AB0515 COMPARISON OF THE PERFORMANCE OF THE 1990 ACR CLASSIFICATION CRITERIA AND THE GIACTA INCLUSION CRITERIA FOR THE CLASSIFICATION OF GCA: RESULTS FROM A SINGLE-CENTRE STUDY.

Abstract

Background:The classification criteria currently used to define giant cell arteritis (GCA) were developed in 1990 by the American College of Rheumatology (ACR), and strongly focus on patients with cranial manifestations. Patients with large-vessel GCA (LV-GCA) have less frequently cranial symptoms and a positive temporal artery biopsy, and are less likely to be captured by the ACR criteria. GiACTA, a trial of tocilizumab in GCA, has recognized the concept of GCA as a clinical syndrome, and included patients with cranial and/or polymyalgic symptoms as long as GCA diagnosis was supported by either biopsy or appropriate LV imaging results. However, these inclusion criteria were elaborated by experts and were not validated in patients with GCA.Objectives:To compare the performance of the 1990 ACR classification criteria and the GiACTA inclusion criteria for the classification of GCA in a single-center cohort of patients with GCA.Methods:All consecutive patients with a diagnosis of GCA seen between January 2008 and December 2016 in our center were included (GCA cohort). Control cohort consisted of consecutive patients with a negative temporal artery biopsy (TAB) performed in the same time period and a final diagnosis different than GCA. For both study cohort, the final diagnosis was made at the end of the follow-up period by consensus by 2 rheumatologists, who retrospectively evaluated all the medical records from symptoms’ onset to December 2019, last visit, or death. Subjects were classified by each of the different criteria. TABs showing inflammation limited to adventitial or periadventitial small vessels were considered negative for both ACR and GiACTA criteria.Two-by-two classification tables were generated to estimate sensitivity and specificity, and receiver operating characteristic (ROC) curves with corresponding areas under the curve (AUC) were calculated.Results:213 patients were included in the study (75% female, mean age 71.7 years). 55 patients had TAB showing transmural inflammation (TMI); 30 patients had TAB showing inflammation limited to adventitial or periadventitial small vessels (PAI); 67 patients had evidence of LV-GCA at imaging (LV-GCA) and 61 patients had TAB without inflammatory changes (negTAB). 1990 ACR and GiACTA criteria were satisfied respectively by 55 (100%) and 51 (93%) TMI, 18 (60%) and 1 (3%) PAI, 23 (35%) and 31 (46%) LV-GCA and 27 (44%) and none (0%) negTAB patients.After a median follow-up of 52.6 months, 174 of the 213 (84%) patients had a final diagnosis of GCA (55 TMI, 22 PAI; 67 LV-GCA and 30 negTAB) and the remaining 33 patients had a diagnosis different than GCA (2 PAI and 31 negTAB). Sensitivity and specificity of 1990 ACR classification criteria for GCA were 67% and 90%, AUC (95% CI) 0.790 (0.715 – 0.864). Sensitivity and specificity of GiACTA inclusion criteria were 48% and 100%, AUC (95% CI) 0.740 (0.669 – 0.811). By adding systemic symptoms in the symptoms domain of GiACTA inclusion criteria, sensitivity increased to 59% and sensitivity remained 100%, AUC (95% CI) 0.792 (0.730 – 0.854).Conclusion:Both 1990 ACR classification criteria and GiACTA inclusion criteria showed a good specificity but a low sensitivity in classifying patients with a clinical diagnosis of GCA from this large monocentric cohort. There is an urgent need for new classification criteria for GCA.Disclosure of Interests:Francesco Muratore: None declared, Luigi Boiardi: None declared, Elena Galli: None declared, Giulia Pazzola: None declared, Alberto Cavazza: None declared, Giovanna Restuccia: None declared, Carlo Salvarani Grant/research support from: consulting and investigator fees from Abbvie, Pfizer, MSD, Roche, Celgene, Novartis, Consultant of: consulting and investigator fees from Abbvie, Pfizer, MSD, Roche, Celgene, Novartis