Background Vitamin D (Vit D) is increasingly recognized as an important immune modulating agent. No data has been reported on how Vit D status can interfere with the biotherapy response in RA patients.
Objectives The main objective was to determine the Vit D status in a cohort of RA patients and to evaluate the 6 month response to rituximab according to the level of plasmatic 25OHVit D at baseline.
Methods 111 RA consecutive patients (ARA 1987 criteria) were recruited from one single institution between 2005 and 2010 prior to initiation of rituximab. Patients characteristics including ethnic origin, BMI, smoking status, disease duration, autoantibody status (RF, ACPA), number of previous DMARDs, corticosteroid therapy status, Vit D supplementation, season of enrolment were documented. 25OHVit D was measured using baseline serum samples and a commercial ELISA kit (IDS immunodiagnostic system). Patients were classified in 3 categories according to 25OHVit D level as adequate (≥30 ng/ml), insufficient (between 10 and 30 ng/ml), or deficient (<10 ng/ml). Treatment response was evaluated according to DAS28 variation between M0 and M6 (Δ DAS28) (EULAR classification: good, moderate, none). We identified factors associated with the treatment response by univariate analysis (Chi-square or Fisher’s test for categorical variables, and analysis of variance for continuous variables). Factors with p-value less than 20% were entered into the multinomial logistic regression.
Results The cohort included 83% F, mean age 53±10.3 y, 86.5% were Caucasians; mean BMI 27.2±6.8; mean disease duration 12.1±9.7 y; RF positive 85.6%; ACPA positive 80%; mean number of previous biotherapies 1.4±1.1. The mean DAS28 at M0 was 5.2±1.3.The mean dose of prednisone at M0 was 8.6±7.7 mg/d. Season for enrolment was autumn (20.7%), winter (29.7%), spring (27.9%) and summer (21.6%). 64% received daily supplementation with Vit D (800 units/d). The mean level of 25OH Vit D was 17.1±8.7 ng/ml for the entire group. According to 25OH Vit D dosage, 9 patients were classified as adequate, 79 (71.2%) as insufficient and 23 (20.7%) as deficient. At M6 20.7% were in remission (DAS28 <2.6); 9% were in low disease activity (DAS28 <3.2); 47.7% moderate activity (DAS28 ≤5.1) and 22.5% were in high disease activity (DAS 28 >5.1).Response to treatment with rituximab was considered as good for 24.3%, moderate for 38.7% and 36.9% had no response. According to Vit D status at M0, the proportion of no response was higher in patients with 25OHVit D deficiency, 52.2% vs 31.6% for 25OH Vit D insufficiency and 44.4% for 25OH Vit D adequacy (p=0.0459; Fisher test). No correlations were seen between 25OHVit D level and Δ DAS28 or DAS28 at M0. The model of multinomial logistic regression analysis showed an association between a significant response and the season of mabthera initiation (better response frequency if initiated in automn) (p global value =0.0166).
Conclusions Whatever the season, inadequate Vit D status was observed in 92% of RA patients in this series, despite daily supplementation for 64% of them. The effect of normalization of Vit D status prior to the initiation of rituximab should be evaluated in order to minimize the rate of treatment failure.
Disclosure of Interest R. M’Barek: None Declared, T. Dupré: None Declared, F. Tubach: None Declared, P. Dieudé: None Declared, E. Palazzo: None Declared, G. Hayem: None Declared, K. Dawidowicz: None Declared, S. Ottaviani: None Declared, T. Alfaiate: None Declared, V. Leçon-Malais: None Declared, A. Boutten: None Declared, O. Meyer Grant/Research support from: ROCHE