A prospective comparison of prostate cancer at autopsy and as a clinical event: the Hawaii Japanese experience.

Prostate cancer was diagnosed in life among 274 of 8006 (3.6%) members of a cohort of Japanese men in Hawaii between 1965 and 1990. Only 55 (20%) of the 274 diagnosed cases died with prostate cancer, and they accounted for only 2% of the 2893 deaths that occurred among the men during this period. None of the 61 men whose tumor was found incidentally to a transurethral resection died as a result of this cancer, while it was the cause of death of 9 of 106 (8%) men with clinical cancer localized to the prostate. Forty-six of the 107 (43%) men with more extensive disease at the time of diagnosis died from prostate cancer. Step sectioning of the prostate identified prostate cancer in 80 of 293 (27%) autopsied Hawaii Japanese men who died after 50 years of age, reaching a frequency of 63% (10 of 16) among men over 80 years of age. The volume of 48 (60%) of these cancers was less than 150 mm3. These small tumors would probably not have been discovered in a screening program. Tumors larger than 1000 mm3 would probably be discovered using modern diagnostic procedures but were found in only 13 (4.4%) of the autopsied men. It is likely that a screening program to detect and treat such large, unsuspected tumors in this population would have had little impact upon the already low proportion of deaths due to prostate cancer among these Japanese men.