1
Biobank Japan project toward the personalized
medicine
Yusuke Nakamura,M.D.,Ph.D. Director, Human Genome
Center, The University of Tokyo Director, SNP
Research Center, RIKEN
2
The Aims of Biobank Japan Project are to
construct an infrastructure for medical science.

1. Discovery of genes susceptible to various
   diseases, or those related to efficacy or adverse
   reactions of drugs

2. Providing the useful information of molecular
targets for evidence-based drug development

1. Providing the important medical information that
   can be applied for establishment of Personalized
   Medicine

1. To perform genetic and environmental epidemiology
   (interaction) for prevention of diseases or
   progression of diseases

3
What we have been doing are

1. Collection of DNA, serum, and clinical
   information from 300,000 patients in 2003-2007
   (one time for DNA, each year for serum and
   clinical informration)

2. Construction of Biobank Japan (DNAs are
stored at 4oC and sera in liquid N2) (Providing
samples to the researchers in academic and
private research institutes in Japan)
3. Systematic genomics and proteomics analysis
4. Identification of genes of medical importance
4
Medical Effect of Biobank Japan Project

1. Use drugs more effectively
2. Reduce incidences of drug adverse reactions
3. Reduce unnecessary medical cost
4. Develop new diagnosis and treatment
5. Prevent diseases or progression of diseases

Population-based biobank Patient-oriented biobank
Much larger information (clinical information,
life-style information, SNP information, protein
information) should be collected.
5
Sample-flow in the Biobank Japan project
Hospitals
2
3
Patient
Nurse
Individual Code number
Blood
Informed consent
1
MC
Clinical information
4
5
Biobank Japan In IMSUT
DNA Serum
7
6
SNP Research Center RIEKN
University
6
Video program to explain biobank project
7
Physical localization of Collaborative hospitals
66 Hospitals (a total of gt 25,000 beds) are
participating in the Biobank Japan project
8
Security system for DNA bank
Fingerprint reader
Only five people including me are allowed to
enter this facility
9
Privacy Protection for individual clinical
information in PC
Registration of medical coordinator Memorize
MCs fingerprint information in an IC card.
Fingerprint card reader
Log-in for PC is allowed only when the
fingerprint pattern is matched
If the pattern is not matched for THREE
consecutive times
All the data in the hard disk will be erased
promptly and automatically
10
Barcode scanner
2D-barcode
11
(No Transcript)
12
Fully-automated system (DNA bank) 1,000,000 tubes
can be stored in this facility.
13
(No Transcript)
14
Fully automated system (DNA bank)
15
Fully automated system (DNA bank) Barcode Reading
Machine
16
(No Transcript)
17
Serum Bank3,000,000 tubes can be stored in this
facility
50 liquid nitrogen tanks (keeping samples at
-150oC)
18
(No Transcript)
19
Biobank Japan Sample Summary at April 30, 2007
Breast 6,085 Colorectal 5,997 Gastric 5,903 My
oma uteri 5,185 Prostate 4,936 Lung 3,959 Live
r 2,016 Leukemia 1,342 Cervical 1,212 Esophag
eal 1,187 Endometrial 1,008 Ovarian
869 Pancreatic 453 CCC 434 Total
40,386
IC asked 208,013 patients IC obtained 178.724
patients (85.9)
Total Cases 254,516 cases Withdrawn 155
individuals
20
MaleFemale ratio of hepatitis, liver cirrhosis,
and HCC
Cirrhosis
Hepatitis C
100 62
100 86
Sex difference? Alcohol? Other environmental
factors Genetic?
Male Female
Male Female
100 31
HCC
Male Female
21
SNP Research Center in RIKEN Yokohama Institute
22
The SNP Research Center in RIKEN contributed the
largest SNP data among the SNP typing centers in
this phase I paper.
October 27, 2005
23
Whole-genome association study

• 2001-2004 (Gene-based SNPs)
• Multiple-PCR Invader Assay
• 500 M SNP data-points/year
• 2005-2006 (250K HapMap SNPs)
• DNA oligoarray Invader Assay
• 700 M SNPdata-points/year
• 2007- (550K HapMap SNPs)
• Beads array
• 4.5 B SNP data points/year

24
Identification of genes susceptible to diseases
(2005)
200 or more each of randomly selected patients
with each of 50 diseases (600 individuals for
some diseases)
Genotyping for 2000 each of patients and controls
at the candidate SNP loci
5 Bil SNP data points
5000-10.000 SNPs X 2000 indiv. X 50
diseases 1 Bil data points
Genotyping for 10,000 individuals at
250,000/550.000 SNP loci
A gene showing the p value of 10-5 or smaller
Association analysis
200-1000 patients with disease A
Information of 10,000 individuals
Functional analysis of candidate genes and SNPs
Selection of loci that show the p value of 0.01
or small
Second phase
First phase
First phase
25
Genotyping by March 31, 2007
First First Second Second
Disease N SNP N SNP
LC 200 234,209 1,338 3,854
DM 200 236,594 1,503 9,121
ATH 200 240,763 1,500 3,881
HA 200 238,946 1,504 9,532
MMK 200 242,784 1,504 3,849
THY 200 246,131 1,018 3,882
CI 200 245,032 1,491 3,871
CA 200 244,661 1,367 3,883
OS 200 245,898 1,499 3,854
CF 200 239,878 1,484 3,384
UAN 200 238,737 1,470 3,887
NA 200 240,730 1,499 3,869
ASO 200 240,417 793 3,851
COPD 200 240,408 1,087 3,852
AD 200 245,987 1,193 3,851
MK 200 246,679 1,500 3,853
LC 200 247,614 850 3,847
ARRY 200 207,183 1,504 11,327
First First Second Second
Disease N SNP N SNP
NEP 200 220,639 329 11,179
CAT 200 218,877 1,504 11,340
GIN 200 218,967 1,504 10,980
SR 200 221,535
ALS 200 219,699 470 10,932
USTO 200 218,046 1,504 11,213
GRAU 200 231,656 1,504 11,063
COK 200 229,218 1,504 11,266
PRK 200 228,756 1,504 11,324
ILD 200 222,762
DM 200 227,841
ANG 200 228,886
RA 200 237,061
HA 200 236,471
HVC 200 238,427
HLY 200 233,567
LMY 200 234,655
EPI 200 239,032
TOTAL 7,400 1,727,925,000 34,491 235,874,147
3 Bil. Data-points (2007.03.31) To 6 Bil
data-points (2008.03.31)
EM 200 210,879 564 10,521
26
SNP-based association studies in RIKEN and
University of Tokyo Myocardial Infarction Lympho
toxin a (LTA) Nature Genetics, 2002 Galactin 2
(LGALS2) Nature, 2004 PSMA6 Nature Genetics,
2006 Rheumatoid Arthritis Peptidylarginine
deiminase type4 (PADI4) Nature Genetics,
2003 Organic cation transporter (SLC22A4)
Nature Genetics, 2003 Fc receptor homolog 3
(FcRH3) Nature Genetics, 2005 Diabetic
nephropathy Solute carrier family 12, member3
(SLC12A3) Diabetes, 2003 Engulfment and cell
motility 1(ELMO1) Diabetes, 2005 IgA
nephropathy SELL,SELE Am J Hum Genet,
2002 Osteoarthritis AsporinNature Genetics,
2005 Calmodulin 1 Human Molecular Genetics,
2005 GDF5 Nature Genetics, 2007 Disc
herniation CILP Nature Genetics, 2005 Brain
Infarction PRKCH1 Nature Genetics, 2007
The advantage of the whole-genome association
study is that we can identify unexpected genes
for which we are unable to expect its involvement
in diseases through the known information of
genes or gene products.
27
Reumatoid Arthritis ACR Teaching Slide Collection
Drs. Ryo Yamada and Kazuhiko Yamamoto
28
Major findings in RA group in RIKEN SRC
29

• Peptidylargininedeiminase (PADI)
• PADI 1-4 located at 1p36

PADI4
Arginine
Citrullin
10kb
Losing a positive charge of an amino acid
30
What are the Roles of Citrullinated Proteins?

• Citrullination would change the tertiary
  structure of the protein
• Conversion of self-protein to non-self-protein
• change the antigenicity

31
Anti-Citrullinated Peptide Antibodies
PADI4 Enzyme
Anti-citrullinated peptide antibody was found
specifically in more than 90 of RA
patients. Indicating the importance of
citrullination in the etiology of RA
?
PADI4 Gene
PADI4 Substrate Citrullinated Proteins
32
Annual numbers of the subjects who were newly
introduced to the chronic renal replacement
therapy due to diabetic nephropathy or chronic
glomerulonephritis
()
Chronic glomerulonephritis
13,920 (41.3)
9,466 (28.1)
Diabetic Nephropathy
84
86
88
90
92
94
96
98
00
02
04
33
Effects of glucose on the expression of the ELMO1
(genes involved in engulfment and cell motility
1) gene in COS cells


plt0.0001
2



1
0
Glucose(mM)
5.5
25
5.5
5.5
25
5.5
5.5
25
5.5
Mannitol(mM)
0
0
19.5
0
0
19.5
0
0
19.5
day
1
1
1
2
2
2
3
3
3
(Shimazaki, Maeda et al. Diabetes, 2005)
34
The effects of ELMO1 overexpression and
knock-down on ECM gene expressions
Fibronectin
Collagen type 1 (a1)
These ECMs may play a significant roles in
Sclerosis of glomerulus.
7
12
6
10
5
8
4
Fold increase
Fold increase
6
3
4
2
2
1
0
0
Control
ELMO1
Control
Control
Control
Control
ELMO1
ELMO1
Control
ELMO1
ELMO1
ELMO1
siRNA
siRNA
siRNA
siRNA
siRNA
siRNA
ELMO1- Line1
ELMO1- Line1
ELMO1- Line2
LacZ
ELMO1- Line2
LacZ
mean SD
? Plt 0.05 ?? Plt0.01 .vs. cells treated with
control siRNA
35
As verification of association study, we have
applied a cohort study in Hisayama town in
Fukuoka, Japan
Genetic Risk Factor of Cerebral Infarction
Drs. Kubo and Kiyohara in Kyusyu Unoversity
36
SNP association dataage gt40 years?sex and age
matching
(SNP-PRHCH1)
Allele frequency () Allele frequency () P-value (Odds Ratio) P-value (Odds Ratio) P-value (Odds Ratio) P-value (Odds Ratio) P-value (Odds Ratio)
N Case Control 1/2 11/1222 1112/22
All CI 1112 23.1 18.9 5.3x10-4 (1.29) 0.066 (1.46) 8.5x10-4 (1.34)
LA 491 24.0 15.7 4.7x10-6 (1.69) 0.0056 (2.63) 2.1x10-5 (1.77)
AT 369 22.6 19.4 0.13 (1.21) 0.72 (0.88) 0.054 (1.34)
LAAT 860 23.4 17.3 1.0x10-5 (1.46) 0.069 (1.56) 7.9x10-6 (1.57)
LAlacuna?ATatheromatous
37
Expression of PRHCH1and atherosclerotic changes
mm2
plt0.01 vs. DIT
6.0 4.0 2.0 0
Expression of PRKCH1 protein

Area of X expression


DIT I II III IV V VI
AHA classification
38
Hisayama
Kyusyu Univ.
Fukuoka
1960 2000
Hisayama 6,500 7,600
Fukuoka 650,000 1,340,000
39
V374I polymorphism and incidence of CIHisayama
cohort study 14-year follow-up (age-sex adjusted)
0.10
0.08
0.06
p0.030
Accumulated incidence of CI
0.04
0.02
0
0
5
10
15
Years
40
Pharmacogenomics study for biobank samples
Drs. Mitsuaki Kubo and Taisei Mushiroda
41
(No Transcript)
42
(No Transcript)
43
We examined 1,500 SNPs in 220 drug metabolizing
enzymes, transporters, receptors
44
Myelosuppression (plt0.0001)
45
Severe Skin Toxicity
SJS 9 cases TEN 5 cases Severe skin
rash 72 cases Total
86 cases
46
Rhabdomyolysis caused by statin
47
Identification of genes associated with warfarin
sensitivity
Drs. Taisei MUshiroda, Yozo Ohnishi, Takao Suzuki
48
Warfarin (WF)
? The most commonly-used oral anticoagulant for
treatment of thromboembolism in the world ?
Interferes with regeneration of vitamin K in the
vitamin K cycle inhibits activation of vitamin
K-dependent clotting factor II (prothrombin),
VII, IX and X ? Difficult to find an appropriate
dose to each patient due to large interindividual
variation in dose requirement Insufficient dose
? failure of preventing thrombosis Over-dose ?
increase of unexpected bleeding risk
49
Distribution of daily maintenance dose among
Japanese patients
150
Median 2.5 mg Minimum 0.5 mg Maximum 10.5 mg
125
100
75
No. of subjects
50
25
0
0.0
2.0
4.0
6.0
8.0
10.0
12.0
Dose (mg/day)
50
Number of Annual Estimate of Drugs Most Commonly
Implicated in Adverse Events Treated in Emergency
DepartmentsUnited States
Warfarin During 30-60 days after the beginning
of the treatment the high risk of bleeding or
infarction
JAMA 2961858-1865?2006
51
Patient characteristics (n 828)
Age (years, median) Gender Male Female Daily
WF dose (mg, median) VKORC1 genotype (intron
1-136TgtC) TT CT CC CYP2C9 genotype 1/1 1/
3 3/3
68 (range 19-92) 67 33 2.5 mg (range
0.5-10.5) 690 (83) 132 (16) 6 (lt1) 790
(95) 37 (4) 1 (lt1)
52
Daily WF dose for Japanese patients classified
by WFRI on the basis of a combination of VKORC1
and CYP2C9
VKORC1 / CYP2C9
P 4.4 ? 10-13
Dose (mg/day)
Dose (mg/day)
3.5
2.5
2.0
VKORC1 Intron 1-136TgtC
TT
TT
TT
CT
CT
CC
0 (n35)
1 (n658)
2 (n135)
CYP2C9
33
31
11
31
11
11
WF-responsive index
(n1)
(n34)
(n655)
(n3)
(n129)
(n6)
53
Medication in the near future
This kind of machine is essential for point-of
-care medical system
54

• Fully automated SNP genotyping equipment
• (in collaboration with Shimadu and Toppan)

Reaction card
Start reaction
55
Fully automated SNP genotyping equipment (in
collaboration with Shimadu and Toppan)
A drop of blood
30-min reaction
SNP
The accuracy is very high as 100 so far we
tested.
56
New cases who have warfarin treatment in
USA 2,000,000 cases
Avoidable ADRs(bleeding) 85,000 cases 1.
brain bleeding 2. intestinal
bleeding Avoidance of brain infarction due to
inappropriate treatment 17,000 cases
Reduction of medical cost per year 1.1 billion
USD (100M---2000M USD)
57
There are still many high hurdles that need to be
cleared, and we often suffer some disappointments
and setbacks along the way. However, we can say
with confidence that the way to provide a more
effective and less harmful treatment to patients
can be achievable.
Thank you!