Sunday, April 27, 2014

The puzzle of arthritis susceptibility, have we found the missing pieces?

 I was at a meeting of rheumatologists in Shanghai in April this year. This brought together scientists and clinicians from around the world to discuss the role of the genome and relevant biological pathways that influence susceptibility to arthritis.

Arthritis has many guises, it can result in painful hot and inflammed joints. If severe enough, the inflammation in patients with ankylosing spondylitis can result in a fused and deformed backbone and chronic disability.

There are a number of additional conditions that are associated with arthritis such as psoriasis and gut inflammation. This group of diseases are collectively known as the spondyloarthropthathies (SpA) since they exhibit shared but also distinct features.

Why does the clinician categorise and group arthritis conditions you might ask? The clinician’s desire to label and categorise is based on a need to understand the disease. Following a diagnosis, the patient can be treated in a consistent fashion and according to accepted guidelines. Originally a clinician grouped patients with similar features and symptoms and gave the condition a name. Giving a disease a name was a coup for a clinician, very much in the way biologists get excited about naming a new species of insect. As more technological advances were made, clinicians were able to examine the tissue microscopically and the biochemistry of the patients’ tissues and fluids. By comparison with healthy people, specific cells and molecules were found to be altered in patients. These features and biomarkers were helpful to monitor the disease course and test the effectiveness of different treatments. Different patient groups had shared and unique biomarkers, supporting a different mechanism of disease in the different patient groups. Things have moved on considerably over the last 10-15 years and scientists have delved even deeper, studying our genetic blueprint, the genome, a collective word for the DNA contained within all our chromosomes.

Why do some people develop an inflammatory arthritis or other autoimmune/inflammatory diseases and others do not? Is it a purely a chance event, influenced by lifestyle choices or by the environment that we live in? Maybe the events that predispose to arthritis aren’t so stochastic, scientists suggest that predisposition to disease may be more predictable. The critical finding in the 1990s was that diseases of SpA could be inherited and this was done by studying families and tracking the incidence of disease in twins and siblings. These studies discovered that if you have a twin who has ankylosing spondylitis, you are far more likely to develop the disease compared to a non identical brother or sister. This means that the more genes in common you have with an person who has SpA, the more likely it is that you will also develop the disease. Although I must mention that the presence of identical gene or genes only increases the risk of getting the disease, it is not a foregone conclusion.

Given the genetic link with disease, this makes the genome an exciting and 'fruitful' place to explore. New technologies have enabled us to study the genetic map, to find the genes that increase our risk of disease. Researchers have invested considerable time and money in characterising the genes that confer disease risk, namely the gene variants that are more prevalent in people with arthritis compared to the healthy population. These studies also identified the gene variants that were much more prevalent in the healthy population, and so revealed the gene variants that can provide protection from disease. So how are these variant genes identified? Variations in our genes between individuals or even on our pairs of chromosomes can be determined very quickly and easily using a technique called ‘SNP’ analysis. These single Gene Nucleotide Polymorphisms (SNPs) are essentially a tag to identify a region in the genome that has variability. These variable regions might directly identify the region that is conferring risk or track closely with regions of the genome that are associated with an individual having a greater or lower risk of developing disease. Analysing the prevalence of certain SNPs in healthy and patients groups is a really powerful way to find out whether certain ‘SNPs’ are associated with imparting increased risk to inflammatory disease and arthritis. A study that can detect association between variant regions in the genome and susceptibility to disease  is called a genome wide association study (GWAS).

It was the work of Matt Brown and his collaborators in the 2000s that pioneered the use of large cohorts of healthy and ankylosing spondylitis patients to identify ‘SNPs’ that were either under represented or over represented in people that had succumbed to arthritis. Once a disease SNP was identified scientists scrambled to identify which gene it might be tagging. Not all SNPs directly tag a gene and so there maybe a number of candidate genes close to the SNP that might be conferring disease protection or increased risk. The upshot of all this genetic anlysis and 'bean counting' ( ie statistics!) was that several genes were identified. One gene of particular interest was an immune hormone receptor, IL-23R. This provided important evidence that the IL-23 pathway influenced disease susceptibility in humans. This was an exciting finding as IL-23 was also shown to be a critical molecule in many animal models of arthritis

As we look for more and more genes, this is when GWAS gets tricky, when the effects are real but are subtle, so more individuals needed to be studied. Over the last 10 years, larger numbers of patients and healthy controls have been analysed by different research groups around the world and the data combined and analysed to achieve greater statistical power. These ‘super’ analyses essentially increased the chances of identifying many of the important genes that confer disease risk. It is important to bear in mind that this approach won't identify all genes that increase risk, there is a limit to the detection of associations of rare variants with disease and alternative approaches are required to identify these.

Identify the gene(s) and you identify the cause? Well yes maybe but it’s not as easy as you think. We have known since the 1980’s that people who expressed HLA-B27 (a specific subtype of an immune system molecule, see my previous blog) were much more likely to develop the debilitating rheumatological condition called ankylosing spondylitis. Having been aware that this gene imparts a 20 fold increased risk to developing arthritis, researchers have strived over the last 30 years to identify the nexus between this gene and the mechanism of the disease pathway, but this connection has remained elusive.

So has GWAS enabled us to get over this 30 year stumbling block with HLA-B27, I think it has and here is the reason why. It is really important that we pick up most of the genes that confer risk because they can then be mapped onto biological pathways. Following GWAS for ankylosing spondylitis, some very discrete pathways were identified, like a game of ‘join the dots’, several pathways started to light up with multiple hits. This provides important evidence for particular mechanisms of disease susceptibility and also identifies potential therapeutic routes for treatment.  The missing pieces of the puzzle are at last being found and a picture of the disease mechanism is starting to develop. With recent data from the GWAS, we can now be more confident about the  disease pathway that HLA-B27 is acting upon. Recent data has identified the genes that are working with HLA-B27 to increase that risk, and this information has guided the type of research that we do. Like a gold miner, if you are told where to dig for your gold you have a much better chance of finding it rather than digging around aimlessly hoping to 'strike it lucky'. Have arthritis researchers found their gold? This means for us to identify an approach to prevent, treat and cure arthitis. Well not yet, but we are digging in the right places, so watch this space.

Just before I finish, I just wanted to add that although it is becoming increasingly clear that differences in the human genome can contribute increased or decreased risk to susceptibility to arthritis, it is becoming apparent that disease risk may also be imparted by the presence of other genomes residing within our body. The presence of bacteria within our gastrointestinal tract provide us with trillions of genomes . These genomes contained within the micro-organisms that reside within our gut, are now being carefully analysed by DNA sequencing using the latest DNA sequencing technologies. Spurred on by a growing body of evidence that suggests that the diversity of bacteria (coined the microbiome), plays an important part in protection from or susceptibility to inflammatory diseases, arthritis researchers have begun to address whether arthritis patients exhibited similarities in their microbiome. This is a new and exciting area of research and hopefully I will be able to report back on this soon,

I haven't got time to write about the other features or imprints on our genome that exist, which may define the disease more closely or even provide more clues to the disease process, This is world of epigenetics which I will write about another time.

No comments: