Genome wide association studies in inflammatory myopathy and

IIM GWAS reveals genetic risk factors shared
with other autoimmune diseases (AID).
Robert G Cooper, on behalf of UK MYONET & MYOGEN
Salford Royal Foundation Trust & University of Manchester

• None
Conflicts of Interest

Polymyositis
Dermatomyositis
Idiopathic
Inflammatory
Myopathy
Malignancy
Myositis-CTD
overlap
Juvenile
dermatomyositis

Background
• IIM remain a therapeutic challenge –
1) Often refractory
2) Disease mechanisms poorly understood
• Mechanistic research difficult, as IIM so rare:
~ 1 new case per UK rheumatologist every 4-5
yrs.

Early genetics - HLA region

Log OR & 95% CI
Myositis Candidate Gene Studies Pre-2000
HLA-DRB1*0301
HLA-DQA1*0501
= US
Caucasian
= Korean
Rider LG et al. Arthritis Rheum 1999;42:1285-1290

Classification of myositis according to auto-Abs ?
•Interstitial lung disease
•Arthritis
•Mechanic’s hands
•Acute severe muscle
weakness
•Myalgias
•V-sign rash
•Shawl-sign rash
•Cuticular overgrowth
Courtesy of Prof FW Miller 1998

Methodological Issues in Pre-2000
Studies
• Small sample size
• Ethnic heterogeneity
– Caucasian, Jewish & Hispanic patients grouped
together to gain statistical power in case:control
comparisons.
• Sub-type heterogeneity
– DM, PM, JDM & IBM pooled to gain statistical power.
– Ab subgroups

UK-Adult Onset Myositis Immunogenetic
Collaboration (AOMIC) – RGC/WER0
• Only Caucasians studied
• Phenotype details, serum and
DNA
• Myositis probable/definite
according to Bohan/Peter
• PM (n=121)
• DM (n=110)
• CTD/overlap (n=71)
• Controls (n=537)
• [2000 – 2004 ~230 cases]

Aims of AOMIC
• Immediate - To recruit ethnically homogeneous
(Caucasian) myositis subtype cohorts of
sufficient size to allow between-subtype genetic
comparisons.
• Long-term - To elucidate aetiopathogical
mechanisms capable of facilitating future
therapeutic developments.

log OR & 95% confidence interval
HLA class II associations in 230 UK
PM/DM cases
10
6
4
=PM
=DM
2
1
.6
.4
.2
.1
DRB1*03
DRB1*07
DQA1*05
DQB1*02

LD - HLA class II haplotypes
DRB1*03-DQA1*05-DQB1*02
DRB1*07-DQA1*02-DQB1*02
PM DM
anti-synthetase
ILD
anti-Jo-1/PM-Scl
DM-specific
anti-synthetase
ILD
anti-Mi-2

Myositis-specific autoantibodies
Myositis specific autoantibodies
Anti-synthetase syndrome
Fever
Raynauds
Lung fibrosis
+/- DM rash
Myositis
Arthropathy
Mechanics hands
Necrotising
myopathy
High CK
Treatment
resistant
Clinical phenotypes in adults and children
Amyopathic
dermatomyositis
Rash sine myositis
Hypomyopathic
Rash precedes myositis
Dermatomyositis
Rash
Malignancy
Calcinosis/vasculitis
(children)
Jo-1
Anti-SRP
Anti-SAE
Anti-NXP-2
Zo
EJ
PL-7
Anti-synthetases
YRS
Anti-200/100
Anti-MDA-5
Anti-p155/140
Anti-Mi-2
KS
OJ
PL-12
Slide adapted from Dr Gunawardena

So, how to go about a GWAS in a
condition as heterogeneous as
myositis?

Dermatomyositis (DM)
• Myositis with the pathognomonic Heliotrope rash or
Gottron’s papules
• Adult-onset and juvenile-onset (age

Characteristic DM Histology

GWAS approaches 1
• MYOGEN (UK/EU/US), an international myositis
genetics consortium established to study the genetics of
myositis.
• Objectives of the first MYOGEN study were to identify
new genetic risk or protective factors for DM, and to
assess possible genetic overlaps with other AID.
• GWAS performed on subjects of European ancestry
meeting probable or definite Bohan and Peter criteria for
DM, using Illumina platforms.
• Subjects studied were 1178 DM cases (705 with adult
DM and 473 with juvenile DM) who were compared to
4724 geographically- and race-matched controls.

GWAS approaches (2)
• Further assessments on the DM GWAS data for 140
non-MHC SNPs previously associated with AID (RA,
SLE, Crohn’s, type 1 Diabetes & Multiple Sclerosis).
18

Covariate 2
Multidimensional scaling analysis of
European PM/DM GWAS data
MDS Dimensions 1 and 2
0.015
0.01
0.005
0
-0.02 -0.01 0 0.01 0.02 0.03 0.04 0.05
Covariate 1
-0.005
Czech (117)
-0.01
Hungary (89)
Netherlands (27)
-0.015
Sweden (79)
UK (319)
-0.02
-0.025

Covariate 2
Multidimensional scaling analysis of European
PM/DM GWAS data – NARAC controls
MDS dimension 1 and 2
0.02
0.015
0.01
0.005
0
-0.02 -0.01 0 0.01 0.02 0.03 0.04 0.05
-0.005
Covariate 1
-0.01
-0.015
-0.02
-0.025
-0.03
Cases (631) Controls (1140)

Populations Studied
Cases
Controls
Population
Sample
Size
Adult DM
Women
(%)
Juvenile DM
Sample
Size
Women
(%)
Sample
Size
Women
(%)
No. of
Successfully
genotyped
SNPs
Covariates
Genomic
inflation
factor (λ)
Czech/
Hungarian
178 70.80% 23 78.30% 256 57.40% 242530
Population
Structure
1.01
Spanish 43 81.40% 4 50% 259 65.60% 242871
Population
Structure
1.009
Swedish/
Dutch
48 68.80% 4 75% 642 72.40% 242644
Population
Structure
1.021
UK 149 65.80% 159 70.40% 2415 47.80% 236039 None 1
USA 287 76.00% 283 69.60% 1152 70.80% 237155
Population
Structure
1.073
Metaanalysis
705 72.4% 473 70.2% 4724 58.2% 241502 None 1.043
21

Manhattan Plot of DM GWAS Associations
Showing a Major MHC Signal
80 genotyped SNPs across the MHC region, but no other SNPs showed
Whole genome significance (orange line)
22

Published GWAS loci for AID –
Associations with DM (P ≤ 0.015)
Gene
Chr / Position
SNP Marker
in Dataset
SNP Marker
not in Dataset
: LD to
Dataset
Marker
Uncorr
P
B lymphoid tyrosine kinase: BLK 8 / 11381382 rs2736340 0.000065
Chemokine (C-C motif) ligand 21:
CCL21
9 / 34727828 rs2492358 rs951005:1.0 0.000209
CCL21 9 / 34733681 rs951005 0.00031
Protein tyrosine phosphatase nonreceptor
type 2: PTPN2
Signal transducer and activator of
transcription 4: STAT4
18 / 12799340 rs1893217 0.00289
2 / 191672878 rs7574865 0.00496
Interleukin-2 receptor alpha: IL2RA 10 / 6146272 rs7072793 rs706778:0.935 0.00727
ER-resident transmembrane protein:
ORMDL3
17 / 35319766 rs2290400 0.00997
SH2B adapter protein 3: SH2B3 12 / 110492139 rs653178 rs3184504:1.0 0.0138
SH2B3 12 / 110368991 rs3184504 0.0149
Interferon regulatory factor 5: IRF5 7 / 128381419 rs10488631 0.015
All of these associations are novel in myositis - No significant differences in these loci
were noted between males and females or between adult and juvenile DM cases
23

QQ plots showing published GWAS loci
for AID with DM (Left) that are not seen in
lung cancer (Right)
24

• The first GWAS in DM confirmed a strong signal in the MHC region as
expected, but no non-MHC SNPs met GWAS significance in this 1178 patient
sample.
• However, assessment of 140 non-MHC SNPs associated with AID indicates
that DM shares many genetic features with these other diseases and suggests
the presence of additional novel risk loci that require further study and
confirmation in DM.
• B lymphoid tyrosine kinase: BLK
Summary
• Chemokine (C-C motif) ligand 21: CCL21
• Protein tyrosine phosphatase non-receptor type 2: PTPN2
• Signal transducer and activator of transcription 4: STAT4
• Interleukin-2 receptor alpha: IL2RA
• ER-resident transmembrane protein: ORMDL3
• SH2B adapter protein 3: SH2B3
• Interferon regulatory factor 5: IRF5
25

Next Steps
• Immunochip analysis currently being undertaken in DM.
• Generate larger cohorts for DM, PM and the rarer MSAassociated
IIM subgroups.
• So, recruit, recruit, recruit!!!
26

Total number of participants recruited each year
Yearly accrual figures since AOMIC began in 2000, became UK
MYONET in 2008/9 (current recruitment total 922)
250
200
150
100
50
0
1 2 3 4 5 6 7 8 9 10 11 12
Years

MYOGEN Members
• Robert Cooper - The University of Manchester Rheumatic Diseases Centre, Salford, UK – on
behalf of UK MYONET
• Jiri Vencovsky - Institute of Rheumatology, Prague, Czech Republic
• Kati Danko - University of Debrecen, Debrecen, Hungary
• Fred Miller, Lisa Rider, Terrance O’Hanlon - NIEHS, NIH, Bethesda, MD
• Lucy Wedderburn, David Isenberg – Univ. College London, London, UK
• Ingrid Lundberg, Leonid Padyukov - Karolinska Inset, Stockholm, Sweden
• Lauren Pachman - Children’s Memorial Research Ctr., Chicago, IL
• Ann Reed, Steve Ytterberg - Mayo Clinic, Rochester, MN
• Albert Selva O’Callaghan - Val d 'Hebron General Hospital, Barcelona, Spain
• Tim Radstake - Nijmegen Ctr. Mol. Life Sciences, Nijmegen, Netherlands
● Paul Plotz, Mark Gourley - NIAMS, NIH, Bethesda, MD 28

• Hector Chinoy - Musculoskeletal Unit, Manchester HSC, University of
Manchester, Manchester, UK
• Janine Lamb, William Ollier - Centre for Integrated Genomic Med Res,
Manchester Academic HSC, University of Manchester, Manchester, UK
• Bo Peng, Wei Chen, Paul Scheet, Chris Amos - MD Anderson Cancer
Center, Houston, TX
• Annette Lee, Peter Gregersen - Feinstein Institute for Medical Research,
Manhasset, NY
• MYOGEN UK Contributors
MYOGEN Members (cont.)
– AOMIC/UK MYONET contributors (60+) – Coords. RGC/WERO
– UK JDM Research Group – Coord. Lucy Wedderburn
• SPONSORS – NIEHS NIH, EU AutoCure Myositis, UK Myositis Support
Group, ARUK, Cure Juvenile Myositis Association, European Science
Foundation (ESF) and the Wellcome Trust
29