Background

To implement personalized medicine, we established a large-scale patient cohort, BioBank Japan, in 2003. BioBank Japan contains DNA, serum, and clinical information derived from approximately 200,000 patients with 47 diseases. Serum and clinical information were collected annually until 2012.

Methods

We analyzed clinical information of participants at enrollment, including age, sex, body mass index, hypertension, and smoking and drinking status, across 47 diseases, and compared the results with the Japanese database on Patient Survey and National Health and Nutrition Survey. We conducted multivariate logistic regression analysis, adjusting for sex and age, to assess the association between family history and disease development.

Results

Distribution of age at enrollment reflected the typical age of disease onset. Analysis of the clinical information revealed strong associations between smoking and chronic obstructive pulmonary disease, drinking and esophageal cancer, high body mass index and metabolic disease, and hypertension and cardiovascular disease. Logistic regression analysis showed that individuals with a family history of keloid exhibited a higher odds ratio than those without a family history, highlighting the strong impact of host genetic factor(s) on disease onset.

Conclusions

Cross-sectional analysis of the clinical information of participants at enrollment revealed characteristics of the present cohort. Analysis of family history revealed the impact of host genetic factors on each disease. BioBank Japan, by publicly distributing DNA, serum, and clinical information, could be a fundamental infrastructure for the implementation of personalized medicine.

Keywords: BioBank Japan Project, Biobank, Common disease, Clinical information, Family history

Study design

In the present cohort, we focused on 47 common diseases (Table 1). Patients diagnosed with any one of the 47 diseases were recruited from 66 hospitals affiliated with 12 medical institutes between fiscal year of 2003 and 2007. The detailed protocol of the recruitment process has been described elsewhere.² Written informed consent was obtained from all participants. The study protocol was reviewed and approved by the Ethics Committees of all participating institutions, including the Institute of Medical Science, the University of Tokyo, and the Center for Integrative Medical Sciences, RIKEN.

We included patients who had been diagnosed with the diseases by physicians at the cooperating hospitals (eTable 1). As this project registered not only patients with newly developed diseases but also patients who were diagnosed and treated before starting the project, some participants were enrolled several years after disease onset or diagnosis.² We excluded patients who had received a bone marrow transplant and those who were not of East Asian descent.

Clinical information

Clinical information including common clinical variables, disease-specific variables, prescriptions, and drug side-effect information, was collected from each participant. The detailed methods of the collection of clinical information has been described elsewhere.² The clinical database was updated every year until 2012. After a thorough review and data-cleansing of clinical variables,² clinical information of 199,982 participants with 47 diseases at enrollment was established on March 31 2015 and used in the current study.

Japanese database

The Ministry of Health, Labour and Welfare in Japan conducts a Patient Survey every three years and a National Health and Nutrition Survey every year. We obtained the results of the Patient Survey of 2005⁵ and those of the National Health and Nutrition Survey of 2006.⁶ Table 65 in the Patient Survey was used to estimate Japanese patient numbers, stratified by sex and age for each disease. Distributions of BMI categories, hypertension prevalence, smoking history, and alcohol intake history in the general Japanese population were calculated from Tables 23, 49-2, 97, and 91 of the National Health and Nutrition Survey, respectively.

Analysis of clinical information

The distributions of BMI, hypertension prevalence, smoking history, and alcohol intake history in the BBJ cohort were adjusted for sex and age group for each table in the national public survey when we compared the distributions among the 47 diseases and Japanese database. BMI category and hypertension were defined according to World Health Organization (WHO) criteria as follows: BMI < 18.5 was defined as underweight, 18.5 ≤ BMI < 25 as normal, 25 ≤ BMI < 30 as overweight, and 30 ≤ BMI as obese⁷; hypertension was defined as systolic blood pressure ≥140 mmHg and diastolic blood pressure ≥90 mmHg or when participants were prescribed antihypertensive drugs. Multivariate logistic regression analyses were performed to assess the association between each target disease and positive family history associated with the target disease, adjusted for sex and age. SAS 9.4 software was used for the data analysis. Ap-value of <0.05 was considered statistically significant.

Basic characteristics (Age and sex)

We characterized the BioBank Japan cohort at enrollment by analyzing common clinical variables of age and sex across the target diseases. Mean age at enrollment, across the entire cohort or for each disease, was comparable between both sexes, but varied among the diseases (Table 1). The highest mean age was observed in men with prostate cancer and in women with osteoporosis (72.60 and 73.77 years, respectively), while the youngest mean age was observed in men and women with febrile seizures (4.16 and 4.35 years, respectively), reflecting the typical age of onset of each disease. A greater number of men were registered in the BBJ cohort compared to women (53.05% vs. 46.95%), while sex ratios varied according to the diseases (Table 1).

To highlight sex and age characteristics of the BBJ cohort, we further compared the sex and age distribution for each disease with the Patient Survey. We included participants with 42 out of the 47 diseases for the comparison, as we obtained the relevant clinical data from the Patient Survey (eTable 2). Almost all diseases displayed equivalent age distributions, while lower proportions of participants <20 years of age were observed in three diseases (bronchial asthma, atopic dermatitis and hay fever), which are likely to occur in younger populations (eFigs. 1.1–1.5 and eTable 3). The proportion of male participants with dyslipidemia was considerably higher in the BBJ cohort (50.76%) than in the Patient Survey (33.55%), although both age distributions appeared equivalent. The low proportion of female patients with heart failure aged ≥80 years resulted in a lower proportion of female participants in the BBJ cohort. We also observed a low proportion of elderly female participants with cerebral infarction, chronic obstructive pulmonary disease (COPD), peripheral arterial diseases (PAD), unstable angina, stable angina, and myocardial infarction in the BBJ cohort. Varied distributions between the BBJ cohort and Patient Survey were observed in pulmonary tuberculosis and nephrotic syndrome.

Basic characteristics (Lifestyle and physical status)

We also evaluated life style including smoking and alcohol intake history, and physical status including BMI and blood pressure, at enrollment in the BBJ cohort. We included participants ≥20 years of age in this analysis because the frequency of smoking, alcohol intake, and hypertension among individuals under 20 years of age is quite low, and the criteria for underweight or obesity according to BMI in children and teenagers are different from those applied to adults.⁸ Furthermore, we compared the BBJ cohort and the National Health and Nutrition Survey 2006 for physical and life style, after adjusting for sex and age, because sex- and age-distribution varied among diseases.

Smoking history at enrollment (including subjects both with and without information on current smoking status) was positive in 74.98% of male subjects and 21.24% of female subjects in the BBJ cohort, while current smokers accounted for 27.78% of male subjects and 10.45% of female subjects (Table 2). The highest frequency of positive smoking history in both sexes was observed in COPD, followed by PAD in male subjects, and esophageal cancer in female subjects (Table 2). The highest proportion of ex-smokers for both sexes was observed in participants with lung cancer, esophageal cancer and COPD (71.45%, 64.88% and 64.68% in male subjects, and 21.75%, 30.86% and 39.44% in female subjects, respectively), while the highest proportion of current smokers for both sexes was observed in participants with Graves' disease (49.84% in male subjects and 24.92% in female subjects) (Table 2). We then compared age-adjusted smoking history among the 47 diseases. The frequency of smokers was highest among participants with COPD, esophageal cancer, interstitial lung diseases/pulmonary fibrosis, pancreatic cancer, and cardiovascular diseases, in which smoking was shown to be a critical risk factor (Fig. 1 andeTable 4).

A positive alcohol history at enrollment (including those with and without current drinking status) was found in 69.68% of male subjects and 28.20% of female subjects (Table 3). The proportion of current drinkers in the whole cohort was much higher than that of ex-drinkers in both sexes: 52.24% and 13.35% of male subjects and 21.70% and 3.99% of female subjects were current and ex-drinkers, respectively. Among the 47 diseases, the proportion of ex-drinkers was relatively high among participants with liver cirrhosis (34.05% in male subjects and 10.80% in female subjects), liver cancer (34.83% and 10.94%), pulmonary tuberculosis (33.17% and 11.20%), esophageal cancer (25.59% and 16%), and pancreatic cancer (29.72% and 10.14%) (Table 3). Age-adjusted alcohol intake history showed that the frequency of drinkers in esophageal cancer was remarkably higher than that in other diseases for male and female subjects (Fig. 2 andeTable 5). To highlight the smoking and drinking status in the BBJ cohort, the frequency of smokers or drinkers, stratified by sex and age group, was compared between the BBJ and the National Health and Nutrition Survey. The BBJ cohort had a higher frequency of smokers among female subjects across all age groups and among elderly male subjects, particularly among those >60 years of age; the frequency of drinkers was almost equivalent between the BBJ and the National Health and Nutrition Survey for both sexes and across all age groups (eFig. 2A and B and eTables 6 and 7).

Mean BMI at enrollment in the BBJ cohort was 23.51 in male subjects and 22.94 in female subjects. Analysis of BMI in each disease revealed that underweight participants (BMI<18.5) had an increased association of various cancers, while overweight or obese participants (BMI ≥ 25) had an increased association of metabolic and cardiovascular diseases (Table 4,Fig. 3 andeTable 8). When comparing the National Health and Nutrition Survey and the BBJ, there was a greater proportion of participants with overweight or obesity in the BBJ, among male and female subjects and across all age-groups; conversely, similar distribution patterns were found when comparing the BBJ cohort and the Survey, by sex and age-group (eFig. 2C and eTables 6 and 7). In contrast, in the BBJ cohort, there were fewer underweight participants in their twenties (for both sexes) but more underweight participants >60 years (among male subjects) and >50 years (among female subjects) (eFig. 2D and eTables 6 and 7).

Nearly half of the participants of the BBJ cohort had hypertension (51.52% of male subjects and 41.11% of female subjects,Table 4) at enrollment. The frequency of hypertension in cardiovascular diseases, particularly in coronary diseases, was higher than that in other diseases, while the frequency of hypertension among cancer participants tended to be low (Table 4,Fig. 4 andeTable 9). The frequency of hypertension increased with age, similarly to the increase observed in the Survey. However, the frequency of hypertension among subjects <50 years of age was higher and subjects >60 years of age was lower in the BBJ cohort than in the Survey (eFig. 2E and eTables 6 and 7).

Family history

Finally, we performed multivariate logistic-regression analysis using age and sex status as covariates to assess the association between positive family history and disease risk. We were able to obtain the questionnaire-based information regarding family history of 45 diseases out of the 47 diseases (eTable 10). For all the diseases, except for PAD, there was a significant association with a positive family history, with an odds ratio of >1.7 (Fig. 5 andeTable 11). Notably, the odds ratios for keloid, chronic hepatitis B, and Grave's disease were relatively high (149.417, 53.474, and 23.751, respectively) indicating the strong impact of genetic and familial factors on disease onset.

• 1.Nakamura Y. The BioBank Japan project. Clin Adv Hematol Oncol. 2007;5:696–697. [PubMed] [Google Scholar]
• 2.Nagai A., Hirata M., Kamatani Y. Overview of the BioBank Japan project: study design and profile. J Epidemiol. 2017;27:S2–S8. doi: 10.1016/j.je.2016.12.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 3.Asslaber M., Zatloukal K. Biobanks: transnational, European and global networks. Brief Funct Genomic Proteomic. 2007;6:193–201. doi: 10.1093/bfgp/elm023. [DOI] [PubMed] [Google Scholar]
• 4.BioBank Japan, Publications from BioBank Japan.https://biobankjp.org/work/public.html; Updated 30.06.16. Accessed 25 July 2016.
• 5.Ministry of Health, Labour and Welfare, Japan . 2005. Patient Survey.http://www.e-stat.go.jp/SG1/estat/List.do?lid=000001047095 (in Japanese) Accessed 6 June 2016. [Google Scholar]
• 6.Ministry of Health, Labour and Welfare, Japan . 2006. National Health and Nutrition Survey.http://www.mhlw.go.jp/bunya/kenkou/eiyou08/01.html (in Japanese) Accessed 6 June 2016. [Google Scholar]
• 7.World Health Organization . 2016. BMI Classification.http://apps.who.int/bmi/index.jsp?introPage=intro_3.html Accessed 6 June 2016. [Google Scholar]
• 8.Kuczmarski R.J., Ogden C.L., Guo S.S. 2000 CDC growth charts for the United States: methods and development. Vital Health Stat. 2002;11:1–190. [PubMed] [Google Scholar]
• 9.Centers for Disease Control and Prevention . 2016. Health Effects of Cigarette Smoking.http://www.cdc.gov/tobacco/data_statistics/fact_sheets/health_effects/effects_cig_smoking/ Accessed 16 June 2016. [Google Scholar]
• 10.Centers for Disease Control and Prevention . 2016. Fact Sheets – Alcohol Use and Your Health.http://www.cdc.gov/alcohol/fact-sheets/alcohol-use.htm Accessed 16 June 2016. [Google Scholar]
• 11.Centers for Disease Control and Prevention . 2016. Adult Obesity Causes & Consequences.http://www.cdc.gov/obesity/adult/causes.html Accessed 16 June 2016. [Google Scholar]
• 12.Reid G.T., Walter F.M., Brisbane J.M., Emery J.D. Family history questionnaires designed for clinical use: a systematic review. Public Health Genomics. 2009;12:73–83. doi: 10.1159/000160667. [DOI] [PubMed] [Google Scholar]
• 13.Brandi M.L., Gennari L., Cerinic M.M. Genetic markers of osteoarticular disorders: facts and hopes. Arthritis Res. 2001;3:270–280. doi: 10.1186/ar316. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 14.Cole Johnson C., Ownby D.R., Havstad S.L., Peterson E.L. Family history, dust mite exposure in early childhood, and risk for pediatric atopy and asthma. J Allergy Clin Immunol. 2004;114:105–110. doi: 10.1016/j.jaci.2004.04.007. [DOI] [PubMed] [Google Scholar]
• 15.Collaborative Group on Hormonal Factors in Breast C Familial breast cancer: collaborative reanalysis of individual data from 52 epidemiological studies including 58,209 females with breast cancer and 101,986 females without the disease. Lancet. 2001;358:1389–1399. doi: 10.1016/S0140-6736(01)06524-2. [DOI] [PubMed] [Google Scholar]
• 16.Harrison T.A., Hindorff L.A., Kim H. Family history of diabetes as a potential public health tool. Am J Prev Med. 2003;24:152–159. doi: 10.1016/s0749-3797(02)00588-3. [DOI] [PubMed] [Google Scholar]
• 17.Hawe E., Talmud P.J., Miller G.J., Humphries S.E., Second Northwick Park Heart Study Family history is a coronary heart disease risk factor in the Second Northwick Park Heart Study. Ann Hum Genet. 2003;67:97–106. doi: 10.1046/j.1469-1809.2003.00017.x. [DOI] [PubMed] [Google Scholar]
• 18.Johns L.E., Houlston R.S. A systematic review and meta-analysis of familial colorectal cancer risk. Am J Gastroenterol. 2001;96:2992–3003. doi: 10.1111/j.1572-0241.2001.04677.x. [DOI] [PubMed] [Google Scholar]
• 19.Pharoah P.D., Ponder B.A. The genetics of ovarian cancer. Best Pract Res Clin Obstet Gynaecol. 2002;16:449–468. doi: 10.1053/beog.2002.0296. [DOI] [PubMed] [Google Scholar]
• 20.World Health Organization, Geneva . 2015. Guidelines for the Prevention, Care, and Treatment of Persons with Chronic Hepatitis B Infection.http://www.ncbi.nlm.nih.gov/books/NBK305553/ Accessed 16 June 2016. [PubMed] [Google Scholar]
• 21.Ban Y., Taniyama M., Ban Y. Vitamin D receptor gene polymorphism is associated with Graves' disease in the Japanese population. J Clin Endocrinol Metab. 2000;85:4639–4643. doi: 10.1210/jcem.85.12.7038. [DOI] [PubMed] [Google Scholar]
• 22.Ban Y., Tozaki T., Taniyama M. The replication of the association of the rs9355610 within 6p27 with Graves' disease. Autoimmunity. 2013;46:395–398. doi: 10.3109/08916934.2013.780600. [DOI] [PubMed] [Google Scholar]
• 23.Ban Y., Tozaki T., Taniyama M., Tomita M., Ban Y. Association of a C/T single-nucleotide polymorphism in the 5′ untranslated region of the CD40 gene with Graves' disease in Japanese. Thyroid. 2006;16 doi: 10.1089/thy.2006.16.443. 443–436. [DOI] [PubMed] [Google Scholar]
• 24.Furugaki K., Shirasawa S., Ishikawa N. Association of the T-cell regulatory gene CTLA4 with Graves' disease and autoimmune thyroid disease in the Japanese. J Hum Genet. 2004;49:166–168. doi: 10.1007/s10038-003-0120-5. [DOI] [PubMed] [Google Scholar]
• 25.Hiratani H., Bowden D.W., Ikegami S. Multiple SNPs in intron 7 of thyrotropin receptor are associated with Graves' disease. J Clin Endocrinol Metab. 2005;90:2898–2903. doi: 10.1210/jc.2004-2148. [DOI] [PubMed] [Google Scholar]
• 26.Kamatani Y., Wattanapokayakit S., Ochi H. A genome-wide association study identifies variants in the HLA-DP locus associated with chronic hepatitis B in Asians. Nat Genet. 2009;41:591–595. doi: 10.1038/ng.348. [DOI] [PubMed] [Google Scholar]
• 27.Komatsu H., Murakami J., Inui A., Tsunoda T., Sogo T., Fujisawa T. Association between single-nucleotide polymorphisms and early spontaneous hepatitis B virus e antigen seroconversion in children. BMC Res Notes. 2014;7:789. doi: 10.1186/1756-0500-7-789. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 28.Kumar V., Yi Lo P.H., Sawai H. Soluble MICA and a MICA variation as possible prognostic biomarkers for HBV-induced hepatocellular carcinoma. PLoS One. 2012;7:e44743. doi: 10.1371/journal.pone.0044743. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 29.Mukai T., Hiromatsu Y., Fukutani T. A C/T polymorphism in the 5′ untranslated region of the CD40 gene is associated with later onset of Graves' disease in Japanese. Endocr J. 2005;52:471–477. doi: 10.1507/endocrj.52.471. [DOI] [PubMed] [Google Scholar]
• 30.Nakashima M., Chung S., Takahashi A. A genome-wide association study identifies four susceptibility loci for keloid in the Japanese population. Nat Genet. 2010;42:768–771. doi: 10.1038/ng.645. [DOI] [PubMed] [Google Scholar]
• 31.Nishida N., Ohashi J., Khor S.S. Understanding of HLA-conferred susceptibility to chronic hepatitis B infection requires HLA genotyping-based association analysis. Sci Rep. 2016;6:24767. doi: 10.1038/srep24767. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 32.Okada Y., Momozawa Y., Ashikawa K. Construction of a population-specific HLA imputation reference panel and its application to Graves' disease risk in Japanese. Nat Genet. 2015;47:798–802. doi: 10.1038/ng.3310. [DOI] [PubMed] [Google Scholar]
• 33.Tripepi G., Jager K.J., Dekker F.W., Zoccali C. Selection bias and information bias in clinical research. Nephron Clin Pract. 2011;115:c94–c99. doi: 10.1159/000312871. [DOI] [PubMed] [Google Scholar]

Supplementary Materials