This prospective study included consecutive SLE patients from the Department of Internal Medicine at Pitié-Salpêtrière Hospital (http://www.clinicaltrials.gov NCT01413230). Patients were eligible for the study when they met at least four of the 1997 American College of Rheumatology criteria for SLE 15. Inclusion criteria for the study were as follows: 1) inactive disease or mild to moderate active disease indicated by a score ≤ 8 in the Safety of Estrogens in Lupus Erythematosus National Assessment-Systemic Lupus Erythematosus Disease Activity Index (SELENA-SLEDAI), and 2) stable dosage of prednisone and/or immunosuppressive agents for at least 1 and/or 3 months, respectively. Pregnant patients and those planning pregnancy, and patients who had previously received B cell-targeted therapy were excluded. Disease activity was assessed using SELENA-SLEDAI 16.

Between 1 September and 31 November 2010, we assessed 24 SLE patients for eligibility (twenty-two women and two men, mean age ± SD, 31 ± 8 years). Their serum 25(OH)D level was measured. Hypovitaminosis D was defined as serum 25(OH)D < 30 ng/mL, while vitamin D sufficiency was defined as serum levels between 30 and 100 ng/mL 17. Those with hypovitaminosis D (< 30 ng/mL) were placed on the following schedule of oral vitamin D supplementation: 100,000 IU of cholecalciferol per week for 4 weeks, followed by 100,000 IU of cholecalciferol per month for 6 months. All supplemented patients were screened before vitamin D supplementation (Day 0, or D0), and 2 and 6 months (M2 and M6) after the beginning of vitamin D supplementation. All but four patients received hydroxychloroquine (200 or 400 mg daily) and/or oral prednisone (≤ 15 mg/day, median dosage 5 mg/day). Three patients received a stable dosage of immunosuppressive agents. The study was approved by the institutional ethics committee, the Comité de protection des personnes Ile-de-France VI, in the Pitié-Salpêtrière Hospital (Paris, France) and informed consent was obtained from all patients.

At each visit, the SELENA-SLEDAI was recorded. Routine measurements were made of 25(OH)D levels, antinuclear antibodies by indirect immunofluorescence on HEp2 cells (Immunoconcepts, Sacramento CA, USA), anti-dsDNA antibodies levels by ELISA (DiaSorin, Saluggia, Italy), complement C3 and C4 levels, complete blood count, serum creatinine, proteinuria and hematuria. Serum 25(OH)D was measured in serum samples by means of a radioimmunoassay after simple extraction with acetonitrile, as described previously 18.

The primary endpoint of this study was safety. Safety endpoints included the occurrence of hypercalcemia, hyperphosphoremia or lithiasis. Secondary outcomes included changes from baseline in T cell and B cell homeostasis, and cytokines and gene expression profiles, in peripheral blood mononuclear cells (PBMCs), and clinical efficacy assessment using the SELENA-SLEDAI.

PBMC subsets (CD3⁺, CD4⁺, CD8⁺ T lymphocytes, CD19⁺ B lymphocytes and CD3^-CD56⁺ NK cells) counts (cells/μL) were established from fresh blood samples using CYTO-STAT tetraCHROME kits with Flowcount fluorescents beads as an internal standard and tetra CXP software with a Navios cytometer according to the manufacturer's instructions (Beckman Coulter, Villepinte, France). Subsets of these cells were analyzed using multicolor flow cytometry and monoclonal antibodies (mAbs) directly conjugated to various fluorescent markers. PBMCs were also stained with the following conjugated mAbs at predetermined optimal dilutions for 30 minutes at 4°C: CD3-ECD, CD4-PCy7, CD4-ECD, CD8-PCy7, CD8-APC, CD10-APC, CD16-FITC, CD19-ECD, CD27-PE, CD28-FITC, CD45RO-FITC, CD45RA-APC, CD56-PE, HLA-DR-PCy7 (Beckman Coulter), CD25-PE, CD38-PCy7, CD56-FITC, CD62L-FITC, IgD-FITC (BD Pharmingen, Le-Pont-de-Claix, France), CCR7-PE and LAP-PE (R&D systems, Lille, France), CD127-FITC (eBioscience, Paris, France) and GITR-PE and CD25-APC (Miltenyi Biotec, Paris, France). Intracellular detection of FoxP3 and CD152 (CTLA4) was performed on fixed and permeabilized cells using appropriate buffer (eBioscience for FoxP3, BD Pharmingen for CD152). Cell acquisition and analysis by flow cytometry were performed using a Navios Cytometer (Beckman Coulter). Instrument setting parameters (gains, compensations, and threshold) were set with machine software (CXP Software; Beckman Coulter) in conjunction with calibration beads (Flow-set beads, Cytocomp kit, and CYTO-TROL Control Cells). Machine reproducibility was verified with standardized beads (Flow-check). Data were analyzed with CXP analysis software and Kaluza software (Beckman Coulter)

For detection of intracellular cytokine production, PBMCs were stimulated with 50 ng/mL phorbol myristate acetate (PMA) and 1 mM ionomycin in the presence of Golgi-Plug (BD Pharmingen) for 4 hours and then stained with anti-IL-4-FITC, anti-IFN-γ-FITC, anti-IL-10-PE (BD Pharmingen), or anti-IL-17A-Alexa Fluor 647 (eBioscience) after fixation and permeabilization, according to the manufacturer's instructions. Culture supernatants from PBMCs stimulated in the absence of Golgi-Plug were harvested and immediately frozen at -80°C.

We compared measures taken at baseline (before the vitamin D supplementation) with those taken at M2 and M6 after the initiation of vitamin D supplementation, using the Wilcoxon signed-rank test. All tests were two-sided with a 0.05 significance level. Graphing and statistical analyses were performed using Prism software (GraphPad Software, Inc.).

Out of the 24 patients screened for serum 25(OH)D levels, 20 patients (83%) had hypovitaminosis D and were included in this prospective study. The baseline characteristics of the 20 patients are listed in Table 1. All patients completed the 6 months follow-up period.

Serum 25(OH)D levels dramatically increased under vitamin D supplementation from 18.7 ± 6.7 at D0 to 51.4 ± 14.1 (P < 0.001) at M2 and to 41.5 ± 10.1 ng/mL (P < 0.001) at M6 (Figure 1A). Treatment was safe, with no significant increase of serum calcium and phosphorous and no occurrence of lithiasis. Although not statistically significant, disease activity assessed by the SELENA-SLEDAI slightly decreased from 2.9 ± 2.5 at D0 to 2.6 ± 2.5 at M2 (P = 0.67) and to 1.9 ± 1.8 at M6 (P = 0.16) (Figure 1B). C3 complement fraction levels remained stable during follow-up, while anti-dsDNA levels decreased from 177 ± 63 at D0 to 124 ± 67 at M2 (P < 0.05) and to 103 ± 36 IU/mL at M6 (P < 0.01) (Figures 1C). No patients required modification of the prednisone dosage or initiation of new immunosuppressant agents. We did not observe SLE flare during the 6 months follow-up period.

The impact of vitamin D supplementation on the proportion and the absolute number of CD3⁺ T cells, CD4⁺ and CD8⁺ T cells, CD19⁺ B cells and CD3^-CD56⁺ NK cells is shown in Figure 2. The mean proportions and absolute numbers at baseline were 80 ± 8% and 810 ± 274/mm³ for CD3⁺ T cells, 42 ± 7% and 412 ± 157/mm³ for CD4⁺ T cells, 37 ± 12% and 356 ± 128/mm³ for CD8⁺ T cells, 12 ± 6% and 138 ± 89/mm³ for CD19⁺ B cells and 6 ± 5% and 61 ± 46/mm³ for CD3^-CD56⁺ NK cells. At M2 and M6, the proportion and absolute number of CD3⁺ T cells, CD4⁺ and CD8⁺ T cells and CD3^-CD56⁺ NK cells remained stable, while CD19⁺ B cell frequency and counts significantly decreased at M2 (P < 0.05 for both) (Figure 2).

Naïve CD4⁺ T cells increased in frequency without reaching the significance level (P = 0.09 and P = 0.10, respectively) and in absolute number (P = 0.05 and P < 0.01, respectively) at M2 and M6 with vitamin D supplementation, while other CD4⁺ T cell subsets remained stable. Effector memory CD8⁺ T cells decreased in frequency (P = 0.02 and P = 0.01, respectively) but not in absolute number (P = 0.11 and P = 0.55, respectively) at M2 and M6 with vitamin D supplementation (Figure 3A).

Using IgD and CD27 surface markers, we also observed a decrease in IgD^-CD27⁺ class-switched memory B cell frequency (P < 0.05) and absolute number (P = 0.06) at M2 and in IgD^-CD27^- memory B cell frequency (P = 0.11 and P < 0.001, respectively) and absolute number (P < 0.05 and P < 0.01, respectively) at M2 and M6, while IgD⁺CD27⁺ marginal zone-like B cell frequency increased at M6 (P < 0.001) (Figure 3B).

The impact of vitamin D supplementation on CD3⁺CD4⁺CD25^hiCD127^-FoxP3⁺ Tregs and Treg subsets was evaluated (Figure 4A). The percentage and absolute count of Tregs at baseline was 3.5 ± 1.2% and 15 ± 8 cells/μL. The percentage of Tregs was increased at 4.6 ± 1.3% at M2 (P <0.001) and 4.3 ± 1.4% at M6 (P <0.01), and the absolute count of Tregs was increased at 23 ± 14 cells/μL (P <0.05) and 25 ± 14 cells/μL (P <0.01), respectively (Figure 4B). Analysis of CD45RA and CD25 expression on CD4⁺ T cells revealed that this increase concerned both resting and activated memory Tregs (Figure 4B). The increase in Tregs was associated with an increased expression of molecules associated with Treg suppression (that is, CTLA4, GITR and LAP) (Figure 4C).

Decreases were observed in Th1 cells from 16.9 ± 6.7% at D0 to 11.0 ± 5.1% at M2, and to 13.6 ± 6.5% at M6 (P < 0.01 and P < 0.05, respectively), in IFN-producing CD8⁺ T cells from 35.3 ± 13.2% at D0 to 26.8 ± 12.5% at M2, and to 29.7 ± 13.4% at M6 (P < 0.05 and P = 0.06, respectively), and in Th17 from 2.0 ± 1.1% at D0 to 1.6 ± 0.9% at M2 and to 2.0 ± 1.3% at M6 (P < 0.01 and P = 0.81, respectively) after vitamin D supplementation. The Th2 cells remained stable during follow-up (Figure 5).

Vitamin D supplementation using 100,000 IU of cholecalciferol per week for 4 weeks, followed by 100,000 IU of cholecalciferol per month for 6 months, was well tolerated.

Vitamin D induced a preferential increase of naïve CD4⁺ T cells, an increase of regulatory T cells, a decrease of effector Th1 and Th17 cells, and a decrease of memory B cells and anti-DNA antibodies.

This preliminary study suggests a beneficial role of vitamin D in SLE patients and needs to be confirmed in randomized controlled trials.