Background/Purpose:

Anti-citrullinated peptide antibody (ACPA) and rheumatoid factor (RF) define “seropositive” rheumatoid arthritis (RA). Both predict disease course, development of extra-articular features and treatment outcomes. ACPA is more specific than RF for the diagnosis of RA. There is little evidence to determine if repeat testing of ACPA and RF is required. We investigated the incidence of seroconversion to ACPA after commencement of treat-to-target therapy in patients with early RA.

Methods:

DMARD-naïve patients presenting between September 1998 – July 2017 with RA according to the 1987 ACR criteria, with a disease duration of <12 months were enrolled. All patients received combination triple DMARD therapy (methotrexate, sulfasalazine and hydroxychloroquine) according to a treat-to-target strategy. Adjustments to treatment by predefined protocol were contingent on failing to reach a target DAS28-ESR score of <2.6.

ACPA and RF were recorded at baseline and at least 6 monthly. Before 2011, ACPA was analysed using ELISA and then by immunoassay. Before 2011, RF was measured by latex nephelometry and then by turbidimetry with excellent correlation between methods. Normal ranges were defined for ACPA as <6 units/mL and for RF <14 IU/mL. Differences between seroconverters and non-seroconverters were assessed via Mann Whitney test (for continuous variables) and Fisher Exact Test (for categorical variables).

Results:

368 patients had a total of 2889 ACPA and 4170 RF measurements. Median follow up was 272 (IQR 53-467) weeks. Median disease duration at diagnosis was 20.4 weeks (IQR 11-26). At 1 year, 44% of patients were in DAS28 remission and 10% progressed to require a biologic DMARD during study follow up.

At baseline, 154 (41.8%) patients were seronegative for ACPA, 10 (6.5%) of whom seroconverted at some time during follow up. Four (2.6%) patients had persistent seroconversion. Of the 10 ACPA seroconverters, 9 patients were RF positive at baseline. Baseline RF titre predicted seroconversion to ACPA (mean 10IU/ml vs 55IU/ml in non-converters, p<0.001). Of the 107 patients who were RF negative and ACPA negative at baseline, ACPA became positive at 10 IU/ml in a single patient who subsequently reverted to negative. Median time to seroconversion for ACPA was 29 months (range 3-192). No patient seroconverted from negative to positive for both RF and ACPA.

Conclusion:

Persistent seroconversion of ACPA from negative to positive after diagnosis in patients with RA is uncommon. ACPA and RF double negative patients are highly unlikely to ever develop ACPA positivity with a risk of < 1%. It is therefore unlikely to be helpful or cost effective to perform serial ACPA measurements in patients with seronegative RA.

ACR Meeting Abstracts - https://acrabstracts.org/abstract/utility-of-serial-acpa-measurements-in-rheumatoid-arthritis-results-of-a-prospective-cohort-study/