Rheumatology News

8 RHEUMATOID ARTHRITIS

R heumatology N ews • Vol. 4 • No. 1 • 2016

Study challenges protein citrullination as a central cause of RA

RA33 autoantibodies seemed to happen about a decade into the dis- ease course, Dr Konig added. “These data suggest that citrullination may not be required to break tolerance to RA33, and support a model in which RA autoantigens are initially targeted as a native protein, and only later become targets of a citrulline- specific response,” he said. Notably, immunoprecipitation, im- munoblotting, and competitive assays all confirmed a third type of autoan- tibody that cross-reacted against both RA33 and citrullinated RA33. The “RA33 [protein] is targeted by patient sera in three ways: only as a native protein, both as a native and as a citrullinated protein, and only as a citrullinated protein,” Dr Konig concluded. He and his associates are exploring the clinical implications of the finding, particularly because patients with the cross-reactive anti- RA33 autoantibodies seem to have the most rapidly progressive and severe disease, he added. The pres- ence of these antibodies might one day help identify a patient who needs especially aggressive monitoring and treatment, he concluded.

help bring a subset of patients with the most advanced and aggressive disease under better control.” Signs and symptoms are often inconclusive early in the course of RA, leading to the search for reliable biomarkers that can hasten diagno- sis and treatment. Anti-citrullinated protein antibodies (ACPAs) are one hallmark of RA, and protein citrul- lination has been seen as central to autoimmunity in its pathogenesis, said Dr Konig, a postdoctoral fellow in rheumatology at Johns Hopkins University, Baltimore. But patients also have autoantibodies against native or unmodified proteins, including calpastatin, Fc-gamma, peptidylarginine deiminase type 4, and heterogeneous nuclear ribonu- cleoprotein A2/B1 (also known as RA33), which has been correlated with clinical disease activity, radio- graphic evidence of bone resorp- tion, C-reactive protein levels, and erythrocyte sedimentation rate in

BY AMY KARON Frontline Medical News At the American College of Rheumatology annual meeting, San Francisco A utoantibodies in patients with rheumatoid arthritis target both the native and citrullinated forms of the RA33 autoantigen, challeng- ing the idea that protein citrullina- tion underlies loss of tolerance in this disease, Dr Maximilian Konig said at the annual meeting of the American College of Rheumatology. “I think the important thing is that this makes us rethink how RA actually starts,” Dr Konig said in an interview. “We identified three different antibody groups that clinically behave very differently. We identified a group of patients that has cross-reactive antibodies against RA33, and they seem to be the ones with the highest and the most rapid disease progression. Maybe identifying these patients early and targeting themmore would

These data suggest that citrullination may not be required to break tolerance to RA33, and support a model in which RA autoantigens are initially targeted as a native protein, and only later become targets of a citrulline-specific response.

also used immunoblotting and mass spectrometry to study synovial fluid from the patients. The assays identified citrullinated RA33 in the joints of RA patients, and revealed distinct autoantibodies that targeted native and citrullinated RA33. Furthermore, this single anti- body system seemed to change with disease duration, Dr  Konig said. Au- toantibody against native RA33 was almost exclusively found in samples from patients in early-stage disease, whereas patients with long-estab- lished RA had much higher levels of anti-citrullinated RA33 antibodies. The switch from a predominance of anti-RA33 to anti-citrullinated

a few previous studies ( Pediatr Int 2009;51:188–92 and J Immunol Res 2014;2014:516593). Dr Konig and his coinvestigators contended that current models of RA do not adequately account for autoimmunity against native pro- teins. To explore that idea, they tested sera from 196 patients from the ESCAPE RA (Evaluation of Subclinical Cardiovascular Disease and Predictors of Events in Rheu- matoid Arthritis) cohort study and from 56 healthy controls. They used quantitative ELISA to identify autoantibodies, and performed im- munoblotting and immunoprecipita- tion to test antibody specificity. They

Dr Konig had no disclosures.

Don’t stop TNFis during rheumatoid arthritis pregnancy

Rheumatologists aren’t giving methotrexate a fair shot

of 13 mg in the second trimester versus 8 mg in the control group. Perhaps not surprisingly, the mean duration of pregnancy was 37 weeks in the TNFi group, with six women (33%) delivering at or before 37 weeks; women in the control group delivered, on average, at 39 weeks, with four (17%) delivering at or before week 37. The investigators found that the risk of preterm birth increased with every cumulative milligram of prednisolone (OR, 1.08; 95% CI, 1.02–1.15; P < 0.01). “Women with RA who discontinue TNFis at conception face a high risk for flares during preg- nancy, independently of known risk factors like seropositivity. Flares are usually treated with pred- nisolone. We found a dose-dependent, significant increased risk for preterm birth associated with prednisolone. In this era of treat-to-target man- agement of RA, our paradigm for RA pregnancy management may need adjusting. By controlling RA activity with medications considered relatively safe in pregnancy, we may be able to improve both the pregnancy experience and pregnancy outcomes,” the investigators concluded. The women were 33 years old on average, and all had live births; four early miscarriages were excluded from analysis. All the pregnancies were planned, with methotrexate discontinued at least 3 months before conception. There was no statistical difference in the rate of seropositivity between the groups, “which is interesting because we know seropositivity is a risk factor for staying active during pregnancy,” Dr Fischer-Betz said. Two boys born to women who took TNFis had minor malformations, one with nasal bone aplasia, retrognathia, and hydronephrosis, and the other with hypospadias. Both of their mothers had taken etanercept in the first trimester. There was one malformation in the control group, a girl born with hydronephrosis.

BY M. ALEXANDER OTTO Frontline Medical News At the American College of Rheumatology annual meeting, San Francisco

BY M. ALEXANDER OTTO Frontline Medical News At the American College of Rheumatology annual meeting, San Francisco

O ral methotrexate is frequently underdosed, given for an inadequate length of time, and rarely switched to subcutaneous formulations before rheumatologists move on to biologics, according to an analysis of claims data from 35,640 rheumatoid arthritis patients. “There’re some major concerns here. Methotrexate is the anchor drug for rheumatoid arthritis, the best drug we have. More appropriate [use] could lead to better control” and “produce significant cost savings,” said investigator Dr James O’Dell, chief of the division of rheumatology at the University of Nebraska Medical Centre, Omaha. When patients don’t fully respond to lower doses, the ground rules for oral metho- trexate include escalation up to 25 mg or a switch to subcutaneous formulations, which have better bioavailability. Those moves should be considered before turning to biologics, Dr O’Dell explained at the American College of Rheumatology annual meeting. Rheumatologists, by and large, aren’t playing by those rules, according to the analysis. “We need to own these data because the majority of patients, over three-quarters, were treated by rheumatologists. We are not doing a great job,” Dr ODell said. The claims data came from Symphony Health Solutions, which captures about 92% of prescriptions written in the US. The 35,640 rheumatoid arthritis patients in the study were started on methotrexate in 2009 and followed through 2014; 15,599 (43.8%) didn’t need anything else and stayed on oral methotrexate alone throughout the study period. Prescribers, however, gave up on oral methotrexate at a mean dose of 15.3 mg and moved 17,528 patients (49%) straight to a biologic without giving subcutaneous methotrexate a shot. They did that after a median of less than 6 months, and within 3 months in more than 40% of patients. Just 2513 patients (7%) moved on to subcutaneous methotrexate when their oral formulation wasn’t enough. That’s all most of them needed; 1802 (72%) remained on subcutaneous methotrexate alone for the remainder of the study period. The rest moved on to a biologic, but after a median of almost a year, not a few months. When their time on oral and subcutaneous methotrexate was included, their median time to a biologic was more than 2 years. The investigators checked to see if things improved for patients who started on oral methotrexate in 2012. “The answer was no. We didn’t do better,” Dr O’Dell said. He didn’t speculate on why methotrexate is underused in the US. Claims data can’t address why patients were switched from methotrexate and other issues, but such nuances “don’t even begin to explain the doses and the timing of switch that we saw here,” he said.

I t might be best to keep women with rheumatoid arthritis on their tumour necrosis factor block- ers during pregnancy, according to German investigators. They found that women are likely to flare with- out them and need more prednisolone, which is associated with preterm birth and other problems, while an increasing body of evidence suggests that tumour necrosis factor inhibitors (TNFis) are rela- tively safe during pregnancy. “We should” rethink discontinuing TNFis dur- ing pregnancy, as recommended in some quarters. “We do not want women to flare during pregnancy,” said investigator Dr Rebecca Fischer-Betz of the department of rheumatology at Heinrich Heine University in Düsseldorf. She and her colleagues compared birth out- comes in 18 rheumatoid arthritis (RA) patients who discontinued TNFi treatment shortly after they got pregnant against those of 24 women with RA who were never exposed to a TNFi because, in general, they had less severe disease. Twelve of the women (75%) in the TNFi group flared, versus four women (17%) in the control group. Although patients in both groups started with a mean 28-joint Disease Activity Score us- ing C-reactive protein (DAS28-CRP) below 3.0, women in the TNFi group had a rise in activity to a mean of about 3.5 in the second trimester, while disease activity in control patients remained stable. Compared with controls, women who stopped TNFis were also far more likely to flare (odds ratio, 10.0; 95% confidence interval, 2.3–42.8; P = 0.002), even after adjusting for age, DAS28- CRP at conception, rheumatoid factor and cyclic citrullinated peptide status, and other potential confounders. They also relied more heavily on prednisolone, taking, for example, a mean dose

There was no outside funding for the work, and the investigators have no disclosures.

Dr O’Dell is an adviser for AbbVie, Lilly, Coherus, Bristol-Myers Squibb, Antares, and Medac. Other investigators disclosed relationships with those or other companies.