I'm surprised to see an article on rheumatoid arthritis (RA) at the top of HN. RA is a complicated disease. This is interesting but there is already data impli

I'm surprised to see an article on rheumatoid arthritis (RA) at the top of HN. RA is a complicated disease. This is interesting but there is already data implicating infections like P. gingevalis, EBV, Parvovirus B19 with plausible mechanisms like citrullination of proteins, molecular mimicry contributing but there is no direct evidence and issues with these being prevalent in the general population. The question to ask when reading this are: Does this antibody/evidence of prior infection raise or decrease the probably of developing pre-RA? clinical-RA? Is it pathogenic or just an interesting finding? What's the prevalence of these antibodies in first degree relatives without disease or healthy controls in the general population e.g. can you reasonably use it for screening? I'd say the answer is no / we don't know to all of these and they it needs more research before this is clinically interesting. RA is a complicated disease but ultimately a clinical (/imaging) diagnosis based on synovitis and extraarticular features supported by investigations. There are mouse models of arthritis e.g. collagen induced but sometimes hard to translate to humans. If you are interested in the topic. Think of RA in terms of a susceptible host progressing to benign autoimmunity, subclinical disease, then clinical disease. Factors like HLA-DR4/DRB1*0401/0404 increases an individual's lifetime risk but are seen in 20-30% of population vs 0.5-2% of the population having disease so aren't helpful from a screening point of view. Environmental exposures like smoking (the data says in shared epitope + individuals) raise that risk significantly. Other exposures like microbiome are clearly important and they interact with the immune system (e.g. via TLRs) so they probably make a difference but aren't well studied (this work would fit in here trying to address it). There is good cohort data that many patients develop detectable autoantibodies years ('benign autoimmunity) before becoming patients with clinical disease. But even with everything Ive said this is still a two-way street: some of these individuals don't develop disease and can even lose their autoantibodies. In at least one real world cohort I'm aware it's been shown of you can be very high risk with ethnicity, CCP+, RF+, have a family member with RA (in some sense controlling for genetics, environmental exposures) and you still have even odds of progressing to disease vs. losing your double seropositive status and not developing disease. So what do you do with that? (It comes down a clinical diagnosis with prompt recognition, treatment to target of remission.) The (very important) trials so far for treating/preventing preclinical RA in ~similar high-risk individuals have been sadly negative or at best may delay onset slightly (abatacept, hydroxychloroquine).