Biomarkers have supplemented clinical findings and imaging in the diagnosis, prognosis and management of rheumatoid arthritis (RA) for decades: rheumatoid factor was first described more than 75 years ago, in 1939.

Today there are hopes that biomarkers will also prove valuable in guiding the selection of the most effective biologic agents for individual patients, replacing the current ‘trial and error’ approach to the initiation and sequencing of treatments.

Background

Rheumatoid factor (RF) – autoantibodies directed against immunoglobulin G, A, M, E and D isotype antibodies – are detectable in up to 70-80% of patients with RA but they are not specific for the disease. They occur in other systemic diseases including Sjogren’s syndrome and systemic infections, and in about 10% of healthy people.¹ The presence of RF was included as a marker of RA in diagnostic criteria from the American College of Rheumatology (ACR) in 1987.

More recently, the focus has shifted to anti-citrullinated protein antibodies (ACPA). ACPA has about 67% sensitivity for RA and, importantly, 95% specificity. A 2015 review noted that ACPA is useful in RF-negative patients with early RA, and a combination of RF and ACPA has a positive predictive value close to 100%.¹ ACPA status was included in classification criteria released by the ACR and the European League Against Rheumatism (EULAR) in 2011.

In addition, ACPA is detectable up to 10-14 years before the onset of symptoms, raising the prospect of early identification of at-risk individuals. Maturation of the ACPA response over time is associated with the transition to clinical RA.²

ACPA, RA pathology and disease course

Current diagnostic criteria for RA are likely to identify a mixed group of patients with different underlying pathologies.

“ACPA-positive and ACPA-negative disease have been shown to be associated with different genetic and environmental risk factors, fuelling the hypothesis that different pathophysiological mechanisms underlie these two separate disease subsets,” according to Professor Tom Huizinga and colleagues from Leiden University Medical Center in The Netherlands.²

“For example, ACPA-negative RA associates with HLA-DR3,74,75 whereas the HLA shared epitope (SE) alleles predispose to ACPA-positive disease.” In addition, the contribution of smoking to disease risk is mainly con­fined to the ACPA-positive HLA-SE-positive patient group.

Although ACPA-positive and ACPA-negative patients have a similar clinical presenta­tion early in the disease, their subsequent disease course is different. ACPA positivity is associated with an increased risk of extra-articular manifestations nodules, vasculitis, lung involvement, and a more destructive course of RA.

“ACPA-positive patients develop erosions earlier and more abundantly,” they said. “Owing to their more severe disease course, APCA-positive patients require a more aggressive treatment regimen than ACPA-negative patients.”

ACPA and treatment response

Data from randomised clinical trials has shown that ACPA-positive and ACPA-negative patients respond differently to methotrexate treatment, and methotrexate alone may be insufficient for achieving disease control in patients with high APCA levels.² In addition, ACPA-positive status in patients with RA refractory to TNF inhibitors are more likely to respond to the anti-B cell biologic rituximab.

More recently, the AMPLE study, published in September 2015, concluded that the presence of ACPA at baseline in biologic-naive RA patients was associated with a better response to both abatacept and adalimumab.³ Patients with the highest baseline ACPA levels had a better clinical response with abatacept than patients with lower concentrations, but a similar association was not observed with adalimumab.

AMPLE randomised 646 patients to either abatacept 125 mg weekly or adalimumab 40 mg bi-weekly. ACPA concentrations were measured at baseline, and ACPA-positive patients – 77% of the total – were divided into quartiles (Q1 to Q4) according to increasing antibody concentrations.

“At year 2, improvements in disease activity and disability and remission rates were similar across Q1–Q3, but were numerically higher in Q4 in the abatacept group,” the researchers said. “In contrast, treatment effects were similar across all quartiles in the adalimumab group.”

The potential role of ACPA as a predictor of response to therapy has been confirmed in related studies:

• ‘Real-world’ data from the prospective observational ACTION study of 672 biologic-naive patients showed the clinical efficacy of abatacept was superior in patients who were RF and/or ACPA seropositive rather than seronegative, despite differences in other predictors of disease progression at baseline.⁴

• The AVERT study randomised patients with RA to treatment with abatacept, methotrexate or the combination.[Huizinga] Combination therapy had increased clinical efficacy in patients who were ACPA-positive at baseline compared to those who were not, and in those who seroconverted from ACPA-positive to ACPA-negative status over time compared to those who did not. Combination therapy was also associated with a higher rate of conversion from ACPA-positive to negative status than monotherapy with either agent. “These results suggest that the impact of abatacept and MTX therapy on ACPA is associated with clinical benefit and that ACPA positivity may help to identify patients most likely to benefit from abatacept treatment,” the researchers concluded.

• Analysis of 802 patients in the Brigham and Women’s Hospital Rheumatoid Arthritis Sequential Study (BRASS) registry who were ACPA positive at baseline found that those with a reduction in ACPA levels during treatment, regardless of the agents used, had a numerically greater reduction in disease activity levels than those in whom the levels did not reduce.⁶

• Another analysis based on the BRASS registry supported observations from clinical trials that abatacept treatment leads to a reduction in ACPA levels that is not seen with TNF inhibitors (TNF-I).⁷

A total of 74 patients treated with abatacept were matched with 66 patients treated with TNF-I, based on baseline ACPA levels, disease duration and DAS28 (CRP) at baseline. ACPA increased by a mean of 2.11 units at year 1 in the TNF-I group, compared to a reduction of 17.25 units in abatacept-treated patients. Similar trends were observed in the second year: ACPA had increased by 9.25 units from baseline in TNF-I-treated patients, compared to a reduction of 22.90 units from baseline in the abatacept group.

The future

Associate Professor Peter Nash, from the Department of Medicine, University of Queensland at the Sunshine Coast, says that rheumatologists continue to rely on both RF and ACPA in the diagnosis of RA and as prognostic markers, but there is tantalising evidence that ACPA status may also help guide the choice of biologic therapy.

“We know that ACPA-positivity predicts more severe disease, especially in smokers, and suggests an need for early, effective disease control,” he says. “Established markers of aggressive disease also include a family history of RA, being young and female at the time of disease onset, having a high ESR, reduced physical function, and the presence of joint erosions and nodules. We don’t emphasise enough the need to cease smoking, which reduces the efficacy of therapy and drives disease activity.”

A systematic review of the potential for personalised biologic treatment, which did not incorporate the recent AMPLE study, found no existing biomarkers added more than 15% to a prediction of response to a specific therapy.⁸

“The fact is, we have an expanding range of biologic therapies and practising rheumatologists want a biomarker that will predict the response from a specific agent in a specific patient,” Professor Nash said. “But biomarkers have been disappointing so far.

“Ineffective treatments are financially costly, add risk of toxicity, and the damage from the disease continues to accumulate while ineffective treatment is being trialled.”

Understanding the contribution of ACPA to the pathogenesis of RA supports its potential role as an important predictor of disease status and therapeutic response, he says. “ACPA is often present many years in advance of clinical disease, and we know that early treatment in ACPA-positive patients may prevent disease progression in studies using methotrexate and abatacept.”

“Simple and inexpensive biomarkers like the acute phase reactants – ESR and CRP – have been clinically useful in guiding our assessment of disease activity , response to treatment, and in identifying patients who would otherwise have an aggressive disease course. We now have evidence that high ACPA levels predict a response to abatacept as well as rituximab. This is a helpful step in targeting biologics to the right patient.”

References

1. Mohan C, Assassi S. Biomarkers in rheumatic diseases: how can they facilitate diagnosis and assessment of disease activity? BMJ 2015 Nov 26;351:h5079.
2. Willemze A et al. The influence of ACPA status and characteristics on the course of RA. Nat Rev Rheumatol 2012; 8: 144-52.
3. Sokolove J et al. Impact of baseline anti-cyclic citrullinated peptide-2 antibody concentration on efficacy outcomes following treatment with subcutaneous abatacept or adalimumab: 2-year results from the AMPLE trial. Ann Rheum Dis 2015 Sep 10. pii: annrheumdis-2015-207942.
4. Alten R et al. Baseline autoantibodies preferentially impact abatacept efficacy in patients with RA who are biologic naïve: 6-month results from a real-world, international, prospective study. ACR/ARHP Annual Scientific Meeting 2015, Abstract 551.
5. Huizinga TWJ et al. Effect of anti-cyclic citrullinated peptide 2 immunoglobulin M serostatus on efficacy outcomes following treatment with abatacept plus methotrexate. ACR/ARHP Annual Scientific Meeting 2015, Abstract 2668.
6. Alemao E et al. Evaluation of anti-cyclic citrullinated peptide autoantibody levels in clinical practice and its association with disease activity. ACR/ARHP Annual Scientific Meeting 2015, Abstract 2663.
7. Alemao E et al. Evaluation of the impact of disease-modifying antirheumatic drugs on anti-cyclic citrullinated peptide autoantibody levels in clinical practice. ACR/ARHP Annual Scientific Meeting 2015, Abstract 2651.
8. Cuppen BV et al. Personalized biological treatment for rheumatoid arthritis: a systematic review with a focus on clinical applicability. Rheumatology (Oxford) 2015 Dec 29. pii: kev421.

Peter Nash did not receive honoraria for his comments from the limbic or Bristol Myers Squibb Australia.

This publication has been created with an educational grant from Bristol-Myers Squibb Australia. The content is based on published studies and the presenters’ opinions. It may not reflect the views of Bristol-Myers Squibb. Please refer to Full Product Information before prescribing any of the products mentioned in this article. Treatment decisions based on these data are the full responsibility of the prescribing physician.