October 14, 2025 | As people with rheumatoid arthritis (RA) well know, it is a “forever disease” with no single magic-bullet cure or guarantee that they will ever feel like they did before their diagnosis. But a better understanding of what’s happening biologically before the first swollen joint appears could reveal opportunities to markedly modify the course of RA and, ultimately, prevent it, according to Kevin Deane, M.D., Ph.D., professor of medicine at the University of Colorado Anschutz., who has worked extensively with people who are at risk for RA.
As he and his colleagues at the Allen Institute for Immunology and UC San Diego have most recently shown, in the years before the swollen joints of RA appear there are dramatic immune system changes that can be mapped (Science Translational Medicine, DOI: 10.1126/scitranslmed.adt7214). This could enable better prediction of who will develop full-blown disease as well as expose the underlying mechanisms driving those processes.
An anti-CCP (cyclic citrullinated peptide) blood test, also called an ACPA (anti-citrullinated protein antibody) test, already exists for the identification of high-risk individuals, in many cases years before disease onset, says Gary Firestein, M.D., senior associate vice chancellor for health sciences at UC San Diego. The problem is that only about one-third of the people flagged go on to develop RA.
In prevention studies, therefore, two-thirds of enrollees were never going to get the disease anyway, he adds. Better markers for identifying those at the highest risk of progressing to clinical RA would translate into shorter studies requiring fewer participants.
The quest to distinguish ACPA-positive individuals who go on to develop RA undertaken with the latest, seven-year study involved transcriptome and proteome profiling as well as immune cell phenotyping. All the resulting multiomics data is available for use by other researchers who “don’t necessarily have to ask the same questions we asked,” says Mark Gillespie, Ph.D., assistant investigator at the Allen Institute in Seattle and a senior author on the study.
“What we found almost appears to be an overactive immune system very early on before people in the study have the actual disease,” he says. Importantly, some of the changes found mirrored those previously seen in clinical disease, including B cells shifting toward a pro-inflammatory state and T helper cells well above normal levels.
This strongly suggests that RA starts at the molecular level well before diagnosis, says Gillespie. The immune signatures affect multiple immune compartments, including monocytes migrating to the tissues and B and T cells that interact with each other to drive the production of the auto-reactive antibodies that are a key feature of RA.
Particularly interesting was the finding that naïve cells that had never encountered antigen before were being reprogrammed, says Gillespie. “Something is fundamentally ... different about these cells,” suggesting that they had undergone epigenetically controlled changes in their response to RA that is perhaps a common feature across auto-immune conditions generally.
Population-Based Screening
Longitudinally profiling individuals at risk for RA, as was done here with integrative multiomics, came with the difficulty of identifying study participants who have ACPA positivity but not full-blown RA. The research team had to be highly proactive in finding people who were ACPA positive by testing first-degree relatives of patients with RA, or who had already had a positive anti-CCP test, says Firestein. That search extended to health fairs, social media, community outreach, and clinic referrals.
For the average person who doesn’t have a relative with RA, the lifetime risk of developing the disease is around 0.5% to 1%, says Deane. But even for people who have a family member with RA, that figure only goes up to between 5% and 7%. “As a result, most of rheumatoid arthritis occurs in people who don’t have a known relative with the disease.”
Consequently, RA screening processes in the future will likely include testing of family members of patients with RA as well as population-based testing, Deane says, much as is currently done for heart disease with routine blood pressure, cholesterol, and blood glucose checks. But the screening test will need to be more robust than solely depending on the detection of the ACPA marker, perhaps by offering a deep genomic and proteomic assessment of the flagged individuals to confirm their likelihood of progressing to clinical RA.
The current reality is that people with RA start getting treatment only after symptoms have appeared and they have swollen joints, says Deane. “For a lot of people who get rheumatoid arthritis, treatments are quite good, and they can return to normal life,” but 40% to 50% of people with the disease continue to experience persistent symptoms such as pain, joint stiffness, and fatigue.
It has been well documented that “early treatment is more effective than waiting until somebody has longstanding disease,” adds Firestein, “so the earlier you start treatment the better.” It is also more likely that they will go into remission and will have no joint damage going forward.
Collaboration has been key to the team’s RA-related research successes to date, which began after the founding of the Allen Institute for Immunology in 2018 with former Eli Lily executive Tom Bumol, Ph.D., at the helm, says Firestein. One of his early stops was at UC San Diego where he was pitched the idea of following patients at high risk for RA over time and collecting blood samples from them and, whenever possible, synovial tissue lining their joints to undertake the massive multiomics project.
Researchers at UC San Diego could not have done this on their own, he stresses. It was only natural to reach out to established collaborators, including Deane and V. Michael Holers, professor of medicine-rheumatology, at the University of Colorado, and Jane Buckner, M.D., president of the Benaroya Research Institute, as well as Gillespie, who leads the autoimmunity program at the Allen Institute for Immunology. “We were essentially getting the band back together.”
The finding about cellular reprogramming described in their paper refers to a change in the chromatin structure of cells where DNA is balled up, Gillespie explains. For a gene to be expressed, some unwinding is necessary, which is where some of the assays came in to take measurements in open regions of the genome. At that level, immune cells in at-risk individuals were found to have different DNA packaging, and thus different accessibility to transcription factors, than people who were not at risk.
“Some of the next steps, which are already in progress by this group, are to define the cell communication networks because these cells are not floating around in a vacuum but are interacting with each other,” says Firestein. “We’re now trying to define which cells are producing which mediators and which cells are receiving the signals from those mediators, and that’s the essential way that a cell will respond to its environment and ultimately be reprogrammed—either by expressing genes that are associated with inflammation or reprogramming the chromatin in the epigenome to imprint it ... [with] a different phenotype than a resting cell.”
The team is working toward diagnostic and treatment alternatives in parallel, Firestein says. “One of the most interesting aspects of rheumatoid arthritis is there is a diversity of responses to highly targeted agents,” he notes. Among these drugs are agents like tumor necrosis factor blockers and abatacept, which blocks T cell activation, and individually help “a different 30% to 50%” of patients.
The implication is that “there is no single mechanism of disease in rheumatoid arthritis,” Firestein continues. Mapping the pathways, as this team has done, could allow patients to be stratified as well as uncover everyone’s mechanism of disease.
“In some people, it might be B cells, and in some people, it might be monocytes and macrophages, and in others it might be CD4 T cells,” he says. “Once we understand these pathways and communication networks better, not only will it hopefully help us personalize medicine better but also to identify new and interesting therapeutic targets to treat that 40%-plus of individuals that are never fully controlled by current treatment.”
The trick will be how to translate information on the pathways, genes, and cell types implicated in the latest study into a therapeutic, says Firestein. “That oftentimes takes collaboration with the private sector,” which the team is confident will happen based on the new research findings.
The good news, according to Deane, is that a half dozen RA prevention trials have been completed worldwide. “The concept is there ... [but] most of those trials have targeted similar [ACPA-positive] people that we had in our collaborative study ... and used drugs already established in rheumatoid arthritis.”
There were some modest successes in the prevention studies, he adds, but the newly available insights about the immune system in individuals who go on to develop RA “will help inform the next round of clinical trials” that could soon be in the offing. Deane was the lead author on the StopRA prevention study that was recently published in Arthritis & Rheumatology (DOI: 10.1002/art.43366).
This was a phase 2 randomized trial of hydroxychloroquine that has been used without supporting evidence to prevent RA, in at-risk individuals based on elevated levels of anti-CCP antibodies. The trial found that hydroxychloroquine did not prevent RA, which supports avoiding use of hydroxychloroquine in similar populations for RA prevention in the future.
Importantly, the new discoveries in the Science and Translational Medicine paper will aid upcoming studies by improving prediction and providing novel targets for preventive interventions. Data sharing by the Allen Institute should also foster others “to ask questions that we maybe haven’t thought of to drive forward discovery in this area,” says Deane.
“As a reminder, this was not the end of the story ... [but] the beginning,” says Firestein. “The team has provided a roadmap for the journey from healthy to at risk to clinical rheumatoid arthritis, and like many maps there are lots of different routes you can take to that destination. We’re now at the point where we can see the roads, we can identify the intersections and know, hopefully, where they’re taking people.
“But what we don’t know,” he continues, “are the most efficient routes or how to intercede ... to stop people from progressing.” Much work remains to be done now that steps have been taken in that direction.
None of this work happens without people enrolling in studies, points out Firestein, which could be more challenging with prevention trials involving people who are feeling pretty good. “Unless there’s a high index of suspicion, like a family member who has the disease, they may not be thinking about getting ... the ACPA test or participating in prevention trials.”
But individuals could be incentivized to sign up out of altruism, one of the main drivers of research participation. As the latest study suggests, test procedures like blood draws and synovial biopsies won’t be dealbreakers, he continues. “We’ve been very surprised and gratified by the individuals who have agreed to allow us to sample joint tissue.”