Abstract
Psoriasis and psoriatic arthritis (PsA) are heterogeneous diseases. While both have a strong genetic basis, it is strongest for PsA, where fewer investigators are studying its genetics. Over the last year the number of independent genetic loci associated with psoriasis has substantially increased, mostly due to completion of multiple genome-wide association studies (GWAS) in psoriasis. At least 2 GWAS efforts are now under way in PsA to identify novel genes in this disease; a metaanalysis of genome-wide scans and further studies must follow to examine the genetics of disease expression, epistatic interaction, and gene-environment interaction. In the long term, it is anticipated that genome-wide sequencing is likely to generate another wave of novel genes in PsA. At the annual meeting of the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA) in Stockholm, Sweden, in 2009, members discussed issues and challenges regarding the advancement of the genetics of PsA; results of those discussions are summarized here.
- PSORIATIC ARTHRITIS
- PSORIASIS
- GENETIC ASSOCIATIONS
- PHARMACOGENETICS
- GENOME-WIDE ASSOCIATION STUDIES
- COPY NUMBER VARIATION
At the annual meeting of the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA) in Stockholm, Sweden. in 2009, members discussed issues and challenges regarding the advancement of the genetics of psoriatic arthritis (PsA). The major topics and the results of those discussions are summarized in a question and answer format here.
Why is there a paucity of large-scale genetic studies in PsA compared to psoriasis and other related autoimmune diseases (such as inflammatory bowel disease)?
Consensus was that this is partly due to the lower prevalence of PsA. The most recent estimate of the prevalence of PsA is 0.25% (95% CI 0.18%, 0.31%)1. Further, PsA occurs in approximately 30% (range 7% to 42%) of patients with psoriasis2. Fewer investigators are engaged in elucidating the genetics of PsA compared to psoriasis. GRAPPA appears to be an ideal forum to facilitate the interaction and communication among rheumatologists, dermatologists, biomedical researchers, and those with an expertise in genetic analyses. GRAPPA, whose members are primarily interested in PsA and who represent all these disciplines, is positioned to integrate the clinical and genetic data with the biological knowledge, which can then guide the statistical analysis to a more productive outcome.
PsA is a heterogeneous disease with 5 well recognized clinical patterns. What challenges will this create regarding the genetics of PsA?
One challenge is that there may be many patients with psoriasis and PsA who have not yet been diagnosed with PsA. Overall, however, the clinical heterogeneity in PsA will likely result in additional genetic heterogeneity. Further, each of these patterns may not be genetically distinct, as there is considerable clinical overlap between these subgroups. The clearest distinction with respect to the pattern of PsA is peripheral arthritis versus axial involvement. Investigating these 2 subsets is not as straightforward as it seems, as there is no uniform consensus regarding what constitutes axial involvement in PsA. Thus, detailed standardized phenotyping is very important. As DNA sequencing technology rapidly evolves, high quality sequencing data will soon become a reality. It is anticipated that detailed phenotypes will become the major hurdle to gene identification for complex genetic diseases. It is unlikely that a single center will be able to collect sufficient numbers of patients with PsA, so a genetic consortium involving multiple sites is essential to advancing the genetics of PsA. Again, this can be facilitated through GRAPPA.
If there were to be large international collaborative efforts for investigating the genetics of PsA, where would the DNA be stored?
The collection and storage of the genetic material is not straightforward. Numerous ethical and privacy concerns must be addressed. Institutions are also making increasing demands regarding ownership issues related to the biological material as well as intellectual properties that may arise from such studies. Before creation of a DNA biobank, addressing the issues of consent, disclosure, ethics, privacy, and intellectual property are very important. The best solution at present appears to be that the local investigators would be the custodians of their samples and that a memorandum of understanding would be signed between the site investigator and the genetic laboratory or the principal investigator(s) if the samples need to be transferred for genotyping.
The MHC region on 6p provides the strongest linkage with psoriasis and PsA; thus numerous candidate genes have been extensively investigated in this region. A detailed analysis of genomic DNA sequences and recombinant haplotypes demonstrated that HLA-Cw*0602 is the disease allele at PSORS1 for psoriasis. The precise identity of the HLA and non-HLA determinants within the MHC region still remains elusive in PsA.
The MHC (major histocompatibility complex) is a dense gene region that codes for a number of genes important in the immune response. HLA (human leukocyte antigen) studies in PsA have revealed numerous associations with disease susceptibility, expression, and progression. Briefly, HLA-B7, B13, B16 and its splits B38 and B39, and B17(B57), as well as B27 and HLA-Cw*06, have been associated with disease susceptibility in PsA3. HLA antigens appear to be important in disease expression: HLA-B27 is associated with involvement of the spine and B38 and B39 with peripheral polyarthritis, while HLA-Cw*0602 is associated with an earlier age of onset in PsA3.
Multiple non-HLA genes within the MHC region have been implicated in PsA and psoriasis. These include MHC Class I related (MIC) genes, TNF-α, and PSORS1C13. A metaanalysis confirmed an association between TNF-α-238 polymorphism and PsA with an odds ratio of 2.294. However, a recent German study reported that the association with 238 variant of TNF-α is dependent on carriage of the PSORS1 risk allele5. This illustrates the challenges in investigating genes within the MHC region, due to the extensive linkage disequilibrium that exists within this gene-enriched area.
Ideally, large-scale genetic studies with many informative recombinants are needed to convincingly demonstrate an independent association. Fine mapping of the MHC with over 2200 single-nucleotide polymorphisms (SNP) has been completed, and identified an association of PsA with SNP rs1150735 independent of known HLA associations6. The SNP resides 1.5 kb upstream of the gene ring finger protein 39, which has been shown to be associated with AIDS. A more detailed analysis of the MHC region with a much larger cohort is presently being conducted.
A number of genome-wide linkage studies have been completed in psoriasis; however, there is only one in PsA. Do we need more genome-wide linkage studies in PsA?
Multiple genome-wide linkage studies have been completed in psoriasis. The most consistently associated locus in psoriasis is within the MHC region at chromosome 6p21.3 (PSORS1)3. Nine other susceptibility loci also have been identified using the linkage approach, designated PSORS2 to PSORS103. Only one genome-wide linkage scan has been completed in PsA7, and it was later reanalyzed with additional markers in order to demonstrate significant linkage in chromosome 16q, close to a locus identified for psoriasis (PSORS8).
Despite numerous loci demonstrating significant linkage in psoriasis, there is lack of replication of many loci and an inability to consistently identify specific risk variants using this approach. In general, the genome-wide linkage studies were underpowered, given the modest effect size of the causative genes, lack of dense microsatellite coverage, and relatively small sample sizes. Thus, the role of future traditional genome linkage studies is likely going to be limited in PsA, unless we are searching for rare variant alleles.
The current strategy for identifying genetic variants in complex disease is to conduct a GWAS. This approach involves genotyping of up to a million SNP that characterize the common variants in the human genome. Robust non-MHC genes in psoriasis have been identified as a result of these scans. What is the status of GWAS in PsA?
A large GWAS from North America provided strong support for the association of at least 7 loci for psoriasis including HLA-C, 3 genes involved in interleukin 23 signaling (IL23A, IL23R, IL12B), 2 genes that regulate nuclear factor-κB (NF-κB) signaling (TNFAIP3, TNIP1), and 2 genes involved in the modulation of Th2 immune responses (IL4, IL13)8. Associations between higher genomic copy number for beta-defensin genes on chromosome 8, and LCE3B and LCE3C members of the late cornified envelope (LCE) gene cluster on 1q21 and psoriasis have been demonstrated using GWAS9.
Meanwhile, association studies in PsA have primarily involved candidate gene studies. The candidate gene association studies have identified a number of genes outside the MHC region, including the IL-1 gene cluster, killer-cell immunoglobulin-like receptor (KIR) genes, IL12B, and IL23R3. A larger number of non-MHC genes have been proposed and evaluated as potential candidate genes in PsA; however, most of these genes were not independently replicated3. An adequately powered GWAS in PsA has yet to be published, but multiple GWAS in PsA are now under way.
What are the challenges with GWAS and what is left to study with respect to genetic determinants in PsA?
Despite the rapid progress and current enthusiasm for GWAS, this technology has yet to identify the majority of genetic variance in risk for any complex disease. This is because a handful of loci identified from GWAS have been associated with modest risk. In fact, relative risks of most genetic loci identified are between 1.1 and 1.2. In order to identify a larger number of genes of mild to modest effect, we need to significantly increase the number of samples in these studies. It is highly unlikely that a single center can generate enough patients with PsA; a PsA genetic consortium is the best solution.
The assumption underlying the GWAS is that genetic variations with common alleles in the population will account for much of the heritability for complex disease. Although there is some theoretical justification for the common-disease, common-variant hypothesis, they may not be applicable to all human diseases. An alternative hypothesis is the rare-variant hypothesis, which postulates that there are multiple rare variants with strong genetic effect in complex diseases. Identifying these rare variants will be difficult due to the enormous amount of sequence variation present in individual genomes of both patients and controls; thus, large-scale sequencing will likely be necessary.
Copy number variation (CNV) is a common type of genomic variation ranging in size from 1 kb to several megabases and covering at least 25% of the genome. In the human genome, a single CNV can have different forms (i.e., duplication, deletion, inversion) among individuals, and these are likely to account for a significant proportion of human genetic variation. The chips currently used in GWAS can detect only a small percentage of CNV; thus, we must wait for advances in the sequencing technology before we can systematically interrogate for CNV. It is anticipated that CNV will be much more relevant for genetics of disease expression and personalized medicine than for genetics of disease susceptibility; again, careful phenotyping is important.
There is much interindividual variability in drug efficacy as well as adverse drug events with almost any treatment for psoriasis or PsA. Currently, pharmacogenetic studies in psoriasis and PsA cohorts are lacking. As the effect sizes for pharmacogenetic studies are also likely to be modest, each genetic variant in isolation is likely to have limited clinical value. Clinical, laboratory, radiological, and pharmacogenetic variables will likely need to be combined to assess the clinical utility of prospectively genotyping these variants for drug efficacy and/or toxicity.
In summary, PsA is a heterogeneous disease with a strong genetic component. Over the last year the number of independent genetic loci associated with psoriasis has increased substantially, mostly because of completion of multiple GWAS in psoriasis. Similarly, a GWAS in PsA appears to be the most rational approach to identify novel genes in PsA, and multiple efforts are under way in PsA. A metaanalysis of genome-wide scans and further studies must follow that will examine the genetics of disease expression, epistatic interaction, and gene-environment interaction. This will require additional sample collections with detailed phenotypic characterization. In the long term, it is anticipated that genome-wide sequencing is likely to generate another wave of novel genes in PsA.
REFERENCES
- 1.
- 2.
- 3.
- 4.
- 5.
- 6.
- 7.
- 8.
- 9.