Rheumatoid arthritis: prediction of future clinically-apparent disease, and prevention

Abstract

Purpose of Review

This review discusses updates in the prediction and prevention of future rheumatoid arthritis (RA).

Recent Findings

In individuals with musculoskeletal symptoms and elevated antibodies to citrullinated proteins (ACPA) without clinical inflammatory arthritis (IA), a ‘simple’ score has a positive predictive value (PPV) of ~28% for clinical IA/RA within 1 year, and a comprehensive score (including ultrasound) has a PPV of ~71% for clinical RA within 5 years. Controlled clinical trials in individuals at-risk for future RA have been performed using corticosteroids, rituximab, atorvastatin, methotrexate, hydroxychloroquine and abatacept. Abatacept modestly reduced rates of incident clinical RA within the trials (and imaging inflammation), rituximab delayed clinical IA, and methotrexate improved function, symptoms and imaging inflammation. Vitamin D with or without omega 3 fatty acids reduced incidence of autoimmune diseases, including RA. While not proven in controlled clinical trials, observational studies suggest exercise, weight loss and smoking cessation may reduce progression to clinical RA.

Summary

Prediction and prevention of RA is advancing although there are no currently approved interventions for prevention. Future studies should include deeper evaluation of the pathophysiology of RA development to improve prediction and identify key pathways to target in future clinical trials, as well as develop infrastructure to support prevention-related research.

Introduction

It is now well-established that immunologic abnormalities, symptoms and even imaging abnormalities may be present in a ‘pre-RA’ period that occurs prior to the first appearance of findings on physical examination of inflammatory arthritis (IA) that is the hallmark of rheumatoid arthritis (RA)(Figure 1)[1].

Figure 1. Natural history of rheumatoid arthritis development.

In this model, rheumatoid arthritis (RA) can progress from initial breaks in tolerance to systemic autoimmunity without clear articular disease, to articular symptoms and/or imaging abnormalities without clinical finding of inflammatory arthritis (IA), and ultimately to clinical RA. The double-headed arrows indicate that it is possible for an individual to progress or regress from a given stage.

This model of the natural history of RA has been developed through a number of natural history studies that have included retrospective case-control studies using biobanked samples collected from prior to the onset of RA[2–4] as well as prospective studies of groups considered at higher risk for RA such as North American indigenous populations[5, 6], first-degree relatives of individuals with established RA[7–9], as well as prospective follow-up of individuals who present to clinical care with musculoskeletal complaints and are found to have autoantibody elevations and/or imaging abnormalities[10–12]. Notably, the pre-RA period is best studied in the form of RA that is autoantibody positive (or ‘seropositive’) for the markers antibodies to citrullinated protein antigens (ACPA) and rheumatoid factor (RF), although there is growing evidence that symptoms and imaging findings can also identify a pre-RA period in seronegative RA.

Importantly, features such as symptoms, autoantibodies and imaging findings can be used to predict which individuals (who can be termed ‘at-risk’) will develop clinical RA in the future. The ability to predict future RA has led to several clinical trials designed to prevent or delay future RA. In addition, individuals are seen in clinical rheumatologic care who have symptoms and/or autoantibody positivity without clear IA which presents a challenge to clinicians on how to manage these individuals.

This review will provide updates in the identification of individuals who are at-risk for future RA, prediction of future clinical RA, the results of the clinical trials of prevention as well as potential clinical management strategies in individuals who may be at-risk for future RA. This review will also discuss important challenges and opportunities in the area of prevention.

Who is at-risk for future RA?

There are a variety of ways to define who is ‘at-risk’ for future RA (Supplemental Box 1). Notably, one may consider that individuals who have clinically-apparent IA yet who do not fulfill established RA classification criteria[13, 14] to be in a pre-RA or at-risk state although those individuals may be better classified as undifferentiated IA - especially as treatment may be warranted for their existing IA. Furthermore, in some ways, anyone who does not already have a diagnosis of RA may be at-risk for future disease.

However, practically speaking, an individual who does not currently or historically have clinical IA may be considered ‘at-risk’ for future RA based on several definable features. These include symptoms that are suggestive of evolving RA that has been termed in some studies as ‘clinically-suspect arthralgia’[15] (and as discussed below, that may also meet more formal criteria). Autoantibody elevations also can define an ‘at-risk’ state; these are typically RF and ACPA[11, 12, 16, 17] (although some evidence suggests not all ACPA are equally predictive[18]) but may also include other autoantibody systems such as antibodies to carbamylated proteins (anti-CarP)[19], antibodies to peptidyl arginine deiminases (anti-PADs)[20] and antibodies to malondialdehyde-acetaldehyde (anti-MAA)[21]. There is also growing understanding that imaging with ultrasound or MRI can identify ‘subclinical’ inflammatory changes in the joints and/or peri-articular tendons that are indicative of risk for future clinical RA (indeed there is some evidence that tenosynovitis may be one of the first ‘joint’ related manifestations of RA)[11, 22–24]. Other features that are associated with increased risk for RA (yet probably less risk if present by themselves than arthralgia, autoantibody and imaging abnormalities) include family history of RA and/or the presence of genetic factors that may include belonging to a population who is at-risk for RA such as indigenous North Americans[6] or the presence of a specific genetic factor such as the shared epitope[25]. In addition, exposure to certain environmental risks (e.g. tobacco), factors including obesity and diet[26, 27], and emerging risks associated with mucosal processes such as periodontal disease[28], airways disease[29, 30] and potentially certain microbial factors[31] are associated with increased risk for RA. As discussed below, these features may be used individually or in combination to estimate risk of development of clinical RA.

How are individuals identified who are at-risk for future RA?

Rheumatologists in clinical practice encounter individuals who are ‘at-risk’ for RA because these individuals have been referred for evaluation of musculoskeletal symptoms, or perhaps referred from a specialist where RA related autoantibodies are identified because of work-up of another medical condition (e.g. ACPA testing because of lung disease). This is also a common pathway to identify individuals for enrollment in research studies of RA development.

There are also other methods used primarily in research to identify and prospectively study individuals who are at-risk for RA. These may be considered ‘population-based’ and include studies of populations who are at-risk for RA such as North American indigenous populations who have high prevalence rates of RA[5, 6], and studies of first-degree relatives of individuals with clinical RA[7, 9]. There are also studies which have identified individuals at-risk for future RA through approaches such as ACPA testing in health-fairs[16] or population surveys[32].

How to define the likelihood and timing of future development of RA?

A critical aspect of evaluating someone who may be at-risk for future RA is quantifying that risk – in other words, how much risk is someone at? Furthermore, quantifying that risk has several parts that include the determination of overall risk (i.e. will they ever get RA) as well as the expected timeline of development of RA (i.e. when will they get RA), and potentially what type of RA will develop. These aspects of prediction are important for both communicating risk to an individual and devising clinical follow-up plans. In addition, these are important in observational research studies and clinical trials where an accurate estimate of the likelihood and timing of the development of clinical RA is critical to study design.

In terms of building prediction models for future RA, several studies have used case-control or retrospective data and have found that elevations of ACPA and/or RF are associated with high positive predictive values for future RA; however, by nature of the design of these studies they likely overestimate risk. As such, for real-time prediction, better models are those that are developed in prospective studies, and that importantly have formally addressed the overall likelihood of developing RA as well as the rate of development of RA within certain time frames (Box 1).

Box 1.

Formal prediction models for future RA derived in prospective studies

Study	Design	Major findings
Van de Stadt et al 2013	Prospective evaluation of 374 individuals who presented to several Dutch rheumatology clinics with arthralgia and ACPA and/or RF positivity and without IA at baseline. 131 (35%) participants developed IA after a median of 12 months.	Nine features were used to develop a score; scores of 7–13 were associated with rates of development of IA of ~63% by 24 months. Features (points if present): first degree relative with RA (1); drink alcohol (1 if no); symptoms start <12 months prior (1); intermittent symptoms (1); symptoms in upper and lower extremities (1); pain scale >=50 on visual analog scale (2); morning stiffness >=1 hour (1); self-reported swelling in any joint (1); RF-IgM pos/ACPA neg (0); RF-IgM negative and ACPA positive <3x cut-off (2); RF-IgM negative and ACPA positive >=3x cut-off (3); RF-IgM and ACPA positive (4). Note: all features if present (or positive) indicate point(s) with the exception of alcohol intake which if negative = 1 point.
Van Steenbergen et al 2016	A EULAR task force determined a set of clinical characteristics that could be used to identify individuals who are at-risk for progression to RA.	A model was developed that included the following features: Joint symptoms of recent onset (duration of <1 year), symptoms located in MCP joints, duration of morning stiffness >=60 minutes, most severe symptoms present in the early morning, presence of a first-degree relative with RA, difficulty making a fist, positive squeeze test of MCP joints. When evaluated in a real-world prospective cohort, the presence of >=3 of 7 features had a sensitivity of 90.2% and a specificity of 74.4% for the future onset of clinical RA.
Duquenne et al 2023	Prospective evaluation at single center (Leeds, UK) of 455 ACPA(+) individuals who initially presented to primary care with musculoskeletal complaints; 148 (32.5%) participants developed IA after a median of 255 weeks of follow-up.	The investigators presented two models. The first is a ‘simple’ model that was recommended to be applied in primary care and used morning stiffness more than 30 minutes, levels of ACPA, positivity of RF and an elevated ESR; a certain score was associated with a positive PPV of ~27% of developing clinical RA within 1 year. The second ‘Comprehensive’ model that was recommended to be applied in rheumatologic practice included more variables as well as ultrasound imaging; a certain score was associated with an individual having a PPV of ~70% of developing IA within 5 years.

Abbreviations: ACPA=antibodies to citrullinated protein antigens; RA=rheumatoid arthritis; RF=rheumatoid factor; ESR=erythrocyte sedimentation rate; PPV=positive predictive value; MCP=metacarpal-phalangeal joint; Ig=immunoglobulin.

Of these, the most recent is a large study published in 2023 that used single center data (Leeds, United Kingdom) from individuals who presented with any musculoskeletal symptoms to primary care and who were subsequently found to be ACPA(+) but without IA at baseline based on physical examination[11]. In this study, two models were developed that included a ‘simple’ model of four features to develop a score that was associated with a PPV of ~27% for the development of clinical IA within 1 year and a ‘comprehensive’ model that included multiple clinical and biomarker features as well as ultrasound imaging where a certain score was associated with a PPV of ~28% for the development of clinical IA within 1 year, and a PPV of ~71% within 5 years. The authors conclude that these models could be used to guide referral patterns in primary care, or selection of individuals at highest risk for RA for preventive interventions, although they acknowledge further study and validation are needed.

Other studies have evaluated the role of symptoms and examination findings in predicting risk. In particular, as mentioned above, a EULAR-supported project identified a set of factors that could help define individuals which individuals who had musculoskeletal symptoms/arthralgia were most likely to develop RA[15]. In this study, the presence of >=3 features on history and examination has a sensitivity of 90.2% and a specificity of 74.4% for future clinical RA. As part of a validation study[33], this model was evaluated in separate cohort of where it was found that in evaluations in a rheumatology clinic, the presence of >=3 features had a PPV of ~30% for future IA; however, in contrast, when applied in a primary care setting the PPV was ~3%. In this validation study, the authors concluded that to improve the specificity of the model, clinical features needed to be combined with biomarkers.

Importantly, prediction of future RA has been estimated in prospective ‘population-based’ studies of those at-risk for future RA. These populations include North American indigenous populations and relatives of individuals with established RA[6, 17], and individuals identified with ACPA positivity through population based testing[16]. However, these studies have not developed formal prediction models like the the studies presented in Box 1 that evaluated individuals who initially presented to clinical care because of musculoskeletal symptoms. These two approaches (i.e. population based versus ‘clinical care’) may have differences in identifying risk for future clinical RA. For example, someone who presents to clinic because of joint symptoms may be at higher risk for RA (or be closer to onset of clinical RA) than someone who is asymptomatic and found to be ACPA positive through population-based studies. This is an area of active investigation in part because it will be important to understand the risk prediction for future RA from these scenarios especially if the field moves towards population-based screening for RA risk (perhaps similar to blood-based screening for hyperlipidemia as a risk factor for cardiovascular disease which is done currently).

What are results from clinical trials for prevention of RA?

The ability to predict future RA has led to the development of multiple clinical trials with the general goal to prevent or delay the first appearance clinical IA (i.e. the first swollen joint) RA in at-risk individuals, although imaging findings have also been used as outcomes in some trials. These trials are presented in more detail in Box 2[34–40].

Box 2.

Summary of published trials to prevent/delay incident clinical RA in at-risk populations^*

Study	Inclusion criteria	Study design and intervention	Primary outcome	Major findings
Bos et al 2010	RF and/or ACPA positive shared epitope positive; arthralgia	RCT; dexamethasone 100 mg IM x 2 doses vs placebo	Incident clinical IA	17/83 (21%) developed IA after a median follow-up of 26 months; no difference between arms (dexamethasone 21% vs placebo 20%); Dexamethasone use associated with decreased autoantibody levels
Gerlag et al 2019 (PRAIRI)	RF and ACPA positive; CRP >0.6 mg/L; arthralgia	RCT; rituximab 1000 mg x 1 dose (and steroid) vs placebo	Incident clinical IA	30/81 (37%) developed IA after a mean follow-up of 29 months; no significant difference in overall rates of IA between arms (rituximab 14/41 (34%), placebo 14/40 (40%)); rituximab associated with delay of onset of IA
van Boheemen et al 2021 (StapRA)	RF and ACPA positive or ACPA >3x ULN; arthralgia	RCT; atorvastatin 40 mg/day x 3 years vs placebo	Incident clinical IA	15/62 (24%) developed IA after a median follow-up of 14 months; no significant difference between arms (atorvastatin 9/31 (29%) vs placebo 6/31 (19%)
Krijbolder et al 2022 (TREAT EARLIER)	Arthralgia and MRI evidence of joint inflammation but without clinical swollen joint; RF/ACPA not required for inclusion although 33% of participants were RF and/or ACPA positive	RCT; methylprednisolone 120 mg x 1 dose and methotrexate up to 25 mg/week x 1 year vs placebo; 1-year post-drug follow-up	RA by 2010 criteria present at 2 time points 2 weeks apart	44/236 (~19%) developed RA over the 2 years of the trial; no significant differences between arms (methotrexate 23/119 (19%), placebo 21/117 (18%); decreased measures of physical function, pain and MRI inflammation in MTX treated group. The highest rate of RA development was within ACPA+ individuals (27/54=50%), although there was no significant difference in rates between arms at 2 years.
Hahn et al 2022 (VITAL)	Individuals without cancer or cardiovascular disease	RCT; omega-3 fatty acid (1000 mg/day) and/or vitamin D (2000 IU/day) vs placebo x 5 years	Incident autoimmune disease (including RA) by chart review	Vitamin D supplementation for five years, with or without omega-3 fatty acids, reduced aggregate incidence of autoimmune disease, including RA, by ~22%.
Rech et al 2024 (ARIAA)	ACPA positive and MRI evidence of joint inflammation	RCT; abatacept 125 mg SQ weekly x 6 months vs placebo; 12-month post-drug follow-up	MRI inflammatory parameter improvement; clinical RA	At 18 months there was significant MRI improvement in abatacept arm compared to placebo (28/49 (57%) vs 14/49 (29%)). At 18 months there was also significantly less progression to clinical RA in abatacept arm compared to placebo (17/49 (35%) vs 28/49 (57%)).
Deane et al 2022 (StopRA)**	ACPA >=2x ULN	RCT; hydroxychloroquine 200–400 mg/day x 1 year versus placebo; 2 years post-drug follow-up	RA by 2010 criteria	In preliminary analyses, 43/144 (~30%) developed RA over the 3 years of the study; no significant differences between arms (24/71 (34%), placebo 26/73 (36%); trial halted and final analyses pending
Cope et al 2024 (APIPPRA)	ACPA+RF positive or ACPA >=3x normal and arthralgia	RCT; abatacept 125 mg weekly injection x 1 year versus placebo; 1-year post-drug follow-up	RA by 2010 criteria	At 2 years, 65/213 (~31%) of participants developed RA by 2010 criteria; 25% in abatacept arm and 37% in placebo arm. This resulted in differences in mean arthritis-free survival time between arms of ~99 days (p=0.002). In exploratory analyses, the highest risk for progression was individuals with multiple positive autoantibodies; this group also had the best response to abatacept.

^*While the inclusion criteria varied across studies, across the studies the participants could not have clinical RA at baseline defined as the presence on physical examination of a swollen joint consistent with synovitis.

^**The results from StopRAhave only been presented in abstract form.

Abbreviations: RA=rheumatoid arthritis; ACPA=antibodies to citrullinated protein antigens; RF=rheumatoid factor; RCT=randomized controlled trial; ULN=upper limit of normal; SQ=subcutaneous; MRI magnetic resonance imaging; MTX=methotrexate; IM=intramuscular; IA=inflammatory arthritis; IU=international units

At a high level, key take-aways from these studies have been that the short-term use of the agents used studies including corticosteroids, rituximab, atorvastatin, methotrexate and hydroxychloroquine have not significantly prevented clinical IA/RA. However, rituximab delayed the onset of RA[35] andmethotrexate improved function by measures such as HAQ and may be associated with less severe RA[37]. Notably, Two studies (ARIAA and APIPPRA) demonstrated that abatacept improved rates of clinical RA development during the trial period[39, 40], and in one of the studies (ARIAA), abatacept was associated with improved MRI findings of inflammation. Additionally, in the APIPPRA study, in exploratory analyses, a subgroup that appeared to respond best to abatacept were those with positivity for multiple autoantibodies including ACPA, RF, anti-CarP. Furthermore, while not primarily a study of RA prevention, in a prespecified analyses performed in the VITAL study which was designed to evaluate the efficacy of vitamin D and omega-3 fatty acids on cancer and cardiovascular disease prevention[41], vitamin D supplementation for 5 years (2000 international units daily), with or without omega-3 fatty acids (1000 mg daily), led to a 22% decrease in incident autoimmune disease, including RA.

How should a clinician manage an individual who is at-risk for future clinical RA?

Despite the completion of these trials, to date, there are no pharmacologic agents that are yet approved by regulatory agencies (i.e. Food and Drug Administration [FDA]) for an indication to prevent the future onset of RA in at-risk individuals. However, individuals who are in an ‘at-risk’ state for RA (that may include musculoskeletal symptoms/arthralgia) are being seen regularly in rheumatology practices. This leads to a challenge to clinical providers on how to manage and monitor these individuals.

While there are no formal guidelines established for the management of these individuals, some general recommendations are presented in Box 3. These include counseling on personal risk, education about signs and symptoms of RA, and lifestyle factors such as exercise, ideal body weight, Mediterranean diet and avoidance of tobacco[42, 43]. Importantly, while these measures are not formally studied in randomized controlled trials, they may be globally beneficial to health. These individuals may also be followed at regular time intervals to directly assess if IA has developed, and to update them on developments in care.

Box 3.

Potential approaches for the management of individuals who are at-risk for future RA

• Determine if clinical RA is present (may include imaging, with careful interpretation of findings). • If autoantibodies are present, consider evaluation for other conditions that may be related to autoantibody positivity. For example, given some association of ACPA with lung disease, consider symptom driven evaluation for lung disease; if RF positive, consider additional studies to ensure that autoantibody is not related to another condition (e.g. hepatitis C, hematologic malignancy). • Discussion of rates of progression to clinical RA; in general, studies suggest that positivity for ACPA has ~20–30% positive predictive value for the onset of clinical RA within 3–5 years although actual risk may be higher or lower depending on many factors. May apply formal prediction model (e.g. Duquenne et al 2023). • Education on the signs and symptoms of RA; consider routine follow-up to monitor joints and update education as needed. • Consideration of lifestyle interventions that may have broad health benefits and potentially reduce risk for RA (smoking cessation, exercise and optimal body weight, oral health). These interventions have not been proven to reduce RA risk in controlled trials. • Initiate approved treatment(s) if clinical RA develops. • Monitoring for advances in prevention that will impact clinical care.

Abbreviations: RA=rheumatoid arthritis; ACPA=antibody to citrullinated protein antigen; RF=rheumatoid factor.

Notably, while the results of the VITAL study (Box 2) are encouraging, it is not clear if vitamin D supplementation (with or without omega-3 fatty acids) is beneficial to reducing progression to RA, but this is something that clinicians and individuals at-risk for future RA may discuss. Furthermore, while it is emerging that mucosal factors such as periodontal disease[28], lung disease (and in particular airways inflammation)[44], or microbial factors[31, 45, 46] may be important in pre-RA, it is not clear if interventions in these areas could lead to prevention. If possible, the individual could be referred to a research study, although these are not available in all areas.

The role of imaging is challenging in the clinical management of an individual who is determined to be ‘at-risk’ for RA. Specifically, in clinical practice, imaging is often utilized to help identify if inflammation is present if a joint examination does not reveal clear joint swelling. Importantly, as mentioned above, imaging may identify a ‘subclinical’ stage[47] of RA development (Figure 1). Furthermore, two of the prevention trials presented in Box 2 (TREAT EARLIER and ARIAA) used MRI findings of joint inflammation as inclusion criteria, and the prediction model by Duquenne and colleagues[11] demonstrates that ultrasound findings can improve prediction. However, while a survey demonstrated that ~70% of rheumatologists may be willing to start disease modifying therapy in ACPA positive individuals who have imaging evidence of IA (yet no clinical IA)[48], there is still not clear broadly-accepted understanding of how a diagnosis of IA by imaging in absence of IA on clinical examination affects long-term outcomes[24, 49]. Indeed, in additional analyses of the cohort evaluated by Duquenne et al[11] demonstrated that subclinical synovitis seen on ultrasound may resolve over time in a subset of ACPA(+) individuals in absence of immunomodulatory therapy[50]. As such, at this time, imaging should be used cautiously in diagnosis of IA and in driving decisions for treatment.

Another challenge is the management of symptoms in individuals who are in an ‘at-risk’ stage which is some cases may be significant and associated with decreased function[51, 52]. If an autoantibody is present – and as above, if imaging shows inflammation - there may be a tendency to start a disease modifying agent. However, in absence of clear guidelines, disease modifying treatments should be initiated cautiously in at-risk individuals. In particular, while the full results of the StopRA trial (Box 2) are pending, the lack of efficacy for HCQ to prevent or delay the future onset of RA may limit its use in clinical care of at-risk individuals. Notably, palindromic rheumatism may be considered a form of ‘pre-RA’ and it also represents a challenge to clinicians for treatment[53]; additional data is needed in this condition to understand optimal treatments for active symptoms, as well as potentially to prevent the future development of clinical RA.

What are the next steps to advance prediction and prevention?

A list of factors that are important to advancing prediction and prevention in RA is presented in Supplemental Box 2. Of these, of top priority is gaining a better understanding of the natural history and pathophysiology of RA development because that will directly help improve prediction models and identify meaningful targets for preventive interventions. Of particular interest is the growing understanding that pre-RA may include a non-articular stage (Figure 1) where mucosal inflammation (including microbial effects), or other non-articular processes, may be driving the initial breaks in tolerance and propagation of autoimmunity, in part by leading to failures of immune ‘checkpoints’ to resolve inflammation and autoimmunity[46]. Addressing these non-articular processes may require use of interventions that are different than the ‘standard’ agents used to treat clinical RA. Notably, there is also a growing understanding that individuals with clinical RA have different ‘endotypes’ of disease that may require personalized approach. It is highly likely that these ‘endotypes’ are also important in the at-risk stage to define an individual’s risk, as well as to identify what their ‘personalized’ intervention may need to be to prevent progression. For example, as is emerging from the APIPPRA study (Box 3), individuals with multiple autoantibodies may represent a certain endotype that responds best to a certain intervention. As such, understanding individual ‘endotypes’ will likely be critical in developing personalized prediction and prevention in RA. While this may seem daunting, an important and encouraging model for RA (and other autoimmune diseases) is the advances that have occurred in type 1 diabetes, where autoantibodies and genetic factors are highly predictive for future clinical onset of diabetes, and now the anti-CD3 monoclonal antibody teplizumab has FDA approval for use in pre-clinical diabetes states[54].

Conclusion

Prediction and prevention are advancing in RA. With additional study, the field may soon be able to change the paradigm for RA managment to one where we seek out individuals who are at-risk for RA through screening programs, appropriately understand their risk for future RA and their current biologic status in order to institute ‘personalized’ interventions that can reduce their risk for future RA – as well as potentially improve their symptoms now.

Key Points.

• The natural history of rheumatoid arthritis includes a ‘pre-rheumatoid arthritis (pre-RA)’ stage that may include a non-articular stage where mucosal or other processes drive the initiation and early propagation of autoimmunity.

• Individuals who are at risk for future clinical RA can be identified using factors including family history and racial/ethnic background, musculoskeletal symptoms, autoantibody elevations and imaging; the likelihood and timing of the development of future clinical RA can be quantified using models that include familial and genetic factors, musculoskeletal symptoms, autoantibodies (including antibodies to citrullinated protein antigens [ACPA]) and imaging.

• Several clinical trials designed to prevent the onset of clinical RA have been performed and agents including corticosteroids alone, rituximab, atorvastatin, methotrexate and hydroxychloroquine did not significantly reduce progression to clinical RA; however, rituximab delayed the onset of clinical RA, and methotrexate improved function, symptoms and joint inflammation on imaging; in addition, vitamin D supplementation (with or without omega-3 fatty acids) reduced rates of incident autoimmune disease, including RA, and data from two studies in ACPA(+) individuals demonstrates that abatacept significantly reduces the onset of clinical RA, and reduced joint inflammation on imaging.

• There are currently no approved pharmacologic interventions to prevent future clinical rheumatoid arthritis; however, individuals who are at-risk for future RA may be counseled on risk, followed for evolution of disease, and while not robustly proven in controlled clinical trials, interventions such as exercise, maintenance of healthy body weight, Mediterranean diet and smoking cessation may reduce progression to rheumatoid arthritis and are associated with general health benefits.

• Future studies should include deeper evaluation of the pathophysiology of RA development to improve prediction and identify key pathways to target in future clinical trials, as well as develop infrastructure to support prevention-related research.