MANCHESTER, ENGLAND – Anti–carbamylated protein antibodies were associated with both increased disability and greater disease progression over time in patients with early inflammatory arthritis, data from a 20-year retrospective, observational analysis showed.

According to the research, presented at the British Society for Rheumatology annual conference, patients who were also positive for anti–citrullinated protein antibodies (ACPAs) fared worse long term than those who were ACPA negative.

The results showed that anti–carbamylated protein (anti-CarP) antibodies may provide additional prognostic information to current antibody tests in inflammatory arthritis (IA), but their measurement is not yet ready for general use, said Dr. Jennifer Humphreys of the Arthritis Research UK Centre for Epidemiology at the University of Manchester.

“We know that inflammatory arthritis patients who are rheumatoid factor [RF] or ACPA positive have an increased risk of persistent disease and a poor prognosis,” she said.

“In particular, ACPA[s] are associated with severe radiological progression [and] increased disability and mortality. However, not all patients who lack these antibodies do well,” Dr. Humphreys said.

Research at the University of Leiden (the Netherlands) has found that ACPA-negative patients have anti-CarP antibodies in their sera and that these autoantibodies precede the onset of symptoms and predict the development of rheumatoid arthritis (RA) in patients with arthralgia. These antibodies have also been associated with radiological damage, she said.

Because there were few data on the relationship to disease activity and long-term outcomes in the presence of these antibodies, Dr. Humphreys and her associates decided to look at this using data from the Norfolk Arthritis Register (NOAR), she said. The U.K. team worked in collaboration with the Dutch researchers who performed the anti-CarP tests on serum samples taken from 1,995 patients with early IA enrolled in NOAR from 1990 onward.

At baseline, 1,222 patients met American College of Rheumatology/European League Against Rheumatism 2010 criteria for RA. The mean age of onset was 55 years for both the total population with IA and for those with RA. About two-thirds of the participants were women, and their median symptom duration at enrollment was 33 weeks.

Long-term disability was assessed using the health assessment questionnaire (HAQ), and disease activity was evaluated using the disease activity in 28 joints (DAS28) score. The median baseline DAS28 score was 3.81 for the total cohort and 4.45 for patients with RA, and the median baseline HAQ scores were a respective 0.875 and 1.125.

When the team looked at patient baseline characteristics according to baseline anti-CarP status, they found that the 460 (23%) patients who were positive for these autoantibodies were more likely to have higher baseline DAS28 scores (4.23 vs. 3.73), HAQ scores (1.125 vs. 0.875), and C-reactive protein levels (14 mg/L vs. 7 mg/L) than the 1,543 patients who tested anti-CarP negative. Patients positive for anti-Carp antibodies were also less likely to be female (58% vs 68%) and were slightly older at onset (55 vs. 53 years).

Dr. Humphreys noted that there was a lot of overlap among the presence of anti-CarP, ACPA, and RF. Of the 460 who were anti-CarP positive, 124 were positive only for anti-CarP, 148 were also RF positive, 30 were also ACPA positive, and 158 had all three antibodies.

“Patients who were anti-CarP antibody positive had higher HAQ scores at baseline, and this higher level of disability persisted throughout follow-up, and that association remained significant in the multivariate model,” Dr. Humphreys said.

“When we stratified by ACPA status, there was a statistically significant association in the ACPA-negative subgroup but not in the ACPA-positive subgroup,” she added. However, she noted that the beta coefficients for the association between anti-CarP antibodies and HAQ scores over time were similar (0.12 for the total cohort, 0.14 for the ACPA-negative patients, and 0.09 for the ACPA-positive patients) and that confidence intervals were wide.

Similar findings were seen for the association between anti-CarP antibodies and the DAS28 over time, with respective beta coefficients of 0.23, 01.18, and 0.25 for the total cohort, the ACPA-negative subgroup, and the ACPA-positive subgroup. Again, confidence intervals were wide.

The researchers looked at the additive effect of having anti-CarP and ACPA, and found that while there is no association between RF positivity and long-term disability, there were independent associations with both ACPA positivity and anti-CarP antibody positivity, with respective beta coefficients of –0.05, 0.12, and 0.14.

“And we see the exact same pattern with disease activity over time,” Dr. Humphreys reported.

Thus, measuring anti-CarP antibodies “may be a useful test to do in addition to both RF and ACPA,” she suggested, adding that measuring these antibodies “may be particularly valuable in patients who lack other antibodies.”

NOAR is funded by Arthritis Research UK. Dr. Humphreys had no relevant financial disclosures.

Anti-CarP antibodies linked to worse outcome in early arthritis