PROJECT TITLE - ESTABLISHMENT OF A JUVENILE IDIOPATHIC
ARTHRITIS BIOBANK
HREC # 27127
PROJECT PROTOCOL
Section 1: Literature review
Juvenile idiopathic arthritis (JIA) is the most common chronic paediatric rheumatic
disease(1).. It is defined as arthritis affecting from one to all joints that begins before 16
years of age, and persists for at least 6 weeks. It occurs in children as young as 2 years of
age. The mean age at diagnosis is 8 years of age. It is an important cause of short-term and
long-term disability (1). Approximately one third of children still have active disease into
adulthood; one third have long term effects into adulthood, and one third of cases resolve
prior to adulthood. All cases require repeated clinical follow up. In severe cases, the
condition can affect normal growth and development, and can progress to life-threatening
complications.
The prevalence of JIA is likely to be underestimated. In an Australian community-based
survey in which school children were assessed by a paediatric rheumatologist, a prevalence
of four in 1000 was reported, much higher than previous reports based on clinical data (2),(3).
Management of this condition currently includes physical and occupational therapies,
psychosocial support, and pharmacological pain and inflammation management measures.
There is no cure for JIA.
The cause of JIA is poorly understood, but is known to be due to a mixture of genetic and
environmental factors(4).The complex nature of this condition means that determining such
factors is a difficult task, and one best achieved in large, well defined patient registers (5) with
a reference group of children without JIA. There are few such case registers worldwide, and
no such cohorts in Australia (6).A review of the scientific literature demonstrates that, of the
studies designed to identify genetic factors, most are not well designed, lacking thoughtful
genetic analytical approaches and statistical power. The outcome of such studies is that
progress in this area has been slow and unrewarding. In relation to environment, there are
very few studies that have addressed these factors, and we know little about the
environmental component of JIA disease causation.
Without knowledge as to the exact causes of JIA, it is virtually impossible to devise a cure or
determine preventive strategies, and very difficult to advance treatment options.
The single most important step in comprehensively studying JIA to determine its causes is to
have access to large numbers of JIA patients. In any complex disease, there will be a
number of individual contributing genetic and environmental factors which are important in
combination, but which on their own will confer modest increases in risk. Unlike single gene
disorders where inheritance of the predisposing gene will confer high or absolute risk of
developing the disease and is therefore relatively easy to identify in small numbers of cases,
in a complex condition such as JIA the effect of the contributing factor might be far less
absolute. To detect factors that confer modest risk with statistical certainty, it is necessary
to identify the increase in risk in a large sample size. We propose the establishment of a JIA
Biobank, gathering patients first from Victoria, and later throughout Australia, to achieve
this. Our open-ended goal is to collect 1000+ JIA patients for use in our future studies. This
will constitute the largest population-based JIA cohort in the world. This study is an initial
pilot study to be undertaken to establish the practicalities of undertaking a larger study.
The Biobank outlined in this proposal would collect clinical patient information such as age of
onset, severity and disease progression. It would also collect a comprehensive range of
environmental data (for example exposure to toxins such as cigarette smoke), shown to be
an important factor in adult rheumatoid arthritis and other autoimmune conditions (7) via
completion of a survey that asks questions related to early life, disease onset, disease
progression, and outcome. Additionally, we would collect and store a range of biospecimens
such as DNA for genetic analysis and serum for biomarker analysis, from each participant.
Study protocol, version 2
page 1 of 9
date: 22 October 2007
Section 2: Rationale for project
The rationale for the project includes the following:
1. Several autoimmune diseases, with pathogenesis likely to be similar to JIA, have
been shown to be dramatically increased in Victoria and worldwide. JIA has not been
studied in great detail and no large Southern Hemisphere cohorts exist. It is possible
and even likely that JIA is increasing in incidence as well.
2. In these other disease (e.g. Juvenile Diabetes, Crohn’s disease) the trends are
occurring over too short a time period to reflect genetics factors only, so
environmental factors should be examined.
3. Of the postulated adverse environmental factors, Victoria is a good setting to study
vitamin D insufficiency. Over a third (39%) of pregnant women in Geelong had
vitamin D insufficiency, (8) indicating sufficient prevalence for one of our main
exposures of interest, low serum 25OHD, is likely among the Victorian child sample.
This has been associated with Rheumatoid Arthritis, a disease which has many
parallels with JIA.
4. The research team has the mix of past experience and expertise to make a useful
international contribution.
5. Royal Children’s Hospital is the ideal location for this work as most Victorian JIA
cases (estimated to be 85%) are seen at the RCH Paediatric Rheumatology clinic.
6. This will be the first JIA biobank collected in Australia for genetic and epidemiological
research.
7. This will be the largest population-based JIA biobank collected anywhere in the world.
8. The size of the biobank will allow genetic analysis of novel candidate genes with
greater statistical power than previously achieved for this disease.
9. The cohort will yield important new information about the environmental component
to the disease which has not been studied previously.
10. The breadth of expertise in genetics and epidemiology available within the Murdoch
Childrens Research Institute will facilitate a unique multi-faceted approach to the
study of this disease which will yield new knowledge likely to translate to better
health outcomes for future JIA patients.
Section 3: Aims
Our research aims are to:
1. Gather comprehensive environmental profiles (e.g sibling number and day care
attendance as markers of early life infection, or exposure to toxins such as cigarette
smoke) of prospective cases.
2. Build a biobank of biological materials such as DNA and serum from retrospective and
prospective JIA cases for genetic and biomarker research studies.
3. To use the resources gathered in Aims 1 and 2 to compare the genetic and
environmental profile of newly incident JIA cases to control children.
4. To use the resources gathered in Aims 1 and 2 to investigate genotype-phenotype
associations within the JIA biobank.
Study protocol, version 2
page 2 of 9
date: 22 October 2007
Section 4: Hypothesis/research questions
1. The early life environment of JIA cases differs to control children with regard to
several factors including breastfeeding, passive smoke exposure and a reduced
exposure to early life infection using markers such as day care attendance and sibling
number.
2. The early life environment of JIA cases differs to control children with regard to
vitamin D supplementation, dietary intake and sun exposure.
3. The genetic profile of JIA cases differs to healthy controls with regard to several
immune-related genes such as CTLA4 and ILR1.
4. Among cases, genetic factors such as HLA type are associated with disease severity.
Section 5: Methodology and inclusion/exclusion criteria
Participants
Group A: Children and adolescents with newly diagnosed JIA and
Group B: Children and adolescents with JIA (retrospective cases)
Children and adolescents with newly diagnosed JIA (identified prospectively) as well as
children and adolescents with JIA as identified on the Paediatric Rheumatology clinic
database. Collecting both prospective and retrospective cases will provide a number of
immediate and future opportunities for the study of JIA. Participants will be recruited by the
Clinical Research Nurse, and will be asked to complete surveys for collection of
environmental data. Biospecimens such as peripheral blood leukocytes, plasma and serum
will be collected during routine disease management blood tests. Repeated clinical contact
with affected families will provide strong potential for gathering of clinical and environmental
data and biospecimens in a longitudinal fashion. As already stated, all children diagnosed
with JIA require repeated clinical follow up.
Group C: Controls – Day Surgery
Patients on the elective Day Surgery lists will be recruited to form a reference paediatric
control population that will be useful for genetic and environmental comparison. Participants
will be recruited by the Clinical Research Nurse, and will be asked to complete surveys for
collection of environmental data. A blood sample will be obtained whilst the patient is
anaesthetized and biospecimens such as peripheral blood leukocytes, plasma and serum will
then be stored and analysed. We will carefully consider the reasons for day surgery and how
they may impact on case control comparisons and conduct subgroup comparisons
accordingly. By restricting recruitment to Victorian born controls we will also be able to
compare the overall peri-natal characteristics of the sample for that of Victorian births as a
whole, using published Statistical profiles of Victorian live births(9),(10).
Study protocol, version 2
page 3 of 9
date: 22 October 2007
Inclusion Criteria
Group A: Children and adolescents with newly diagnosed JIA

Children / adolescents aged 0-18 years.

A patient of the RCH private or public clinic.

Diagnosed with JIA after the commencement of the study

Are alive during the period of study.
Group B: Children and adolescents with JIA (retrospective cases)

Children / adolescents aged 0-18 years.

A patient of the RCH private or public clinic

Diagnosed with JIA since 1997 (onwards)

Are alive during the period of study.
Group C: Controls – Day Surgery

Children / adolescents aged 0-18 years.

A patient of the RCH attending the Day Surgery Unit for elective surgery

Are alive during the period of study.

Victorian-born
Exclusion Criteria
Group A: Children and adolescents with newly diagnosed JIA

Other medical co-morbidities such as major illness that would forgo attendance at a
normal Victorian school in the one year prior to the study

Major congenital abnormalities.
Group B: Children and adolescents with JIA (retrospective cases)

Other medical co-morbidities such as major illness that would forgo attendance at a
normal Victorian school in the one year prior to the study

Major congenital abnormalities.
Group C: Controls – Day Surgery

Other medical co-morbidities such as major illness that would forgo attendance at a
normal Victorian school in the one year prior to the study

Major congenital abnormalities.

Children born outside Victoria.
Study protocol, version 2
page 4 of 9
date: 22 October 2007
1. Study overview
A matched case-control study (n =1000 cases JIA and n=1000 age and sex matched
controls, N=2000) with extensive environmental and phenotype measures and careful
storage of samples for future epigenetic, genetic and immunological work. The two year pilot
study aims to recruit 100 new cases of JIA and 100 controls.
2. Source Population
This study has multiple aims, and the optimal source populations for each differ.
Source population characteristics relevant to all aims are that children are under the age of
18 years and are residing in Victoria during the study period.
For Aim 4, we will require a large case series to investigate gene-phenotype associations.
Cases, regardless of where they are born, are of interest.
For Aim 3 we will use only the Victorian born cases (a subset of all cases examined in Aim 4)
and compare them to Victorian born controls. This provides several advantages:1. It will insure that the cases and controls both arise from the same population.
2. This means that questions about one of our main environmental exposures, past sun
exposure as measured by time outdoors/in sun, will be standardised for ambient
UVR, at least at the state-level
3. It will rule out problems that may occur if the diagnosis of past conditions (e.g
asthma) are region dependent- thus a child who migrated from India 6 months ago
may have a different probability of disease diagnosis for the main outcome (JIA) or
co-morbidities e.g asthma
4. It will reduce, although not eliminate population stratification issues in the genetic
work
5. It will provide an ability for us to adjust for the non-representativeness of the hospital
controls in the future, considering a comparison to all Victorian live births and
adjusting accordingly to re-weight under or over represented groups.
We have used a similar approach in the Tasmanian Multiple Sclerosis case control program,
where, although some of the genetic work included all cases, regardless of residence (Hum
Genet. 2004 May;114(6):573-80), the case control component, investigating environmental
(BMJ. 2003 Aug 9;327(7410):316, JAMA. 2005 Jan 26;293(4):463-9 ) or geneenvironmental effects (work in progress) was restricted to the subgroup of cases with a
grandparent born in Tasmania and matched controls with the same criteria, for the reasons
outlined in points 1., 2 and 4 above. Here we consider point 5 additionally important as we
are not using a population-based control sample.
3. Juvenile Idiopathic Arthritis cases
Group A & B: Children aged 0-18 years living in Victoria during the study period either newly
diagnosed with JIA (Group A) or diagnosed with JIA since 1997 (Group B)
Using data from previous RCH based Rheumatology research we estimate that
approximately 200 early-onset cases will arise during the two years of study fieldwork, of
which we aim to collect 100.
Note that to ensure we are collecting a case group that is representative of the Melbourne
Metropolitan region as a whole, following a start-up period for this project we plan to recruit
cases from other paediatric rheumatology clinics, in particular Monash Medical Centre, via
the Victorian Paediatric Rheumatology Consortium. We shall apply for an amendment to our
ethics approval prior to commencing this phase of the project.
4. RCH Day Surgery based controls
Controls will be children under the age 0-18 years who were born in Victoria attending the
Day Surgical Unit for elective surgery, and who are residing in Victoria during the study
Study protocol, version 2
page 5 of 9
date: 22 October 2007
period.
5. Study measurements
The questionnaire will include extensive environmental assessment related to pregnancy,
infancy and childhood, consistent with the key hypotheses. Most study measurements will be
collected in a face-to-face interview and with a trained research nurse and participant. The
research nurse will read through the questionnaire with the participant and their
parent/guardian and be present as it is completed to answer any queries. The interviews will
be conducted at the Royal Children’s Hospital, Melbourne. On occasion a follow-up phone call
may also need to be made (eg to check data from child’s clinic book, if this was not available
at interview). The questionnaire has been weighted towards questions that will be less
affected by recall bias, and differences in recall bias between incident and prevalent JIA
cases will be assessed by sensitivity analysis to ensure that any exposure-disease
associations are of a similar magnitude in both groups.
We will collect data on:
5.1 Sun exposure and vitamin D. Validated questionnaire measures of recent summer
and winter sun exposure, which we have previously shown to agree (ICC = 0.62) with
personal sun exposure measured by polysulphone personal UVR monitors will be used for
maternal antenatal and postnatal behaviour. For infant and child sun exposure, we will use
protocols with an emphasis on those developed by ourselves (11), (12-18) and other Australian
groups, (19-21) including questions on daycare sun policy. Lifetime sun exposure and skin
type: tendency to sun-burn, tanning ability, sunburn history, sun-protection behaviours and
past history of outdoor activity levels in summer and winter from birth to current age will be
recorded(17).Data on the timing of exposures by age will also be collected. Natural hair colour
and eye colour will be recorded. We will measure spectrophotometric skin type. In
Caucasians, we have shown the spectrophotometric reflectance at wavelengths 400 - 420
nm correlated highly (r = 0.68) with histological melanin (22).Skin pigmentation also increases
with age in UV-exposed but not UV-unexposed sites, thus a ratio measure can provide a
measure of cumulative UV exposure. Serum 25OHD levels will be measured by
radioimmunoassay (DiaSorin, USA) with an intra- and interassay precisions of these assays
are 6% and 15%, respectively.
5.2 Other related factors. JIA is likely to be multifactorial in aetiology, so we will collect a
range of other exposure measures. We will record sibling structure, dates of birth and sex of
siblings and half siblings – as this allows us to calculate cumulative exposure to siblings of
specified ages prior to a specified age for the index subject, providing a proxy marker for
early life infection that has been extremely useful in investigating MS. (23-24) We will also
collect data on past child infection, day care and other indices of microbial exposure in early
life. Serum will be stored for future possible quantitative titres of IgM or IgG antibodies to
common microbial infections in childhood.
We will record antenatal maternal diet, smoking, alcohol and medication use, and
demographic variables including parental ancestry and ethnicity. Data will be obtained on
past medical history, family history of rheumatic diseases, asthma and other immune
disorders and skin cancer; use of supplements including vitamin D and cod liver oil –
(antenatal, birth to current); breast-feeding history and age of introduction of cow’s milk
and foods, foetal, infant and child exposure to tobacco smoke and child physical activity (25).
We will also measure weight and height.
5.3 Biological samples.
For children with JIA, at the interview, an anaesthetic patch will be administered prior to
blood taking and 5 to 10 ml of venous blood will be collected then centrifuged. The control
groups will have venous blood (5-10 ml) taken when already anaesthetized for their primary
purpose of having a day care procedure. DNA will be extracted. Serum specimens for 25OHD
at child interview and other tests will be stored at -80 degrees Celsius.
Study protocol, version 2
page 6 of 9
date: 22 October 2007
5.4 Genotyping. The approach to detecting genes associated with JIA will be similar for
each gene chosen (based on information from other autoimmune diseases, including VDR,
HLA, PTPN22, CTLA4 etc). We shall use publicly available data on single nucleotide
polymorphism (SNP) linkage disequilibrium throughout each candidate gene to select ‘tagSNPs’, defined as the minimum number of SNPs required to comprehensively examine
sequence variation within, and surrounding, each gene of interest. These SNPs will be
genotyped using the Sequenom MassARRAY system in each case and control participant.
Statistical techniques such as regression will be employed to determine whether there are
significant differences between the allele frequencies of each SNP between cases and
controls that would indicate a difference in risk attributable to the inherited allele. The
number of tag-SNPs required to be genotyped will be different for each gene studied, for
example, 80 SNPs require genotyping for the VDR gene.
6. Timetable
One hundred children with incident JIA and 100 controls would be recruited in the first 2
year period. Thereafter (funding permitting), we will recruit 60-100 incident cases and 100 150 retrospective cases, and 200 - 250 matching controls per year for a further four years,
to achieve our goal of 1000 JIA cases in six years. We will be in a position to begin genetic
and epidemiological analyses of some factors once a total of 100 cases and 100 controls are
reached (estimated to occur before the end of year 2). From this time point, laboratory
studies will be performed in 6 monthly to one year batches to allow the collection of
genotypic information to keep pace with recruitment.
7. Statistical Power
The following calculations are based on a power of 90% and statistical significance at the 5%
level. In the first two years of the study, with 100 cases and controls we will be aiming to
identify factors with odds ratios of 3 or greater and the power (beta) for doing so varies as
follows by exposure prevalence: 10% exposure, 80% power; 20%, 93% power, and 40%,
96% power. The magnitude of the odds ratio is high, but this is a novel study and odds
ratios greater than three have been found for markers of a hygienic early life (sibling
patterns, daycare) and low vitamin D intake for another child autoimmune disease, type 1
diabetes. In fact, no vitamin D supplementation was associated with more than 4.5-fold
increase in T1D risk (26).
With regard to the eventual sample of 1000 cases and 1000 controls, individual gene
variants or environmental factors with an odds ratio of 1.5 for JIA will be detectable with
power varying from 84% for exposure present in 10% of the population and higher if
exposure levels are higher. All power calculations use the US National Cancer Institute’s
method of assessing power in gene and environment case control studies (27).
Section 6: Study analyses
The distributions of exposures will first be examined. Logistic regression will be the primary
method for analysis of the case control study, providing odds ratios (95% CI) for
associations between gene variants or environmental factors and JIA after adjustment for
relevant confounding factors (28). In all analyses, careful attention will be placed on assessing
the potential for bias due to disease-related changes in recall, behaviour or biomarkers and
potential bias introduced by the selection of controls from elective day care lists. We will also
assess whether third variables, such as early life infection, are true confounders, partial
intermediates or effect modifiers of the observed associations using standard epidemiological
and biostatistical techniques to assess causal pathways (29).
Study protocol, version 2
page 7 of 9
date: 22 October 2007
Section 7: References
1.
2.
3.
4.
Ravelli A, Martini A. Juvenile idiopathic arthritis. Lancet 369:767-78, 2007
Manners & Diepeveen, Pediatrics 98:84, 1996;
Manners & Bower, J Rheumatol 29:1520, 2002).
Phelan JD, Thompson SD, Glass DN. Susceptibility to JRA/JIA: complementing general
autoimmune and arthritis traits. Genes and Immunity 7:1-10, 2006
5. Hattersley AT, McCarthy MI. What makes a good genetic association study? Lancet
366:1315-23, 2005
6. Phelan & Thompson, Curr Opin Rheumatol 18:482, 2006).
7. Costenbader & Karlson, Lupus 15:737, 2006
8. Morley R, Carlin JB, Pasco JA, Wark JD. Maternal 25-hydroxyvitamin D and parathyroid
hormone concentrations and offspring birth size. Journal of Clin Endocrinol Metab
2006;91:906-12.
9. Riley M, Griffin O. Validating a statewide data collection: differences in information
technology resources between hospitals. Health Inf Manag. 1997 Jun-Aug;27(2):67-8.
10. Riley M, King J. Births in Victoria 2001-2002; 2003
11. van der Mei IAF, Ponsonby AL, Dwyer T, Blizzard L, Simmons R, Taylor BV, Butzkueven
H, Kilpatrick T. Past exposure to sun, skin phenotype, and risk of multiple sclerosis: case
control study. BMJ 2003; 327 (7410): 316-322.
12. Jones G, Dwyer T, Hynes KL, Parameswaran V, Greenaway TM. Vitamin D insufficiency
in adolescent males in Southern Tasmania: prevalence, determinants, and relationship
to bone turnover markers. Osteoporos Int. 2005; 16(6):636-41
13. Jones G, Dwyer T. Bone mass in prepubertal children: gender differences and the role of
physical activity and sunlight exposure. Journal of Clinical Endocrinology and
Metabolism. 1998;83:4274-9.
14. Blizzard CL, Dwyer T, Ashbolt R. Changes in self-reported skin type associated with
experience of sunburning in 14-15 year old adolescents of Northern European descent.
Melanoma Research. 1997;7:339-346.
15. Dwyer T, Blizzard CL, Ashbolt R. Sunburn associated with increased number of nevi in
darker as well as lighter skinned adolescents of Northern European descent. Cancer
Epidemiology, Biomarkers and Prevention. 1995;4:825-830
16. Dwyer T, Blizzard CL, Gies PH, Ashbolt R, Roy C. Assessment of Habitual Sun Exposure
in Adolescents via Questionnaire - A Comparison with Objective Measurement Using
Polysulphone Badges. Melanoma Research. 1996; Vol 6 No. 3:231-239.
17. van der Mei IAF, Blizzard L, Ponsonby A-L, Dwyer T. Validity and reliability of adult recall
of past sun exposure in a case-control study of Multiple Sclerosis. Cancer Epidemiol
Biomarkers Prevention. 2006;15(8):1538-44
18. van der Mei I, Ponsonby A-L, Dwyer T, Blizzard L, Kilpatrick T, Butzkveven H, McMichael
AJ. Vitamin D levels in people with multiple sclerosis and community controls in
Tasmania, Australia. J Neurology 2007 May;254(5):581-90
19. Harrison SL, Buettner PG, Maclennan R. The North Queensland "Sun-Safe Clothing"
study: design and baseline results of a randomized trial to determine the effectiveness of
sun-protective clothing in preventing melanocytic nevi. American Journal of Epidemiology
2005;161(6):536-45.
20. Whiteman DC, Brown RM, Purdie DM, Hughes MC. Melanocytic nevi in very young
children: the role of phenotype, sun exposure, and sun protection. Journal of the
American Academy of Dermatology2005;52(1):40-7.
21. Fritschi L, Battistutta D, Strutton GM, Green A. A non-invasive measure of photoageing.
Int J Epidemiol 1995;24(1):150-4.
22. Ma X, Buffler PA, Layefsky M, Does MB, Reynolds P. Control Selection Strategies in CaseControl Studies of Childhood Diseases. Am. J. Epidemiol. 2004;159:915-921
23. Dwyer T; Ponsonby AL; van der Mei I; Blizzard L; Taylor B; Kemp A; Kilpatrick T.
Exposure to Infant Siblings During Early Life and Risk of Multiple Sclerosis—Reply. JAMA
2005; 293: 2089 - 2090.
24. Ponsonby A-L, Dwyer T, van der Mei I, Kemp A, Blizzard L Taylor B Kilpatrick T
Simmons R. Asthma onset prior to multiple sclerosis and the contribution of sibling
exposure in early life. Clinical Experimental Immunology 2006;146(3):463-70
25. Rockell JE, Green TJ, Skeaff CM, et al. Season and ethnicity are determinants of serum
Study protocol, version 2
page 8 of 9
date: 22 October 2007
25-hydroxyvitamin D concentrations in New Zealand children aged 5-14 y. J Nutr
2005;135:2602-8.
26. Hypponen E, Laara E, Reunanen A, Jarvelin MR, Virtanen SM. Intake of vitamin D and
risk of type 1 diabetes: a birth-cohort study. Lancet 2001;358(9292):1500-3.
27. Garcia-Closas M, Lubin JH. Power and sample size calculations in case-control studies of
gene-environment interactions: comments on different approaches. Am J Epidemiol
1999;149(8):689-92.
28. Hosmer DW, Lemeshow S. Applied Logistic Regression. 2nd ed. Toronto: John Wiley and
Sons Inc., 2000.
29. Rothman KJ, Greenland S. Modern Epidemiology. Second ed: Lippincott-Raven
Publishers: Philadelphia, 1998.
Appendices
Appendix
Appendix
Appendix
Appendix
Appendix
Appendix
Appendix
Appendix
1 Protocol
2A Information sheets and
2B Information sheets and
3A Information sheets and
3B Information sheets and
4 Project Budget
5 Child Questionnaire
6 Parent Questionnaire
Study protocol, version 2
consent
consent
consent
consent
forms
forms
forms
forms
page 9 of 9
(participant) – JIA cases
(participant) – controls
(parent/guardian) – JIA cases
(parent/guardian) – controls
date: 22 October 2007