Objective: To characterize the time trend of the male liver cancer incidence rate in Japan.
Methods: We obtained data on male liver cancer incidence rates from the 'Cancer Incidence in Five Continents (CI5) Series'. Data from the population-based cancer registries of Miyagi, Osaka, Nagasaki, Hiroshima, Saga and Yamagata between 1962 and 1997 were combined and used as the data for the Japanese. To characterize the time trend in rate, we chose and combined the data on Japanese-Americans from the cancer registries of Hawaii and Los Angeles County, California between 1968 and 1997. Data on US whites who participated in the Surveillance, Epidemiology, and End Results program in 1973-1997 were obtained from the Data Series. The age-standardized incidence rate (ASR) and birth-cohort-specific rate were calculated in the three groups using a computer program in 'CI5 Vols I-VIII'.
Results: Among Japanese males in Japan, the ASR increased sharply starting in the mid 1970s and leveled off in the mid 1990s. In contrast, among both the Japanese-Americans and US whites, the ASR continued to increase throughout the observation period. Among the US whites, an increasing trend was more apparent during 1983-97 than during 1973-87. The trend by birth cohort among Japanese males in Japan clearly showed that there was a peak incidence among men aged 45-59 years. They had been born between 1931 and 1935.
Conclusions: The present calculations clarified the distinctive time trend of liver cancer between the 1970s and 1990s in Japanese males. A possible explanation for the observed trend is discussed.