LAUSR

Background High-throughput high-sensitivity ELISAs for autoantibodies associated with CTD, such as extractable nuclear antigens (ENA), are used widely. Anti-Jo-1 (anti-histidyl-tRNA synthetase), one of this panel, is believed to confer a poor prognosis due to an association with interstitial lung disease (ILD) and myositis. Objectives To describe: the pattern of anti-ENA positive tests; frequency of repeated requests; stability and repeatability of anti-Jo-1 tests; clinical characteristics of anti-Jo-1 +ves compared with controls; and diagnostic value of anti-Jo-1 for ILD. Methods All anti-ENA test requests, from any hospital department, between Jan 2013 and Dec 2014 were identified. Serum samples are screened for ENA (Quanta Lite® ENA profile, Inova Diagnostics) and positive samples have specific ENA antibodies levels quantified. Data from anti-Jo-1 positive patients and controls was extracted from electronic records allowing a minimum of 12 months after first test. Results Jo-1 Positive (n=40) Controls (n=80) P value* Age, mean years (range) 53 (19–86) 52 (17–87) – Sex (% female) 70% 79% 0.37 Dead 13% 4% 0.12 Current or previous malignancy 10% 10% 1.0 Raynauds 17.5% 6.3% 0.10 Inflammatory arthritis 20% 19% 1.0 Clinical myositis diagnosis 5% 1.3% 0.26 CPK >1000 units/liter 5% 1.3% 0.26 Interstitial lung disease 12.5% 6% 0.30 CT chest done during study period 17/40 20/80 0.06 ANA (≥1:100) 18/38 (47.4%) 22/79 (27.8%) 0.06 RF 8/25 (32%) 12/44 (27.3%) 0.78 CCP 0/19 (0%) 3/33 (9.1%) 0.54 Anti-dsDNA (Crithidia +ve) 7.5% 1.3% 0.11 Scl70 7.5% 0% ** SSA/Ro 10% 0% ** SSB/La 10% 0% ** RNP 10% 0% ** *Fisher's exact test, two tailed. **Statistical analyses were not done on these comparisons as by definition controls were negative for ENA antibodies. 4009 samples from 3581 patients were tested. The first sample tested, chronologically, was designated test of interest. 616 (17.2%) patients were anti-ENA screen +ve, and 40 (1.1%) anti-Jo-1 +ve (>20 AU/mL). Anti-ENA tests were done more than once for 350/3581 (9.8%) patients (428/4009 (10.7%) samples) and for 7/40 (17.5%) of anti-Jo-1 +ve patients. The median interval between 1st and 2nd requests: 124 days (IQR 233 days). The Table shows data for anti-Jo-1 patients and randomly selected ENA -ve controls. The frequency of ILD, myositis and Raynaud's was comparable. Sensitivity and specificity of Jo-1 for ILD, a key feature of “anti-synthetase syndrome”, were 50% (CI 19–81%) and 68% (CI 59–77%) respectively. Positive predictive value 12.5% (CI 4 to 27%) and negative predictive value 93.8% (CI 86–98%). Of patients with the highest anti-Jo1 titres (≥40 AU/mL, 10/40 patients, 25%): 3 had ILD, 1 myositis and 2 had a malignancy (disseminated melanoma and CML). Bland-Altman plots show that anti-Jo-1 values remained stable when patients were re-tested at another time but re-testing available stored samples from +ve patients showed important variation (Figure). Conclusions Anti-Jo-1 is uncommon in a heterogenous hospital population and is only weakly predictive for ILD. When tested repeatedly levels remain stable over many months. Repeated testing for anti-ENA is common and potentially unnecessary. Controls over repeated requests could yield cost savings. Disclosure of Interest None declared