Dear Sir,Jonsson et al. 1 analyzed the association of prostate cancer with height in a cohort of 8,998 male twins. In this cohort, in which 331 prostate cancer cases occurred, height was not associated with risk of prostate cancer. We examined the association of body height with prostate cancer incidence in a cohort of U.S. twins. The National Academy of Sciences-National Research Council (NAS-NRC) Twin Registry is a national populationbased twin registry in the United States that includes 15,924 white male twin pairs born in the years 1917-1927, both of whom served in the armed forces. 2 Zygosity of these twins was determined by questionnaires, eye and hair color, fingerprints, anthropometrics and, if possible, blood type. 3 In 1972, men in this twin registry received a questionnaire in which they were asked about information on anthropometrics: 27,502 twins were mailed the questionnaire, and 83.9% sent it back. 4 In our analysis, we included 9,922 men (4,961 twin pairs), among which 430 prostate cancer cases had occurred. Prostate cancer cases were ascertained up to 1994 by questionnaire, telephone screening interview, mortality records and computerized inpatient records of Medicare and the Department of Veteran Affairs. 5 We used generalized estimating equations (GEEs) to calculate age-adjusted odds ratios (ORs) of having prostate cancer in taller compared to shorter men separately by zygosity and adjusted for zygosity overall.We observed statistically significantly higher odds of prostate cancer in taller men compared to shorter men [fourth quartile (>1.80 m) compared to first quartile (<1.72 m) OR 5 1.45, 95% confidence interval (CI) 1.04-2.03; p trend 5 0.02]. Similar, although not statistically significant, associations were seen in monozygotic (OR 5 1.54, 95% CI 0.95-2.49; p trend 5 0.06) and dizygotic (OR 5 1.35, 95% CI 0.85-2.14; p trend 5 0.20) twins. In the analysis, we took into account correlated individuals. When we performed a matched analysis for monozygotic twin pairs who were discordant for prostate cancer, to directly take into account shared genetics, the results were similar to those of the GEE analysis (data not shown).Height has been positively associated with prostate cancer in several cohort studies, 6-13 with effect estimates ranging 1.14-1.80. It has been suggested that height and prostate cancer are associated because prostate development starts at the same age as adolescent growth in height. Men who are exposed to higher levels of testosterone and growth-promoting factors, such as IGF-I, are taller and might have a higher risk of prostate cancer because these same factors stimulate prostatic epithelial cell division.Yours sincerely,