Therapeutic strategies in preclinical stages of rheumatoid arthritis

Introduction

Rheumatoid arthritis (RA), the most common rheumatic disease, significantly impacts patients’ quality of life due to a loss of joint function from chronic progressive erosive joint disease, and also affects morbidity and mortality.¹

While modern pharmacologic therapies can greatly benefit individual RA patients, they also contribute to rising treatment costs. Considering both the decreased quality of life and increased medical expenses, RA imposes a substantial socioeconomic burden. This holds true even in light of diminishing indirect costs, such as fewer sick leaves, since the total treatment costs have increased. The direct costs of RA treatment in Germany between 2002 and 2011 increased from 4914 EUR to 8206 EUR per patient aged between 18 and 64 years, particularly due to rising expenses for pharmacotherapy.²

Consequently, initiating therapy as early as possible might be beneficial. Therefore, the concept of the “window of opportunity,” which refers to a period in the very early stage of RA when pharmacologic therapy has the greatest beneficial impact, is highly relevant for reducing the individual and socioeconomic burden of the disease. Early treatment could modify or even prevent occurrence and progression of RA.³

In addition to this strategy, the treat-to-target principle yields superior outcomes, as compared with standard RA therapy.^4,5 This approach can lead to long-term remission in patients with RA, potentially allowing for dose reduction or discontinuation of treatment during remission periods.⁶ Unfortunately, prospective real-world data indicate that only a fraction of patients achieve the treat-to-target therapy goals of sustainable remission or low disease activity.⁷

Therefore, the concept of primary prevention, involving treating the disease before it fully manifests, is a promising approach. Pharmacologic therapy in the preclinical phase of RA or for patients at high risk is intended to impede the onset and progression of the disease in the long term. In this review, we summarize the concept of primary prevention, discuss relevant trials, including employed therapeutic strategies and their findings, and offer a critical appraisal.

Definition

Long before the clinical manifestation of RA, characterized by palpable painful joint effusion, elevated serum autoantibody levels can be detected in both asymptomatic individuals and those reporting arthralgia and polyarthralgia.⁸ The phase of serologic activity (seropositivity for anti–citrullinated protein antibody [ACPA] and / or rheumatoid factor [RF]) that may be accompanied by arthralgia or polyarthralgia but without palpable joint effusion is called pre-RA or at-risk RA. On the other hand, clinically suspect arthralgia (CSA) refers to a typical anamnesis and certain clinical pattern of arthralgia or polyarthralgia independently of RF and ACPA positivity. These terms are often used interchangeably to describe a precursor of RA, but they do differ. From a semantic perspective, the term at-risk RA appears more appropriate than pre-RA as it does not imply a certain progression to the disease.

Pathogenesis and disease progression

The precise pathogenesis of RA is still unknown. However, our understanding of RA etiology and pathomechanism is growing.¹ Recently, the European Alliance of Associations for Rheumatology (EULAR) Study Group for Risk Factors for Rheumatoid Arthritis named 5 groups of predominant risk factors for RA that may be seen as phases of pre-RA: 1) genetic factors, 2) environmental factors, 3) associated systemic autoimmunity, 4) symptoms without signs of clinical arthritis, and 5) unclassified arthritis. Retrospective studies including patients with RA could adopt this stage-based model of preclinical RA and indicate development phases before the actual clinical occurrence of the disease.⁹ It is important to note that these phases do not occur in all patients who progress to RA nor do they occur in the same order in all patients.⁹ An overview on the hypothesized pathogenesis of RA is provided in Figure 1.

Seropositivity for ACPA and / or RF has the highest positive predictive value for RA. The predictive value is up to 80% over 5 years, modulated particularly by the presence of high-titer ACPA or the presence of both ACPA and RF. However, autoantibodies such as serum RF immunoglobulin M or ACPA can be detected in affected individuals up to 5 years before the onset of clinical manifestation.¹⁰ The predictive value is far from perfect: even in individuals with arthralgia and high titers of RF and ACPA (highest-risk group), approximately 60% did not develop arthritis after 1 year and about 20% after 5 years.¹¹

The formation of ACPA occurs due to citrullination of endogenous proteins. Environmental factors play a propagating role in the development of autoimmunity, interacting with genetic factors such as the “shared epitope.” Significant environmental risk factors promoting citrullination include continued nicotine abuse and microorganisms triggering local mucosal lesions related to inflammation.¹² Interestingly, the generation of ACPA does not occur in the joint itself, but is speculated to take place in the mucosa of the lungs, oral cavity, or intestines.^13,14 Studies in humans and in animal models have shown that, for example, Porphyromonas gingivalis and Aggregatibacter actinomycetemcomitans initiate local inflammatory processes in the oral cavity resulting in protein citrullination.^15,16 Similarly, patients with pre-RA have been found to have a higher incidence of upper respiratory tract infections than the general population.¹⁷

The evidence for the microbiome as a potential pathogenetic cause is limited, with differences observed between the microbiome of the general population and treatment-naive pre-RA patients. In particular, Prevotella copri was significantly more abundant in the pre-RA group.^18,19 At present, due to limited data, causality of dysbiosis in the development of RA cannot be conclusively established.

Progression probability is not influenced by serologic findings only. It can also be defined by patient clinical characteristics. An EULAR task force defined a phenotype of CSA that can be identified based on the clinical history, and that has a 90% sensitivity and 74% specificity for the future diagnosis of RA.^20,21 CSA is defined as presence of at least 3 of the following anamnestic and / or clinical criteria: recent onset of symptoms, metacarpophalangeal (MCP) joint pain / symptoms, morning stiffness lasting 60 minuntes or longer, early morning severity of symptoms, first-degree relative with RA, incomplete / painful fist closure, and positive squeeze test of MCP joints.

A combination of serologic and clinical findings is associated with improved predictive value. Individuals seropositive for ACPA with new-onset arthralgia have a 30% probability of developing RA within the next year.²²

Imaging studies may also be helpful in the identification of individuals at a greater risk for progression into RA, but most likely cannot be used as a criterion alone. It was found that even healthy individuals present inflammatory lesions in the joints on imaging without clinical relevance.²³ Even “true” subclinical joint inflammation only progresses to clinical arthritis and RA in 30% of cases when considered alone. Therefore, vigilant evaluation is required because assessing radiologic signs without considering clinical findings might lead to overinterpretation and misdiagnosis.^24,25 For instance, in a study that included patients who fulfilled the EULAR definition of CSA, subclinical inflammation was found on magnetic resonance imaging (MRI) in 51% of tender and 39% of nontender joints.²⁵ Imaging is best used in combination with other risk factors in at-risk or CSA cohorts. ACPA-positive probands with ultrasound-detected pathologic synovial hypervascularization had an increased risk of RA even if joint swelling was absent on clinical examination.²⁶ Moreover, it has been determined that MRI aids in predicting progression in CSA. Tenosynovitis, particularly MCP-extensor peritendinitis, appears to be one of the first inflammatory manifestations of RA. Incorporating this feature and a number of locations with subclinical inflammation improved prediction of the disease progression with positive predictive values up to 63%–67%.²⁷

It is very likely that resilience factors might exist, since only a certain percentage of ACPA-positive patients develop RA over time. A study on indigenous ACPA-positive patients with RA and their first-degree relatives has shown a heterogenic distribution. Whereas some ACPA-positive relatives developed RA, the disease never occurred in others, even with persistent ACPA-positivity. Some even lost their seropositivity over time.²⁸ Interestingly, RA can emerge in patients who have lost their seropositivity and vice versa—RA can develop in seronegative patients, who then become seropositive over time. Therefore, it should be emphasized that autoantibodies alone do not promote the development of RA, but are part of a multifactorial pathogenesis. The same applies to clinical and imaging findings: only a certain proportion of patients with undifferentiated arthritis and MRI-proven inflammation in hands progress to clinical RA.²⁹

The various preclinical forms of RA have different rates of progression to a clinically manifested disease. An overview is presented in Figure 2. Pre-RA and at-risk RA have the highest rate of progression, with 30%–60% of patients developing clinically detectable arthritis within 1 year and 80% within 5 years. In the cases of RF and / or ACPA positivity, the progression rate is up to 70% within 5 years. The progression rate for CSA is 30% to develop arthritis within 2 years and 22% to fulfill the classification criteria of RA. In the presence of subclinical inflammation on imaging, the progression rate is 30% within 1 year. Undifferentiated arthritis has the lowest rate of progression to fulfill the RA classification criteria at 18%–27% within 5 years^{10,11,20,30,31}

Preventive interventions in preclinical stages of rheumatoid arthritis

Over the past decade, several preventive therapeutic studies have been initiated in the preclinical stages of RA. Their results and therapeutic implications are summarized hereafter (Table 1). So far, statins, dexamethasone, conventional synthetic disease-modifying antirheumatic drugs (csDMARDs), biologic DMARDs (bDMARDs), and lifestyle changes have been investigated.

Table 1. Overview of published intervention studies on the prevention of rheumatoid arthritis development

Study	Intervention	Inclusion criteria	Intervention	Primary end point	Results
STAPRA34 (Dutch Trial Register: NTR5265 / Trial NL5036)	Atorvastatin	•Arthralgia •High-titer ACPA (>3 × ULN) or RF and ACPA positivity	Atorvastatin 40 mg/d vs placebo over 3 years	Progression to RA: ≥1 effused joint	Study prematurely terminated: most likely no effect
Effect of dexamethasone on autoantibody levels and arthritis development in patients with arthralgia: a randomized trial35	Dexamethasone	•Arthralgia •ACPA or RF positivity	Dexamethasone, 100 mg, intramuscularly, initially and after 6 weeks vs placebo	Autoantibody reduction by 50% or normalization after 6 months	Primary end point reached by 1 participant. No difference in RA development between both groups
SAVE43	Methylprednisolone	•Inflammatory arthritis of ≥1 joint •Symptom duration <⁠16 weeks	Methylprednisolone, 120 mg, intramuscularly, once vs placebo injection	Clinical remission at week 12 and 52 without further glucocorticoids or DMARD treatment	Neither remission nor development of RA is delayed by glucocorticoid treatment
STIVEA44	Methylprednisolone	Inflammatory poly-arthritis •Symptom duration of 4–10 weeks	Methylprednisolone, 80 mg, intramuscularly, once weekly for 3 weeks vs placebo	Initiation of DMARD therapy required at the 6-month assessment	Prevention of RA progression and start of DMARD therapy observed in 1 of 10 patients
StopRA37 (NCT02603146)	Hydroxychloroquine	•ACPA >40 U/l •No clinical joint effusion or synovitis	Hydroxychloroquine, 200–400 mg/d, weight-adapted vs placebo over 1 year	Occurrence of RA (ACR/EULAR criteria) or arthritis with ≥1 bone erosion on X-ray	No difference between intervention and placebo with progression to RA
PROMPT45 (Dutch Trial Register: NTR73 / Trial NL44)	Methotrexate	•Classification of probable RA •DMARD-naive •Symptom duration <⁠2 years	Methotrexate, 15 mg/week (increased by 5 mg/week every 3 months if DAS >2.4) vs placebo over 1 year	Progression of RA after 5 years	No difference in RA progression but delay of progression to RA in ACPA-positive patients
TREAT EARLIER38 (Dutch Trial Register: NTR4853 / Trial NL4599)	Methotrexate	•Arthralgia of hands or feet •Unilateral MRI-detected joint inflammation in hands or feet	Methotrexate, 25 mg/week, orally vs placebo	Occurrence of RA (EULAR 2010 criteria) or clinical arthritis (SJC-66) in ≥2 joints within a period of ≥2 weeks during a follow-up of 2 years	No difference in the intervention or placebo group considering RA occurrence. Improvement in MRI-detected inflammation, and related symptoms in intervention group
PRAIRI39 (Dutch Trial Register: NTR1969)	Rituximab	•ACPA and RF positivity •Elevated CRP (>0.6 mg/l) or subclinical MRI- / ultrasound-detected synovitis	Rituximab, 1000 mg intravenously, single shot vs placebo (+ initial shot of methyl-prednisolone 100 mg intravenously in both groups)	Occurrence of RA in 29-month follow-up	No difference in the intervention or placebo group considering RA occurrence, but mean delay of 12 months in the intervention group
APIPPRA40 (ISRCTN 46017566)	Abatacept	•Arthralgia •ACPA >3 × ULN •ACPA and RF positivity	Abatacept, 125 mg/week subcutaneously vs placebo over 12 months	Occurrence of RA (ACR/EULAR criteria) or detection of joint effusion in ≥3 joints within 24 months	Significant reduction in the development of RA between intervention and placebo group after 12 and 24 months
ARIAA41 (NCT-02778906)	Abatacept	•ACPA positivity •MRI-detected synovitis / tenosynovitis / osteitis	Abatacept 125 mg/week subcutaneously vs placebo over 12 months	Improvement of MRI detected inflammation markers according to RAMRIS score within 24 months	Significant improvement of inflammation markers on MRI and development of arthritis in the intervention group
Abbreviations: ACR/EULAR, American College of Rheumatology / European League Against Rheumatism; CRP, C-reactive protein; DAS, disease activity score; DMARD, disease-modifying antirheumatic drug; RA, rheumatoid arthritis; RAMRIS, Rheumatoid Arthritis Magnetic Resonance Imaging Score; SJC, swollen joint count; ULN, upper limit of normal; others, see Figure 2

Statins

In 2004, a randomized, double-blind TARA trial (Trial of Atorvastatin in Rheumatoid Arthritis) showed that an additive statin therapy in patients with RA on DMARD therapy had positive effects on the disease activity (as measured by Disease Activity Score-28 [DAS28]).³² Moreover, a large-population study from Israel showed a correlation between statin therapy and reduced risk of RA.³³ Following these findings, the double-blind, multicenter, placebo-controlled STAPRA trial (Statins to Prevent Rheumatoid Arthritis) evaluated the effect of atorvastatin on RA in patients with high-titer ACPA or with ACPA and RF over 3 years (1 year of treatment and 2 years of follow-up without treatment). The primary end point was progression to RA defined as a minimum of 1 swollen and painful joint (DAS44). The study was prematurely terminated due to lower than expected enrollment of only 62 of originally planned 220 randomized patients. In the follow-up period, 15 individuals (24%) developed arthritis: 9 (29%) in the atorvastatin group; 6 (19%) in the placebo group, showing no significant difference (HR, 1.4; 95% CI, 0.5–3.95 vs HR, 1.19; 95% CI, 0.38–3.76).³⁴ In this small set of high-risk individuals, no protective effect of atorvastatin on RA development could be demonstrated. However, the statistical analysis is at best inconclusive because of the mentioned low enrollment rate.

Dexamethasone

In 2009, a Dutch multicenter, randomized, placebo-controlled trial demonstrated that 2 separate intramuscular 100-mg injections of dexamethasone 6 weeks apart did not reduce autoantibody levels at week 24 in 83 patients with arthralgia and ACPA or RF positivity. In a mean follow-up of 26 months, both the placebo and the control groups showed a nearly equal RA incidence (20% vs 21%, respectively).³⁵

Conventional synthetic disease-modifying antirheumatic drugs

The StopRA (Strategy for the Prevention of Onset of Clinically Apparent RA), a randomized, double-blind, placebo-controlled, multicenter study from the United States evaluated the effect of a 12-month, body weight–adapted therapy with hydroxychloroquine in patients with an elevated ACPA level greater than 40 U/l alone and the absence of clinical joint effusion or synovitis over 3 years after the initiation of treatment. Hydroxychloroquine is known for its anti-inflammatory, immunomodulatory, and antiproliferative effects, and nowadays it is investigated even outside rheumatology.³⁶ First results (interim analysis) were presented at the 2022 American Council of Rheumatology meeting.³⁷ Between April 2016 and November 2022, 144 patients were randomized with a total of 142 included in the analysis. The 3-year follow-up was calculated based on the available results (41 patients had developed RA). The study was subsequently terminated as approximately 34% of patients in the hydroxychloroquine arm and 36% in the placebo arm had developed RA at month 36 (P = 0.844).

Another recently published Dutch multicenter, randomized, placebo-controlled TREAT EARLIER trial (Treat Early Arthralgia to Reverse or Limit Impending Exacerbation to Rheumatoid Arthritis) analyzed the effect of a 1-year course of oral methotrexate (≤25 mg/week) plus a single intramuscular glucocorticoid injection (120 mg of prednisolone) in patients with at-risk RA (arthralgia and signs of subclinical arthritis on MRI imaging) on the development of RA and disease activity over a 2-year follow-up. The primary end point was the occurrence of RA (fulfilling the EULAR 2010 rheumatoid arthritis classification criteria or involving 2 or more swollen and tender joints) persisting for at least 2 weeks. The frequency of the primary end point was similar between the groups: 23 of 119 participants (19%) in the treatment group and 21 of 117 (18%) in the placebo group (HR, 0.81; 95% CI, 0.45–1.48). Secondary end points showed improved physical functioning in the treatment group during the first 4 months. Reduced pain, morning stiffness of joints, and MRI-detected joint inflammation showed sustained improvement in the treatment group, as compared with the placebo group. In conclusion, methotrexate plus a single glucocorticoid injection did not prevent the development of clinical arthritis but modified the disease course.³⁸

Biologic conventional synthetic disease-modifying antirheumatic drugs

Often, RA is described as the prime example of a chronic autoimmune disease. Therefore, several trials evaluated the effect of a B-cell and T-cell modulating therapy to prevent RA.

The randomized, double-blind, placebo-controlled, multicenter PRAIRI trial (Prevention of Clinically Manifest Rheumatoid Arthritis by B-cell Directed Therapy in the Earliest Phase of the Disease) investigated the impact of a single-dose 1000-mg rituximab injection on the disease progression of at-risk RA probands. The study included 81 probands that did not present any clinical signs of arthritis but had arthralgia, positive ACPA and RF titers, and an elevated C-reactive protein level (>0.6 mg/l) or synovitis detected on MRI or ultrasound. Both the rituximab group and the control group additionally received an intravenous 100-mg dose of methylprednisolone. The onset of clinical arthritis was the primary end point, and no difference in the occurrence of that end point was noted between the groups during the follow-up of 29 months (34% in the treatment vs 40% in the placebo group). Nevertheless, 75% of patients in the active treatment group progressed to RA 12 months later, as compared with the controls (P <⁠0.0001). Overall, there were more adverse events in the active treatment group, but these required only symptomatic treatment.³⁹

The British and Dutch APIPPRA study (Arthritis Prevention in the Pre-Clinical Phase of RA with Abatacept), whose findings were published recently in 2024, recruited patients with reported arthralgia and increased ACPA levels defined as at least 3-fold upper limit of normal or elevated ACPA and RF. The active treatment group received weekly injections of abatacept for 1 year, and placebo injections were administered in the control group. Follow-up lasted 1 year. The primary end point was the development of at least 3 swollen joints or fulfilment of the 2010 American College of Rheumatology / EULAR classification criteria for RA using swollen but not tender joints. Patients in the intervention group progressed to RA significantly less often than those in the placebo group after 12 (6% vs 29%, respectively) and 24 months (25% vs 37%, respectively; P = 0.044).⁴⁰

Similarly, a German multicenter, randomized, double-blind, placebo-controlled trial (ARIAA: Abatacept Reversing Subclinical Inflammation as Measured by MRI in ACPA Positive Arthralgia) investigated abatacept therapy in at-risk RA probands. The study included 98 of 139 recruited autoantibody-positive probands with MRI-detected joint inflammation without swollen joints. The treatment group received a weekly subcutaneous injection of 125 mg of abatacept for 6 months with a follow-up of 1 year. The primary end point was defined as an improvement in at least 1 of the MRI inflammation parameters (any change from baseline >0 assessing synovitis, tenosynovitis, and osteitis) according to the Rheumatoid Arthritis Magnetic Resonance Imaging Score. A total of 57% of patients from the active treatment group, as compared with 31% of those from the placebo group, reached the primary end point. Progression to RA was observed in 17 patients from the placebo group and 4 patients from the active treatment group (P = 0.0016) after 6 months. The difference between the 2 groups in terms of detectable inflammation on MRI and the progression of RA at a comparable rate of adverse effects was still present even after the 12-month follow-up following the 6-month intervention (P = 0.018).⁴¹

Lifestyle changes

In addition to trials investigating therapeutic interventions, there have been several observational trials on the impact of lifestyle characteristics on RA prevention. The strongest evidence for a modifiable lifestyle risk factor or behavior associated with a reduction of progression to RA is available for cessation of continued nicotine abuse in ACPA-positive individuals, followed by weight loss in the presence of obesity.⁴²

An overall healthy lifestyle consisting of high consumption of omega-3 fatty acids, reduced alcohol intake, reduced caffeine intake, and fewer sugar-sweetened soft drinks was associated with a lower incidence of RA. The role of physical activity in developing RA remains unclear, but high activity levels may reduce the risk for RA.⁴²

Undifferentiated arthritis

Various studies were performed 10 to 15 years ago in a population of patients with undifferentiated arthritis (patients with clinically swollen joint(s) who did not meet the classification criteria for RA). Primary end points of these studies included progression to RA or remission without the need for csDMARDs or glucocorticoids over time.

The SAVE (Stop Arthritis Very Early) trial was a randomized, double-blind, placebo-controlled, international multicenter study that investigated the effectiveness of a single intramuscular injection of methylprednisolone in a dose of 120 mg in reducing RA without further glucocorticoid or DMARD therapy at week 12 and 52. Patients with inflammatory arthritis in at least 1 joint and symptom duration shorter than 16 weeks were included. No difference in clinical remission was noted between the 2 groups.⁴³

The STIVEA trial was a randomized, double-blind, placebo-controlled, multicenter study showing that a 3-week course of intramuscular glucocorticoid injections (methylprednisolone, 80 mg weekly) in patients with very early inflammatory polyarthritis could prevent the progression to RA requiring DMARD therapy in 1 out of 10 patients after 12 months (adjusted odds ratio, 0.42; 95% CI, 0.18–0.99; P = 0.048).⁴⁴

The PROMPT (Probable Rheumatoid Arthritis: Methotrexate vs Placebo Treatment) study was a randomized, double-blind, placebo-controlled multicenter trial evaluating the effect of a 1-year methotrexate treatment on progression to RA in patients with unclassified arthralgia over a period of 5 years. Patients with unclassified arthralgia were randomized to receive either oral methotrexate in a dose of 15 mg/week or placebo. No DMARDs or steroids were allowed during treatment. Every 3 months, methotrexate dose was increased by 5 mg to a maximum of 30 mg/week as long as DAS28 was greater than 2.4. After 12 months, methotrexate was tapered and discontinued in patients who had not developed RA. After 5 years there was no lasting benefit of a 1-year course of methotrexate as compared with placebo in patients with unclassified arthritis. Progression to classifiable RA was not suppressed (25 patients in the methotrexate group and 29 in the placebo group progressed to RA; P = 0.45), drug-free remission was equally induced, and the progression of radiologic damage was similar in both groups.^45,46

None of these trials reached the primary end point. Early drug intervention with glucocorticoids or methotrexate for a limited (short) time in the phase of undifferentiated arthritis did not result in a difference with regard to progression to RA or csDMARD- or glucocorticoid-free clinical remission when comparing the intervention groups and the placebo groups.

Overview of various intervention studies on different stages of rheumatoid arthritis pathogenesis The intervention studies discussed in this review focused on different stages of RA progression. Figure 3 presents an overview of these divisions. As mentioned, the pathogenesis of RA begins with a phase of predominant autoimmunity and presence of autoantibodies in the absence of clinical symptoms (assessed in the STAPRA study³⁴). Next phase is characterized by arthralgia and subclinical synovitis. This is divided into CSA in the presence of arthralgia without the evidence of synovitis (the focus of the StopRA³⁷ and TREAT EARLIER³⁸ studies), at-risk RA in the presence of arthralgia and subclinical inflammation on imaging (assessed in the APIPPRA⁴⁰ and ARIAA⁴¹ studies), and pre-RA in the presence of autoantibodies, arthralgia, and subclinical evidence of inflammation (the PRAIRI study³⁹). The presence of clinical synovitis without meeting the classification criteria for RA is referred to as undifferentiated arthritis (the SAVE⁴³, STIVEA⁴⁴, and PROMPT⁴⁵ studies), which may progress to RA.

Conclusions

The concept of preventing RA with at best a single shot of medication or with a medication administered over a short period of time is very appealing in many ways. Not only could we prevent individual suffering due to RA but we could help lessen the socioeconomic burden RA places on health care systems. Therefore, several trials investigated pharmacologic treatment in preclinical stages of RA. These trials included healthy individuals seropositive for RF/ACPA fulfilling the criteria for at-risk RA and CSA as well as patients with (yet) unclassifiable arthritis. Thus far, statins, glucocorticoid, csDMARDs, and bDMARDs have been investigated.

As tempting as the concept might be, these trials do not warrant a paradigm shift from the treatment of a clinically apparent RA to early prevention of this autoimmune disease. All investigated preventive interventions with atorvastatin, glucocorticoids, csDMARDs (hydroxychloroquine, methotrexate), and rituximab did not prevent RA in the long term, although there was some benefit noted in patients receiving active treatment instead of placebo.

The most promising results were obtained in the 2 preventive studies, ARIAA and APIPPRA, using T-cell modulation with abatacept, demonstrating a significant reduction in RA development between the intervention and placebo groups, which persisted over time for up to 24 months.

Furthermore, it is important to note that not all patients with a high-risk profile develop RA. Overtreatment should be prevented at all costs, as pharmacologic therapies may be associated with serious adverse events. A strict risk-benefit analysis is warranted before trial results are generalized and put into clinical practice. Moreover, it seems crucial that before other pharmacologic preventive approaches are considered, current risk models are reassessed to reduce overtreatment. In this context, novel biomarkers should be assessed to improve risk models and thus possible effect sizes of preventive strategies.

Ultimately, the patient’s perspective should also be considered. While it is easier to accept pharmacologic treatment in the context of arthralgia and loss of functionality, as in the TREAT EARLIER study, individuals who feel healthy may be less eager to do so. This could have been observed in the STAPRA study, which was prematurely terminated due to low enrollment. The absence of severe symptoms probably gave little occasion to take medical consultation or consider pharmacologic treatment, even with such a “mild” drug as atorvastatin.

From our point of view, patients presenting with polyarthralgia in hands and feet and / or with joint swelling should be tested for RF and ACPA seropositivity. Broad screening is not recommended because only a percentage of individuals positive for RF and ACPA experience progression to RA. The serologic markers are not sensitive and specific enough to be used as screening tools. Patients with polyarthralgia and RF/ACPA positivity in whom other diagnoses (such as osteoarthritis) had been ruled out are considered “suspected” of developing RA and should be monitored regularly to detect joint swelling. If joint swelling does occur, a diagnosis of RA can and should be made, followed by the initiation of treatment based on the EULAR and national guidelines.

Given the evidence we have discussed, currently, no therapy should be administered to individuals with at-risk RA or CSA in clinical practice (outside of clinical studies). The assessed studies have shown that treatment in the preliminary phases of RA for a short period does not definitely prevent the progression of the disease. Therefore, treatment should only commence upon a confirmed diagnosis of RA.

In conclusion, the concept of RA prevention and the recently published trials assessing csDMARD and bDMARD treatment in individuals at risk for RA are landmark studies and are opening a new line of thinking. Nevertheless, prevention is elusive. Before investigation of other medical interventions in this population, risk models should be reassessed and tweaked to yield better predictive value to essentially reduce overtreatment.