Although the incidence of testis cancer has risen markedly in many Western populations over the past half-century, it is not clear whether rates in other populations also have increased. To clarify this issue, we examined testis cancer incidence rates over the 25-year time period of 1973-1997 for selected populations around the world. Age-standardized incidence rates for 21 registries in the Americas, Asia, Europe and Oceania over successive 5-year time periods were obtained from volumes [4][5][6][7][8] Among many Western populations, testicular cancer incidence increased markedly over much of the past 20th century, 3,4 although recent data from some cancer registries suggest that rates may be plateauing. 5,6 A divergence in trends for different histologic subtypes of testis cancer has also been reported. The majority of testis cancers are germ-cell tumors, which are grouped histologically into seminomas and nonseminomas. 3 In Canadian and U.S. white populations, the incidence of nonseminoma has been increasing less rapidly than that of seminoma. 7,8 No adequate explanations for these trends have been identified, as the etiologies of these malignancies remain largely unknown.As investigations into secular trends of testis cancer have generally focused on populations of European ancestry, it is not clear whether rates in low-incidence populations also have increased over time. A study on international cancer trends between 1973-1987 by the International Agency for Research on Cancer (IARC) clearly demonstrated rising testis cancer rates in Japan and Puerto Rico but not in other low-incidence registries. 4 Since that time, IARC has published rates for additional time periods, enabling the analysis of secular trends over a longer time span. 2,9 To better understand how testis cancer incidence has changed over time across different populations, we examined incidence data over the 25-year period 1972-1997 from 21 populations in the Americas, Asia, Europe and Oceania.
Material and methods
Incidence dataTo examine the secular trends in the incidence of testis cancer (ICD-8, ICD-9 186), 10,11 age-standardized (World Population) 12,13 incidence rates in 21 populations were obtained from volumes 4-8 of Cancer Incidence in Five Continents (CI5). 2,9,14,15,16 The CI5 volumes include incidence data reported by selected populationbased cancer registries covering areas within Asia, Oceania, Africa, Europe and the Americas. Volumes 4-8 generally provided data for the 5-year time periods 1973-1977, 1978-1982, 1983-1987, 1988-1992 and 1993-1997, respectively. Rates for volumes 5-8 were based upon the ICD-9 classification of testis cancer, and volume 4 used the ICD-8 classification; there were no changes in the coding of testis cancer between the eighth and ninth ICD revisions. Incidence rates for different histologic subtypes of testis cancer (seminoma, nonseminoma, other/unspecified) were abstracted from volume 8.Populations were chosen for inclusion in our analysis on the basis of the following criteria: (i) the availability ...