. Brown, Roger R. Connelly, and Joseph F. Fraumeni, Jr. (Received 2 May 1990; acceptedin revised form 23 July 1990) We compared the descriptive epidemiology of several urinary tract cancers, utilizing incidence data from the United States and international sources. The patterns of cancers of the renal pelvis, ureter, and urethra were more similar to those of bladder cancer than to cancer of the renal parenchyma in several ways: (i) transitional cell carcinoma is the predominant histologic type in the renal pelvis, ureter, urethra, and bladder, whereas the vast majority of renal parenchyma neoplasms are adenocarcinomas; (ii) in situ tumors often appear in all these sites except the renal parenchyma; {iii) rate ratios for renal pelvis/ureter cancers among blacks and Hispanics, relative to whites, are doser to those for bladder than to those for renal parenchymal cancers; (iv) rates among US men and women for cancers of the renal pelvis and ureter are more highly correlated with those for bladder cancer than with those of the renal parenchyma across racial groups; and (v) similar correlations occur among women across geographic areas within the US and internationally. However, the patterns for cancers of the renal pelvis and ureter do not always resemble more closely those for bladder than renal parenchyma cancers and occasionally appear different from one another. These findings indicate the importance of distinguishing tumors based on specific primary site and cell type. Key words: Bladder neoplasms, kidney neoplasms, renal pelvis neoplasms, ureter neoplasms. Introduction Cancers of the urinary tract account for seven percent of all newly diagnosed cancers, 1 and their incidence is increasing, z The urinary tract includes several organs and tissues which differ in function and cellular composi- tion) The bulk of the kidneys (renal parenchyma) is composed of glandular cells that filter and secrete wastes from the blood, whereas the lower urinary tract is a collection system, composed of the renal pelvis, ureter, urinary bladder, and urethra, lined primarily by trans- itional epithelial cells. Cells of the upper and lower urinary tract may vary in their degree of exposure and response to carcinogenic agents, and some studies have pointed to differences in etiologic factors by anatomic site. <5 Cancers of the renal pelvis, ureter, and urethra have not been well studied because they are rare and, for statistical purposes, often grouped with cancers arising in the renal parenchyma. Our purpose is to compare the descriptive epidemiology of malignancies arising in the various parts of the urinary tract to determine whether cancers of the renal pelvis, ureter, and urethra more closely resemble cancers of the bladder or renal parenchyma. Materials and methods We analyzed data for more than 66,000 incident cases of urinary tract cancer, diagnosed during 1973-85, among residents of nine areas participating in the Surveillance, Epidemiology, and End Results (SEER) program. 6 The population-based registries in Atlanta, Detroit, San Francisco-Oakland, Seattle-Puget Sound, Connecticut, Hawaii, Iowa, New Mexico, and Utah include about 10 percent of the US population. Overall, the diagnosis of 96 percent of these malignancies was microscopically confirmed: more than 97 percent for cancers of the renal pelvis, ureter, and bladder, and about 88 percent for cancers arising in the renal parenchyma, a category that includes tumors labeled kidney, not otherwise specified Drs Devesa, Silverman, McLaughlin, and Fraumeni are in the Eptdemiology and Biostatistics Program, Division of Cancer Etiology, and Dr Brown and Mr Connelly are in the Biometry Branch, Division of Cancer Prevention and Control- both Divisions of the US National Cancer Institute, Bethesda MD, USA. Address repn)~t requests to Dr Devesa at Executive Plaza North - Room 415, Bethesda MD, USA 20892. © 1990 Rapid Communications of Oxford Ltd. 133