Japanese Journal of Clinical Oncology Advance Access published online on February 12, 2008
Japanese Journal of Clinical Oncology, doi:10.1093/jjco/hym170
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© The author (2008). Published by Oxford University Press. All rights reserved
Height, Urban-born and Prostate Cancer Risk in Japanese Men
1 Division of Community Health, Tohoku University School of Health Sciences, Sendai
2 Division of Urology, Miyagi Cancer Center Hospital, Natori
3 Division of Pathology, Miyagi Cancer Center Research Institute, Natori
4 Division of Nursing, Miyagi Cancer Center Hospital, Natori
5 Division of Epidemiology, Miyagi Cancer Center Research Institute, Natori, Japan
For reprints and all correspondence: Yuko Minami, Division of Community Health, Tohoku University School of Health Sciences, 2-1 Seiryo-machi, Aoba-ku, Sendai 980-8575, Japan. E-mail: adym{at}mail.tains.tohoku.ac.jp
Received September 11, 2007; accepted December 7, 2007
 |
Abstract |
|---|
 TOP Abstract INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Background: Height and early-life environments have received attention as risk factors for prostate cancer. However, the evidence is sparse in Japan. To elucidate the associations of height and early-life factors with prostate cancer risk in Japanese men, we conducted a hospital-based case–control study. In addition, to investigate whether the associations vary between prostate cancer and other major cancers, we conducted a comparative study within the same case–control study.
Methods: Study subjects consisted of 282 prostate cancer cases, 584, 461, 231, and 156 male stomach, lung, colon and rectal cancer cases, respectively, and 1730 male hospital controls, aged 50 and over admitted to a single hospital in Miyagi Prefecture from 1997 to 2003. Information on height and early-life factors including birthplace and stature at 12 years was collected using a self-administered questionnaire. Odds ratios (ORs) and 95% confidence intervals (95% CIs) were estimated for each exposure variable.
Results: A significant positive association was found between height and prostate cancer risk (OR, 1.52; 95% CI, 1.00–2.31, between the highest and lowest quartiles; P for trend = 0.03). A significant association of urban-born with prostate cancer risk was also found (OR, 1.48; 95% CI, 1.03–2.13). Analyses by stage revealed that height might be more strongly associated with the risk of advanced prostate cancer. For other major cancers, no significant association with height and early-life factors was observed.
Conclusions: Height and early-life factors were significantly associated with prostate cancer risk. Compared with other major cancers, these associations were specific to prostate cancer.
Key Words: case–control study • growth • height • Japanese • prostate cancer
|
INTRODUCTION |
|---|
|
TOPAbstractINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Japanese men have a much lower risk of prostate cancer compared with men in Western countries (1,2). However, prostate cancer incidence rates in Japanese men have substantially increased over several decades (3,4). During 1975–2000, the age-adjusted prostate cancer incidence rate to the world population increased 3-fold (3). The age-adjusted rate in 2000 was 14.9 per 100 000 in Japan (3).
In Western countries where prostate cancer is one of the most common male cancers, various putative risk factors have been presented (5,6). During recent decades, height and early-life environments have received attention as risk factors for prostate cancer. Some Western studies have provided evidence that height and early-life factors affecting child growth might be related to prostate cancer risk (7,8). On the basis of such findings, the increase in the prostate cancer incidence rate in Japan may be correlated to the trend of height and early-life factors in the population. Further, the patterns of the correlation in prostate cancer may be different from those in other major cancers in which the incidence has been gradually changing. However, the associations of height and early-life factors with prostate cancer risk in Japanese men are poorly understood.
To investigate the associations of height and early-life factors with prostate cancer risk in Japanese, we conducted a hospital-based case–control study in Miyagi Prefecture, Japan. In the study, birthplace and stature at 12 years of age were taken as early-life factors, which may be related to child growth. Analyses were separately conducted for localized and advanced type, as well as for total prostate cancer. In addition, to investigate whether the effects of height and early-life factors are different between prostate cancer and other major cancers, we analyzed other major cancer sites within the same hospital-based case–control study and compared the results with those of prostate cancer.
|
SUBJECTS AND METHODS |
|---|
|
TOPAbstractINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Data Collection
In January 1997, we began a questionnaire survey in connection with the present study. Information on lifestyle, personal history and anthropometric measures has been collected from all patients on their first admission to the Miyagi Cancer Center Hospital (MCCH), using a self-administered questionnaire. The questionnaire is distributed to the patients on the day of reservation for first admission before any definite diagnosis or treatment at the MCCH. Between January 1997 and December 2003, the questionnaire was given to 12 929 first admitted patients; of which, 11 682 responded (response rate, 90.4%). Details of the data collection procedures have been described elsewhere (9).
The questionnaire covers items on demographic characteristics, including birthplace and usual occupation (if retired, former occupation), general lifestyles before the development of current symptoms, personal and family histories of cancer and other diseases, current height and weight, and subjective judgment as to stature at 12 years. Around 12 years, pubertal growth begins and physical changes appear (10). Items related to the referral base (without referral, with referral, from screening, other) were also included.
This study was approved by the ethical review board of Miyagi Cancer Center.
Study Subjects
Since prostate cancer is a disease of older men, cases and controls were selected from male patients aged 
50, who responded to the above-mentioned questionnaire survey. To identify incident cases of cancer, a list of the patients was linked with the hospital cancer registry which records all cancer cases confirmed by clinical, cytological and/or histopathologic examination at the MCCH. Two-hundred and eighty-two prostate cancer cases were identified. Data on clinical stage of prostate cancer were also retrieved from the registry. Further, 584 stomach, 461 lung, 231 colon and 156 rectal cancer cases were identified.
Controls were selected from male non-cancer patients aged 50 without any past history of cancer. Patients with benign tumor have been classified as non-cancer patients in the present study. Accordingly, 1730 non-cancer patients were identified as controls. The diagnoses among controls were as follows: benign tumor 814 (47.0%), cardiovascular disease 88 (5.1%), digestive tract disease 280 (16.2%), respiratory tract disease 85 (4.9%), urological disease 88 (5.1%), other benign disease 107 (6.2%) and no abnormal findings 268 (15.5%). The sites of benign tumor were stomach 99, lung 5, colon 468, rectum 22, urological organ 13, bone or connective tissue 109 and other 98. The diagnoses in the controls were regarded to be unrelated to height and early-life factors.
The response rate to the questionnaire survey according to case and control was 91.6% for prostate, 94.3% for stomach, 82.9% for lung, 94.7% for colon, 94.5% for rectum and 91.0% for control group.
Assessment of Height and Early-Life Factors
Current height was self-reported in the questionnaire survey. However, since there were some incomplete or missing data in the self-reported data, we compensated those with measured height retrieved from the clinical database at the MCCH. Finally, the data including 3114 self-reported height (prostate 258, stomach 520, lung 395, colon 215, rectum 143 and control 1583) and 265 measured height (prostate 13, stomach 57, lung 61, colon 14, rectum 12 and control 108) were used for statistical analysis. The correlation coefficient between self-reported and measured height in a subsample in the present study was 0.92.
For early-life factors, birthplace comprises five categories (urban, rural, mountainous, seaside and other area) and stature at 12 years comprises three categories (short, moderate and tall). Among them, stature at 12 years was subjectively rated by study subjects, i.e. self-rated.
Statistical Analysis
To evaluate the associations of height and the early-life factors (birthplace, stature at 12 years) with the risk of cancer in prostate and the four other major sites, we used unconditional logistic regression models (11). In the models, adjustments were made for year of birth, year of survey, family history of index cancer in parents and siblings and other related factors including major risk factors for male cancer. Height was categorized into four classes, and odds ratios (ORs) and 95% confidence intervals (95% CIs) for each height category were estimated, taking the lowest category as a reference category. In the categorization of height, we set similar cut points used in a cohort study recently conducted in Japan (
159, 159 < to < 164, 164 to < 168, 168 cm) (12). The cut points are relatively consistent with quartile points in control subjects. For the height categories, linear relation was also tested by treating median values of the category as a continuous variable. Birthplace was regrouped into two classes (urban, rural and other), and OR for urban-born was estimated. In the analysis on stature at 12 years of age, ORs for each category were estimated, taking ‘short’ as a reference category.
Additionally, prostate cancer cases were divided into localized and advanced groups based on clinical stage, and ORs were estimated according to the subgroup. Localized cancers include organ-confined cases with no evidence of extracapsular extension. Advanced cancers comprised cases with extracapsular extension or metastatic spread.
|
RESULTS |
|---|
|
TOPAbstractINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Table 1 shows distributions of background characteristics in study subjects. Cases of prostate and lung cancer are older compared with controls. Although the distribution of the area of residence was comparable between stomach cancer cases and controls, slight differences were observed between cases in other sites and controls. There were variations in referral patterns between cases and controls. About 15% of prostate cancer cases were derived from screening. The frequencies of ever-smokers and alcohol drinkers were low in prostate cancer cases. In all cancer sites listed, the frequency of cases with a family history of index cancer was higher than that of controls. Some differences in occupational history were observed between cases and controls. In cases of prostate and colon cancer, the frequency of professional workers was high.
|
Table 2 shows ORs and 95% CIs of height for each cancer site. The ORs were adjusted for year of birth, year of survey and confounding variables, which were listed in Table 1. In the analysis, unknown data in histories of smoking and alcohol drinking were, respectively, classified into one category and entered into the models (ever, never, unknown). Occupational history was reclassified into two categories (professional or clerical work, other work). Increasing height was significantly associated with an increased risk of prostate cancer (P for trend = 0.03). The highest category (
168 cm) was marginally associated with an increased risk of prostate cancer (OR, 1.52; P = 0.05). In other sites, no significant association of height with cancer risk was observed. Although an increased OR of colon cancer was observed for the highest category, statistical test showed non-significance (P = 0.26).
|
The results on early-life factors were shown in Table 3. Urban-born was significantly associated with an increased risk of prostate cancer. Although an increased OR for urban-born was observed for colon and rectal cancer, the statistical test showed non-significance. The associations of stature at 12 years of age with cancer risk varied among prostate cancer and the four other sites. Although an increased OR of prostate cancer was observed for tall subjects at 12 years of age, a linear association of stature at 12 years with prostate cancer risk was unclear.
|
Tables 4 and 5 present the results on prostate cancer, according to clinical stage. The 282 prostate cancer cases were classified into localized (125), advanced (141) and stage unknown (16) cases. Although not shown in the tables, cases with advanced cancer were older than those with localized cancer (mean age; localized 69.5 years, advanced 73.3 years), and the frequency of cases who were admitted through screening was higher in localized cancer (localized, 22.4%; advanced, 7.8%). The ORs presented in Tables 4 and 5 were adjusted for covariates including age and referral base. Although the association with height was not observed for localized cancer, the risk of advanced cancer was marginally related to increasing height (P for trend = 0.07). For the highest category of height, a greater OR of advanced cancer was observed (OR = 1.70). Urban-born was significantly associated with an increased risk of advanced cancer. Stature at 12 years of age was positively associated with localized cancer risk (P for trend = 0.03).
|
|
|
DISCUSSION |
|---|
|
TOPAbstractINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The present study found a significant association between height and the risk of prostate cancer in Japanese men, whereas no association with height was observed for other major cancer sites. Some associations between early-life factors such as birthplace and prostate cancer risk were also found. Analyses on prostate cancer according to clinical stage revealed that height might be associated with the risk of advanced cancer.
Before interpreting these findings, methodological problems merit consideration. In hospital-based case–control studies, some methodological problems are likely to influence the results. First, we considered problems in comparability between cases and controls. Theoretically, eligibility criteria for cases define the source population in case–control studies, from which controls should be selected. Namely, if hospitalized cancer cases are a biased sample of the general population, controls should also be a biased sample in the same way (13,14). In the present study, we selected controls from patients admitted to the same hospital as cases. To improve the comparability between cases and controls, statistical analyses appropriately controlled background characteristics such as area of residence and referral patterns. Furthermore, it is unlikely that diagnoses in the controls are related to exposure variables, i.e. height and early-life factors. Secondly, we considered the possibility of information bias. Bias in measuring height might have distorted the results, and some self-reported information on exposure in the distant past including stature at 12 years might have suffered from misclassification. In the present study, information on height was mainly obtained from self-reported data at admission. Some incomplete or missing values were compensated by measured data obtained from the clinical database. However, there was a good correlation in height between these two data, and an additional analysis limited to self-reported height showed similar results (data not shown in tables). The effect of bias in measuring height would be small. Regarding measuring height, the possibility of another limitation was also considered. Height was reported or measured around admission time, so data on height might have been underestimated by aging, pre-existing diseases or cancer occurrence. However, year of birth and year of survey were adjusted for in the analysis. Additional adjustment for pre-existing diseases, such as hypertension, diabetes mellitus and heart disease which were included in the questionnaire, did not substantially alter the results. It is unknown whether the reduction level of height differs among men with cancers and other diseases. Thus, it is unlikely that this limitation distorted the results. As for the misclassification of exposure in the distant past, we considered as follows. Since the hypothesis that birthplace and stature at 12 years of age may be associated with prostate cancer risk is unknown to study subjects, any bias for self-report would be non-differential. Thus, ORs will be toward unity (13). It is unlikely that the quality of the self-reported information might have seriously distorted the results.
The association between height and prostate cancer risk has been investigated in numerous Western studies (7,15–23). Most, but not all, of these Western studies have showed that tallness might be associated with an increased risk of prostate cancer (7,16,18,19,22,23). Recent meta-analysis also showed an association of height with prostate cancer risk (24). The present study on Japanese subjects revealed a similar association of height with prostate cancer risk. A greater risk of prostate cancer for tall men was found (OR = 1.52, between the highest and lowest categories). In Japan, to our knowledge, only two studies have evaluated the relations between height and prostate cancer risk (12,25), and a positive association with height was never observed. One study that used a case–control study method showed that the mean height in prostate cancer cases was less than that in controls (25). The other study that used a cohort study method showed null association between height and prostate cancer risk (12).
Regarding other major cancers, it has been suggested that tallness might be associated with colon cancer risk (16,26,27). In the present study, a moderate high risk of colon cancer was observed in the highest category of height. However, compared with prostate cancer, the effect of height is likely to be weaker for colon cancer. Thus, the comparisons in the effects of height on cancer risk between prostate and other major sites indicate that the association with height is more specific to prostate cancer. Prostate cancer may have some different epidemiologic characteristics from other major cancers.
Analysis on prostate cancer according to clinical stage showed that height was more strongly associated with the risk of advanced prostate cancer. Although few studies have examined the effect of height according to clinical stage (8,18,21,23), most of these studies reported similar associations to our study. From several viewpoints, we considered a difference in associations of height between the risk of localized and advanced prostate cancer. First, localized cancers are the sum of clinically overt and latent tumors, which might affect the risk estimates for localized cancer (28). In particular, some screen-detected cancers are likely to be latent. It is possible that the difference in the frequency of screen-detected cancer and characteristics of screened subjects between studies may produce different risk estimates for localized cancer. To investigate whether the risk estimate of localized cancer might be modified by screening history, we attempted separate analysis by referral base (data not shown in tables). No significant association with height was found among screened subjects as well as among non-screened subjects, which was similar to the summary result shown in Table 4. Regardless of referral patterns, the association of height with localized cancer risk seems unity or weak. In contrast, it is likely that the most advanced cancers are clinically overt. In the present study, the frequency of cases from screening was much lower in advanced cancer compared with localized cancer. The positive association of height with the risk of advanced cancer may represent a typical characteristic of prostate cancer. A greater OR of advanced cancer for the highest category of height suggests that height may play an important role in developing advanced cancer.
Height is mainly determined by genetic traits and nutrition during prenatal and childhood period, under the complex control of various hormones. Some biologic studies suggest the possibility that taller men may have been exposed to higher levels of insulin-like growth factor-I (IGF-I) and testosterone during childhood and adolescence (29–31). Height may act as surrogates for these hormones. It is known that IGF-I may directly enhance prostate tumorigenesis by inhibiting apoptosis and by stimulating cell proliferation (31). Serum testosterone, controlling prostatic growth, may influence the effects of IGF-I. Epidemiologic studies have shown that a higher level of IGF-I might be associated with prostate cancer risk (32,33). For advanced cancer, a stronger association with the level of IGF-I has been observed (34). A higher risk of advanced cancer among tall men may be explained by hormonal mechanisms including the effect of IGF-I.
The above-mentioned findings on height suggest the possibility that early-life factors determining child growth may have some effects on future prostate cancer risk. The present study showed a significant association between urban-born and prostate cancer risk. Furthermore, the analysis by clinical stage revealed that the association was limited to advanced cancer. An elevated risk for urban-born was also observed for colon and rectal cancer; however, statistical testing showed non-significance. Although the results of hospital-based studies like ours must be carefully interpreted, these findings indicate that the distributions of birthplace may vary among prostate cancer cases and controls residing around the catchment area of the hospital. To our knowledge, two Western case–control studies have showed the association between history of urban living and prostate cancer risk (8,20). It is likely that environments in early life may play a role in the development of prostate cancer. Urban-born, for example, might have influenced dietary habit promoting child growth and endocrine events at puberty, which may modify prostate cancer risk. Although we could not obtain lifestyle data in the distant past, the reports of the nutritional survey in the 1950s showed that urban residents consumed much more nutrients such as protein and fat than rural residents (35). Special reports on child growth showed that infants in urban area tended to be taller than those in rural area (36). Furthermore, it is reported that infant mortality rate in urban area has been lower than that in rural area, suggesting that living conditions in urban area may be favorable for child growth (37). Besides, stature at 12 years of age tended to be positively related to the risk of prostate cancer. In the analysis according to clinical stage, a significant higher risk of localized cancer was found among tall men at 12 years of age. Although this finding is discrepant to that on height, men who attain a greater height at an earlier age are likely to have a higher risk of localized cancer. Attained height reflects hormonal environment during the growth period. The timing of hormonal effects may also play an important role in the development of prostate cancer (38). However, since studies regarding the effect of growth pattern have been few (39), we cannot fully explain the mechanisms for our findings. To clarify the effects of early-life environments and growth pattern, further studies are required.
The trend in prostate cancer incidence in Japan may be explained based on the present findings to some extent. Recently, height in Japanese adults and boys has been increasing (40,41). Mean height in middle- and older-aged Japanese men increased 
5 cm during 1975–2000 (41). These drastic changes in Japanese men might have contributed to the increased incidence of prostate cancer, together with urbanization.
In summary, the present hospital-based case–control study revealed a significant positive association between height and the risk of prostate cancer in Japanese men. Analyses according to clinical stage revealed that height might be more strongly associated with the risk of advanced prostate cancer. Regarding early-life factors, a significant association of urban-born with the risk of advanced prostate cancer was found. The association with tallness at 12 years of age was also observed for localized prostate cancer. For other major cancers, no significant association with height and early-life factors was observed. These findings indicate that height and early-life factors may have greater impacts on prostate cancer risk in Japanese men.
|
Acknowledgments |
|---|
The authors are grateful to all staff members of the Miyagi Cancer Center who generously cooperated in this study. Support for the study came in part from a Grant-in-Aid for the Second Term Comprehensive 10-Year Strategy for Cancer Control from the Ministry of Health, Labour and Welfare, and a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture, Japan.
Conflict of interest statement
None declared.
|
References |
|---|
|
TOPAbstractINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
1 Parkin DM, Whelan SL, Ferlay J, Raymond L, Young J. Cancer Incidence in Five Continents, vol. (1997) VII. Lyon: International Agency for Research on Cancer. 386–9. (IARC scientific publications, no. 143).
2 Zaridze DG, Boyle P, Smans M. International trends in prostatic cancer. Int J Cancer (1984) 33:223–30.[Web of Science][Medline]
3 Marugame T, Kamo K, Katanoda K, Ajiki W, Sobue T. Cancer incidence and incidence rates in Japan in 2000: estimates based on data from 11 population-based cancer registries. Jpn J Clin Oncol (2006) 36:668–75.
4 The Miyagi Prefectural Cancer Registry. Cancer incidence in Miyagi Prefecture 1998–2002. (2007) Sendai: The Miyagi Prefectural Cancer Registry. (in Japanese).
5 Adami HO, Hunter D, Trichopoulos D. Textbook of Cancer Epidemiology. (2002) Oxford: Oxford University Press. 400–28.
6 Bostwick DG, Burke HB, Djakiew D, et al. Human prostate cancer risk factors. Cancer (2004) 101(10 Suppl):2371–490.[CrossRef][Web of Science][Medline]
7 Andersson S-O, Wolk A, Bergström R, et al. Body size and prostate cancer: a 20-year follow-up study among 135006 Swedish construction workers. J Natl Cancer Inst (1997) 89:385–9.
8 Andersson S-O, Baron J, Wolk A, Lindgren C, Bergström R, Adami H-O. Early life risk factors for prostate cancer: a population-based case-control study in Sweden. Cancer Epidemiol Biomarkers Prev (1995) 4:187–92.[Abstract]
9 Minami Y, Tateno H. Associations between cigarette smoking and the risk of four leading cancers in Miyagi Prefecture, Japan: a multi-site case-control study. Cancer Sci (2003) 94:540–7.[CrossRef][Medline]
10 Odell WD, Puberty. Endocrinology. DeGroot LJ, ed. (1989) vol. 3, 2nd edn. Philadelphia: W. B. Saunders. 1860–72.
11 Breslow NE, Day NE. The analysis of case-control studies. (1980) vol. 1. Lyon: International Agency for Research on Cancer. 192–246. Statistical methods in cancer research (IARC scientific publications, no. 32.)
12 Kurahashi N, Iwasaki M, Sasazuki S, Otani T, Inoue M. Tsugane S for the JPHC Study Group. Associations of body mass index and height with risk of prostate cancer among middle-aged Japanese men. Br J Cancer (2006) 94:740–2.[Web of Science][Medline]
13 Rothman KJ, Greenland S. Modern Epidemiology. (1998) 2nd edn. Philadelphia: Lippincott-Raven.
14 Schlesselman JJ. Case-control Studies. Design, Conduct, Analysis. (1982) New York: Oxford University Press.
15 Albanes D, Jones DY, Schatzkin A, Micozzi MS, Taylor PR. Adult stature and risk of cancer. Cancer Res (1988) 48:1658–62.
16 La Vecchia C, Negri E, Parazzini F, et al. Height and cancer risk in a network of case-control studies from Northern Italy. Int J Cancer (1990) 45:275–9.[Web of Science][Medline]
17 Andersson S-O, Baron J, Bergström R, Lindgren C, Wolk A, Adami H-O. Lifestyle factors and prostate cancer risk: a case-control study in Sweden. Cancer Epidemiol Biomarkers Prev (1996) 5:509–13.[Abstract]
18 Giovannucci E, Rimm EB, Stampfer MJ, Colditz GA, Willett WC. Height, body weight, and risk of prostate cancer. Cancer Epidemiol Biomarkers Prev (1997) 6:557–63.[Abstract]
19 Hebert PR, Ajani U, Cook NR, Lee I-M, Chan KS, Hennekens CH. Adult height and incidence of cancer in male physicians (United States). Cancer Causes Control (1997) 8:591–7.[CrossRef][Web of Science][Medline]
20 Hsieh CC, Thanos A, Mitropoulos D, Deliveliotis C, Mantzoros CS, Trichopoulos D. Risk factors for prostate cancer: a case-control study in Greece. Int J Cancer (1999) 80:699–703.[CrossRef][Web of Science][Medline]
21 Norrish AE, McRae CU, Holdaway IM, Jackson RT. Height-related risk factors for prostate cancer. Br J Cancer (2000) 82:241–5.[CrossRef][Web of Science][Medline]
22 Habel LA, Van Den Eeden SK, Friedman GD. Body size, age at shaving initiation, and prostate cancer in a large, multiracial cohort. Prostate (2000) 43:136–43.[CrossRef][Web of Science][Medline]
23 Sequoia JSP, Wright ME, McCarron P, et al. A prospective investigation of height and prostate cancer risk. Cancer Epidemiol Biomarkers Prev (2006) 15:2174–8.
24 MacInnis RJ, English DR. Body size and composition and prostate cancer risk: systematic review and meta-regression analysis. Cancer Causes Control (2006) 17:989–1003.[CrossRef][Web of Science][Medline]
25 Furuya Y, Akimoto S, Akakura K, Ito H. Smoking and obesity in relation to the etiology and disease progression of prostate cancer in Japan. Int J Urol (1998) 5:134–7.[Medline]
26 Shimizu N, Nagata C, Shimizu H, et al. Height, weight, and alcohol consumption in relation to the risk of colorectal cancer in Japan: a prospective study. Br J Cancer (2003) 88:1038–43.[CrossRef][Web of Science][Medline]
27 Minami Y, Nishino Y, Tsubono Y, Tsuji I, Hisamichi S. Increase of colon and rectal cancer incidence rates in Japan: trends in incidence rates in Miyagi Prefecture, 1959–1997. J Epidemiol (2006) 16:240–8.[CrossRef][Web of Science][Medline]
28 Begg CB. Methodological issues in studies of the treatment, diagnosis, and etiology of prostate cancer. Semin Oncol (1994) 21:569–79.[Web of Science][Medline]
29 Keenan BS, Richards GE, Ponder SW, Dallas JS, Nagamani M, Smith ER. Androgen-stimulated pubertal growth: the effects of testosterone and dihydrotestosterone on growth hormone and insulin-like growth factor-I in the treatment of short stature and delayed puberty. J Clin Endocrinol Metab (1993) 76:996–1001.[Abstract]
30 Juul A, Bang P, Hertel NT, et al. Serum insulin-like growth factor-I in 1030 healthy children, adolescents, and adults: relation to age, sex, stage of puberty, testicular size, and body mass index. J Clin Endocrinol Metab (1994) 78:744–52.[Abstract]
31 Kaaks R, Lukanova A, Sommersberg B. Plasma androgens, IGF-I, body size, and prostate cancer risk: a synthetic review. Prostate Cancer Prostatic Dis (2003) 3:157–72.[Medline]
32 Chan JM, Stampfer MJ, Giovannucci E, et al. Plasma insulin-like growth factor-I and prostate cancer risk: a prospective study. Science (1998) 279:563–6.
33 Allen NE, Key TJ, Appleby PN, et al. Serum insulin-like growth factor (IGF)-I and IGF-binding protein-3 concentrations and prostate cancer risk: results from the European prospective investigation into cancer and nutrition. Cancer Epidemiol Biomarkers Prev (2007) 16:1121–7.
34 Chan JM, Stampfer MJ, Ma J, et al. Insulin-like growth factor-I (IGF-I) and IGF binding protein-3 as predictors of advanced-stage prostate cancer. J Natl Cancer Inst (2002) 94:1099–109.
35 Omori K. Investigation on nutritional level in Japan. (1954) Daiichi-Shuppan. (in Japanese).
36 Ministry of Health and Welfare. Statistical report on infant and child growth 1970. (1971) Tokyo: Printing Bureau, Ministry of Finance. (in Japanese).
37 Araki S, Murata K. Social life factors affecting stillbirth, and infant and adult-disease mortality. Tohoku J Exp Med (1986) 149:213–9.[Web of Science][Medline]
38 Hsing AW. Hormones and prostate cancer: what's next? Epidemiol Rev (2001) 23:42–66.
39 Giles GG, Severi G, English DR, et al. Early growth, adult body size and prostate cancer risk. Int J Cancer (2003) 103:241–5.[CrossRef][Web of Science][Medline]
40 Ministry of Education, Culture, Sports, Science and Technology. School Health Statistics 2000. (2003) Tokyo: Printing Bureau, Ministry of Finance. (in Japanese).
41 Bureau of Public Health, Ministry of Health and Welfare. The National Nutrition Survey in Japan 2000. (2002) Tokyo: Daiichi-Shuppan. (in Japanese).
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. Sung, Y.-M. Song, D. A. Lawlor, G. D. Smith, and S. Ebrahim Height and Site-specific Cancer Risk: A Cohort Study of a Korean Adult Population Am. J. Epidemiol., July 1, 2009; 170(1): 53 - 64. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||