Oral supplementation with cholecalciferol imrproves vitamin D status in HIV-infected children and young adults

Cholecalciferol Supplementation in HIV- Infected Youth with Vitamin D Insufficiency: Effects on Vitamin D Status and T-Cell Phenotype: A Randomized Controlled Trial Vania Giacomet,1 Alessandra Vigano,1 Valeria Manfredini,1 Chiara Cerini,1 Giorgio Bedogni,2 Stefano Mora,3 Manuela Borelli,4 Daria Trabattoni,4 and Gian Vincenzo Zuccotti1 1Department of Paediatrics, Luigi Sacco Hospital – Università degli Studi di Milano, Milan, Italy; 2 Clinical Epidemiology Unit, Liver Research Center, Basovizza, Trieste, Italy; 3 Laboratory of Pediatric Endocrinology and BoNetwork, Division of Metabolic and Cardiovascular Sciences, San Raffaele Scientific Institute, Milan, Italy; 4Immunology, Università degli Studi di Milano, DISP LITA Vialba, Milan, Italy Objectives: In addition to its known effects on bone metabolism, vitamin D may regulate immune function. Design: We performed a randomized controlled trial (RCT) to test whether cholecalciferol supplementation can improve vitamin D status and affect the T-cell phenotype in HIV-infected youth with vitamin D insufficiency. Methods: Fifty-two HIV-infected patients aged 8 to 26 years and with serum 25(OH) D <30 ng/mL were randomized to receive orally vitamin D3 100,000 IU or placebo every 3 months for 4 doses. Serum 25(OH)D, 1,25(OH)2D, PTH, and CD4+ T cells were assessed 3 months before baseline and at 0, 3, 6, 9, and 12 months, while Th1-, Th2-, Th17-, and Treg-subsets and T-lymphocyte vitamin D receptor were assessed at 0, 3, and 12 months. Results: Forty-eight subjects (25 receiving vitamin D and 23 receiving placebo) completed the RCT. Cholecalciferol supplementation produced an early (3 months) decrease in PTH, a concomitant increase in 25(OH)D, and a later (6 months) increase in 1,25(OH)2D levels, all persisting at 12 months. The frequency of vitamin D insufficiency at 12 months was 20% versus 60% in the inter- vention versus placebo group (P = .007). Cholecalciferol supplementation had no effect on CD4+ T-cell counts but was associated with a decreased Th17:Treg ratio at 3 months. Conclusions: In our cohort of HIV-infected youth, a 12-month chole- calciferol supplementation increased 25(OH)D and 1–25(OH)2D and decreased PTH levels but had no effect on CD4+ T-cells. However, it was associated with changes in CD4+ T-cell phenotype, warranting further investigation. Key words: adolescents, children, HIV, immunity, vitamin D There is increasing evidence that hypovita- minosis D is common in the general popu- 1,2 lation. Low dietary intake of vitamin D and reduced exposure to sunlight are probably the major risk factors. However many facets of this condition are still unclear, and recommenda- tions for correction of vitamin D deficiency vary widely.3–12 A high prevalence of hypovitaminosis D has been described in HIV-infected adults13,14 and children.15,16 HIV infection itself and antiretroviral (ARV) treatment may be responsible for alteration of vitamin D metabolism.17,18 Studies have shown a significant decrease in serum 25-hydroxyvitamin- D [25(OH)D] concentration in adults receiving non-nucleoside reverse transcriptase inhibitors (NNRTIs)19,20; however, a sure interference of ARV treatment on vitamin D metabolism has not been 21,22 confirmed in more recent studies. Because of the importance of vitamin D in bone health and the greater risk for bone disease in HIV-infected subjects, randomized controlled trials (RCTs) have been performed to test whether vitamin D Address for correspondence: Gian Vincenzo Zuccotti, Department of Pediatrics, L. Sacco Hospital, Università degli Studi di Milano, Milan, Italy; phone: +39 0239042258; fax: +39 0239042254; e-mail: gianvincenzo.zuccotti@unimi.it HIV Clin Trials 2013;14(2):51–60 © 2013 Thomas Land Publishers, Inc. www.thomasland.com doi: 10.1310/hct1402-51 51 52 HIVCLINICALTRIALS • 14/2 • MAR-APR2013 supplementation can improve vitamin D status and bone mineral metabolism in HIV-infected children and adolescents.23,24 Along with its effects on bone metabolism, vita- min D is an important modulator of the immune system. The vitamin D receptor (VDR) is found in high concentrations in activated T lymphocytes and in small amounts in monocyte/macrophage cells; B lymphocytes do not contain detectable amounts of VDR.25 Experimental studies have shown that the active di-hydroxylated metabolite of vitamin D [1,25(OH)2D] is able to skew the T-cell compart- ment into a more anti-inflammatory state, with inhibition of Th1 and Th17 cells and promotion of Th2 and T regulatory (Treg) subsets.26,27 In the context of HIV infection, in which Th1 sub- populations inhibit viral replication,28 any altera- tion of the Th1/Th2 balance would be of concern. Although all the biological effects of vitamin D are mediated by the 1,25(OH)2D, the 25(OH)D needs to be routinely quantified because of its longer half- life.29 However, HIV-infected subjects may have a defective 1α-hydroxylation of 25(OH)D.30 Thus, it is important to evaluate the effects of vitamin D supplementation in terms of both 25(OH)D and 1,25(OH)2D responses. Our study was aimed to test the hypothesis that oral supplementation with cholecalciferol (vitamin D3) is able to increase serum 25(OH) D and 1,25(OH)2D levels and to affect T-cell phenotype. METHODS Subjects Vertically HIV-infected children and young adults with stable disease,31 followed at the Depart- ment of Paediatrics – Luigi Sacco Hospital (Milan, Italy), were evaluated for inclusion into the study as of April 2011. Inclusion criteria were age ≤30 years and serum 25(OH)D concentration <30 ng/mL.29 Exclusion criteria were hyperparathyroidism, as detected by an intact serum parathyroid hormone (PTH) ≥65 pg/mL; Black ethnic group; supplementation with vitamin D in the previous 12 months; use of any treatment known to alter vitamin D status in the previous 6 months (excluding ARV); and any concomitant severe illness. Persons of African descent were excluded to avoid possible confounding due to racial differ- ences in vitamin D metabolism status, for which genetic explanations beyond possible skin color gradations have been identified.32 Patients with 25(OH)D <30 ng/mL with con- comitant PTH level above normal limits were not considered eligible for the present study and were referred to treatment. Study Design A repeated-measures parallel-group RCT was performed. Eligible subjects were allocated to receive oral cholecalciferol or placebo using a 4-block randomized design.33 Serum 25(OH)D, 1,25(OH)2D, PTH, calcium, urinary calcium, and CD4+ T cells were assessed 3 months before enrollment (–3 months), at baseline (0 months), and at each visit thereafter (3, 6, 9, and 12 months). T-lymphocyte VDR expression and Th1-, Th2-, Th17-, and Treg-lym- phocytes were measured at 0, 3, and 12 months. Adverse events were defined as those injuries related to or caused by the treatments under study. Patients and/or legal guardians were specifically asked about adverse events at each study visit. The main outcome was the frequency of hypovitaminosis D at the end of the trial, 3 months after the last dose. Twenty-five subjects per group gave a power of 83% to detect a reduction of 40% in the frequency of hypovitaminosis D in the vitamin D versus the placebo group at an alpha level of 0.05 (Fisher exact test). The study was approved by the Luigi Sacco Hospital Ethical Committee and registered as trial number 2011–00059354 in the Eudract Registry. Written informed consent was obtained from the subjects or their legal guardians if they were younger than 18 years. Intervention Patients were randomized to receive orally vita- min D3 100,000 IU (Dibase, ABIOGEN PHARMA SpA) or matching placebo oil suspended in 2 mL of olive oil in sealed plastic syringes labeled with the unique identification numbers. All the study participants, outcome assessors (laboratory techni- cians and immunologists), and personnel, except VITAMIN D3 SUPPLEMENTATION IN HIV+ YOUTH • GIACOMET ET AL 53 the pediatrician who administered the treatment, were blinded to it. Direct observed treatment (DOT) was adminis- tered at the study site at baseline and at months 3, 6, and 9 for a cumulative dose of 400,000 IU. Blood Collection Blood was collected by venipuncture using EDTA-containing Vacutainer tubes (Becton Dick- inson, Italy), and peripheral blood mononuclear cells (PBMC) were separated on lymphocyte separation medium (Organon Teknica, USA). Serum was obtained by centrifugation, and aliquot samples were stored at –80°C pending further analysis. Vitamin D and Related Measures Serum 25(OH)D was measured by immunoassay (25-hydroxy-vitamin D EIA; Immunodiagnostic Systems Ltd, Boldon, UK). The sensitivity of the assay was 5 nmol/L, with intra-assay precision ranging from 5.3% to 6.7% and inter-assay preci- sion from 4.6% to 8.7%. Serum 1,25(OH)2D was measured by radio- immunoassay (1,25-dihydroxy vitamin D RIA; Immunodiagnostic Systems Ltd). The sensitivity of the assay was 5 pmol/L, with intra-assay precision ranging from 11.9% to 20.0% and inter-assay preci- sion from 8.6% to 16.6%. Serum PTH was measured by immunoassay (Elecsys PTH Test System, Roche Diagnostics GmbH, Mannheim, Germany). The sensitivity of the assay was 1.2 pg/mL; intra-assay precision ranged from 2.6% to 6.5% and inter-assay precision from 1.1% to 4.1%. Serum and urinary calcium (spot sampling) were measured using standard laboratory techniques. T-Cell Immunophenotyping PBMCs were incubated for 18 hours in the pres- ence/absence of a pool of gag+env peptides (HIV). During the last 6 hours, 10 μg/mL of brefeldin A (Sigma-Aldrich, St. Louis, MO, USA) was added to cell cultures for cytokine analyses. We measured VDR expression on CD4+ T cells as previously described.34 Briefly, 0.5 x 106 PBMC were washed in phos- phate-buffered saline (PBS) and stained for CD4 FITC monoclonal antibodies (Beckman-Coulter, Fullerton, CA, USA) for 15 minutes at room temperature in the dark. Cells were fixed in 1% paraformaldehyde (Sigma-Aldrich, St. Louis, MO, USA) for 15 minutes at 4°C, washed, resuspended in 0.5% saponin (Sigma-Aldrich), stained for VDR antibodies (R&D Systems Inc., Minneapolis, MN, USA), and coupled to phycoerythrin (PE) (Innova Biosciences, Cambridge, UK). After a 45-minute incubation at 4°C in the dark, cells were washed and fixed in 1% paraformaldehyde. The follow- ing antibody panels were used to define Th1, Th2, Th17, and Treg cells: anti-CD4 PE-Cy7, anti-IL2 PE, anti-IFN-γ FITC (Beckman-Coulter, Fullerton, CA, USA), anti-Tbet PE-Cy5 (eBioscience, San Diego, CA, USA), anti-CD4 PE-Cy7 (Beckman-Coulter), anti-GATA3 eFluor660 (eBioscience), anti-IL-4 PE (R&D Systems Inc), anti-CD4 PE-Cy5 (Beckman- Coulter), anti-ROR-gamma PE, anti-IL17A FITC (eBioscience), anti-CD4 PE-Cy7, anti-CD25 PE (Beckman-Coulter), anti-Foxp3 PE-Cy5 (eBiosci- ence), and anti-IL10 FITC (R&D Systems Inc.). Stimulated PBMCs were washed in PBS (Euroclone, Pero, Italy) and stained with surface antibodies. PBMCs were then washed, fixed, permeabilized (FoxP3 Staining Buffer Set; eBiosci- ence), and stained with intracellular antibodies. Cells were finally washed with permeabilization buffer and acquired on an FC500 flow cytometer (Beckman Coulter, Brea, CA, USA). Th1 cells were identified as IFNγ- or/and IL2-secreting Tbet+ CD4+ T cells, Th2 cells were identified as IL4- or/ and IL10-secreting GATA3+ CD4+ T cells, and Th17 cells were identified as IL17-secreting RORγt+ CD4+ T cells. Data were analyzed by first gating on the lym- phocyte population as defined by forward and side light scatters. From this population, a single color CD4 or CD8 histogram was made, and T cells were selected and inserted into a 2-dimensional dot plot. Statistical Analysis Most variables were not normally distributed, and baseline values are reported as 50th, 25th, and 75th percentiles. Categorical variables are reported as the number of subjects with the char- acteristic of interest. Between-group comparisons of categorical variables were performed using Fisher exact test. The longitudinal changes of the outcomes of interest were evaluated using mixed 54 HIVCLINICALTRIALS • 14/2 • MAR-APR2013 linear regression models employing (1) treatment (categorical: 0 = placebo, 1 = vitamin D); (2) time [categorical: –3, 0, 3, 6, 9, and 12 months for 25(OH) D, 1,25(OH)2D, PTH, serum calcium, urinary cal- cium, and CD4+ T-cells and 0, 3 and 12 months for square-root transformed VDR, Th1-lymphocytes, Th2-lymphocytes, Th1:Th2 ratio, Th17-lympho- cytes, Treg-lymphocytes, and Th17:Treg ratio], (3) a Treatment x Time interaction (Categorical x Categorical), and (4) the baseline value of the outcome as covariates and the patient as random effect.35 Standardized residuals and level-1 residu- als were evaluated to check model fit.36Although model fit improved substantially after square-root transformation of VDR, Th1-lymphocytes, Th2- lymphocytes, Th1:Th2 ratio, Th17-lymphocytes, Treg-lymphocytes, and Th17:Treg ratio, we none- theless used robust 95% confidence intervals to relax the assumption of homoscedasticity of level-1 residuals. Although not strictly needed for the other outcomes, we calculated robust 95% confidence intervals for all models for reasons of uniformity. Using maximum likelihood estima- tion, the mixed models allowed us to manage missing data under the assumption of missing at random (MAR).37 The analysis was intention-to- treat (ITT) for all outcomes. Statistical analysis was performed using Stata 12.1 (Stata Corp, College Station, TX, USA). RESULTS The selection of patients is depicted in Figure 1. Out of 90 vertically HIV-infected patients aged 8 to 26 years who underwent vitamin D status assessment 3 months before starting the trial (April 2011), 79 (87.7%) had vitamin D insuf- ficiency [25(OH)D < 30 ng/mL]. Eleven subjects were excluded because of Black ethnicity; 11 patients with hypovitaminosis D and hyperpara- thyroidism were not eligible for randomization and were referred to treatment. Of the remaining 57 patients, 52 agreed to par- ticipate in the trial and were randomized into the vitamin D (n = 26) or placebo group (n = 26). One patient in the vitamin D group developed epilepsy and 1 in the placebo group moved away from Milan before starting the study, so 50 patients were actually evaluated. Baseline patients’ characteristics are given in Tables 1 and 2. At enrollment, the 2 groups were comparable in both categorical (Table 1) and continuous measurements (Table 2). Twenty- one subjects (84%) in the vitamin D group and 22 patients (88%) receiving placebo had been under stable ARV therapy for at least 6 months before enrollment and all of them had undetectable HIV RNA levels (Table 1). The majority (52%) of the patients were receiving an NNRTI-based regimen; 17 (34%) subjects were receiving a protease inhibi- tor–based regimen. Only a few patients were naïve to ARV (Table 1). Vitamin D and related measures as well as immunological features were also comparable in the 2 groups (Table 2). Forty-four patients performed all scheduled visits. Given the DOT strategy, we reported no missed doses of vitamin D. Four patients in the vitamin D group missed 1 visit, while 2 in the placebo group were lost at follow-up at 3 and 6 months. Only 65% of the planned urinary calcium measurements were performed, whereas all other measurements were obtained in virtually all cases (from 98% to 100%). No patients were excluded from analysis at any time point, and the analysis was ITT. Vitamin D and Related Outcomes Changes in vitamin D and related outcomes dur- ing the study are reported in Table 3. There were no changes in 25(OH)D, 1,25(OH)2D, and PTH levels at 0 (baseline) versus -3 months. At the first assessment after baseline (3 months), 25(OH)D levels were higher and PTH lower in the vitamin D group, but no between-group difference in 1,25(OH)2D levels was detected. At the second assessment (6 months), an increase in 1,25(OH)2D concen- tration was evident in the vitamin D group in addition to the increase in 25(OH)D and decrease in PTH. At the third (9 months) and fourth (12 months) assessments, these changes were still evident and stable. Serum calcium showed a tendency to increase in the vitamin D group at 3 months, but the effect size was very modest and not replicated at the following time points. No changes in urinary calcium levels were seen in the 2 groups at any time. However, the changes in urinary calcium concentration need to be interpreted with caution because of the high number of missing data (35%). VITAMIN D3 SUPPLEMENTATION IN HIV+ YOUTH • GIACOMET ET AL 55 n = 90 HIV-infected patients (April 2011) n = 79 25(OH)D < 30 ng/mL n = 11 25(OH)D ≥ 30 ng/mL n = 57 potentially eligible n = 11 excluded because of hyperparathyroidism n = 11 excluded because belonging to Black ethnic group n = 52 accepted to participate n= 5 refused to participate n = 26 randomized to vitamin D n = 26 randomized to placebo n = 25 actually enrolled n = 1 patient developed epilepsy Figure 1. Selection of patients. Figure 2 plots 25(OH)D values in the 2 groups during the study period. (These values were pre- dicted from the mixed regression model of Table 3 and take into account missing data.) The frequency of vitamin D insufficiency at 12 months (main outcome) was 60% in the placebo group and 20% in the vitamin D group (P = .007). Data were calculated on 48 patients, because 2 patients in the placebo group were lost to follow-up during the study (see above). Immunological Outcomes No changes were observed in CD4+ T-cell count or percentage in the 2 groups (Table 3). A decrease of the Th17:Treg ratio (square-root transformed) was detected at 3 months in the vitamin D versus the placebo group, but it became nonsignificant at 12 months (Table 4). Safety No adverse events were reported in either group. n = 25 actually enrolled n= 1 moved away 56 HIVCLINICALTRIALS • 14/2 • MAR-APR2013 Table 1. Baseline categorical measurements of the participants Vitamin D Placebo Category (n = 25) (n = 25) CDC clinical category N+A 169 B+C 916 Current CDC clinical category N+A 25 25 B+C 00 CDC immune category 185 2 1010 3 710 Current CDC immune category 1 1921 264 300 Undetectable HIV-RNA (< 37 copies mL) Treated with NNRTIs Treated with PIs Naïve 18 21 13 13 8 9 4 3 Note: CDC = Centers for Disease Control and Prevention; NNRTIs = non-nucleoside reverse transcriptase inhibitors; PIs = protease inhibitors. Table 2. Baseline continuous measurements of the participants Age, years 25(OH)D, ng/mL 1,25(OH)2D, pg/mL PTH, pg/mL Serum calcium, mg/dL Urinary calcium, mg/dL CD4+ T cells, cells/mm3 CD4+ T cells, % Th1 lymphocytes, % Th2 lymphocytes, % Th1:Th2 ratio Th17 lymphocytes, % Treg lymphocytes, % Th17:Treg ratio Lymphocyte VDR, % Vitamin D (n = 25) P50th (P25–P75th) 20 (14–23) 15 (12–19) 97 (86–113) 40 (25–48) 9 (9–10) 11 (7–21) 663 (507–796) 37 (27–40) 0.1 (0.1–0.4) 0.5 (0.2–1.1) 0.2 (0.1–0.7) 0.3 (0.1–0.3) 0.3 (0.2–0.4) 1.0 (0.4–1.5) 6.9 (2.4–11.9) Placebo (n = 25) P50th (P25–P75th) 18 (15–23) 15 (14–19) 93 (70–108) 37 (32–41) 10 (9–10) 12 (7–30) 673 (601–773) 34 (25–40) 0.1 (0.0–0.2) 0.7 (0.3–1.1) 0.2 (0.0–0.4) 0.2 (0.0–0.3) 0.4 (0.2–0.7) 0.2 (0.0–0.7) 8.6 (4.2–13.7) Note: Values given as % (range) or n (range). Px = xth percentile (P50 is the median); PTH = parathyroid hormone; VDR = vitamin D receptor. Table 3. Changes of vitamin D and related outcomes for vitamin D vs placebo group Change vs –3 months for vitamin D vs placebo Month 25-OH-D, ng/mL 1,25(OH)2D, pg/mL PTH, pg/mL Serum calcium, mg/dL Urinary calcium, mg/dL CD4+ T cells, cells/mm3 CD4+ T cells, % 0 3 6 9 0.0 [−3.0,3.0] 13.0 [−4.4,30.4] 13.6 [−3.3,30.5] 27.9** [10.1,45.8] 23.9* [2.7,45.1] 27.0** [10.0,44.0] –3.6 [−10.9,3.7] –10.3** [−17.6,–3.0] –12.4** [−20.3,–4.5] –13.3** [−23.2,–3.4] –0.1 [−0.3,0.1] 0.2* [0.0,0.4] 0.1 [−0.2,0.3] 4.0 [−3.8,11.9] 0.3 [−0.2,0.9] –1.2 [−6.1,3.6] –5.7 [−19.3,8.0] 2.5 [−6.2,11.1] 5.6 [−1.6,12.8] 7.4 [−0.8,15.5] –72.4 [−186.8,42.1] 7.0 [−87.1,101.2] –1.4 [−4.6,1.8] 0.9 [−2.1,3.9] –2.7 [−8.3,2.8] –1.2 [−5.1,2.6] 12 12.5*** [5.9,19.0] –9.8* [−18.2,–1.5] –77.0 [−180.2,26.3] 4.8* [1.5,8.1] 11.1** [3.8,18.3] 8.9*** [5.9,11.9] 19.9 [−119.6,159.3] 48.6 [−68.9,166.0] 58.1 [−114.5,230.7] Note: 95% robust confidence intervals in brackets. Point estimates and confidence intervals were obtained from a mixed linear model. *P < .05. **P < .01. ***P < .001. 50 45 40 35 30 25 20 15 10 Placebo Vitamin D VITAMIN D3 SUPPLEMENTATION IN HIV+ YOUTH • GIACOMET ET AL 57 VDRa Month 3 0.6 [−0.2, 1.4] Month 12 0.5 [−0.2, 1.3] Th1a –0.3 [−0.6, 0.1] –0.1 [−0.7, 0.5] Th2a –0.1 [−0.4, 0.2] 0.3 [−0.1, 0.6] Th1:Th2a –0.5 [−1.0, 0.1] –0.7 [−2.2, 0.8] Th17a –0.2 [−0.5, 0.0] 0.1 [−0.2, 0.5] Trega 0.1 [−0.2, 0.4] 0.7 [−0.2, 1.5] Th17:Trega –0.6 [−1.0,–0.1] –0.4 [−0.8, 0.1] −3 0 3 6 9 12 Months Figure 2. 25(OH)D levels during the study: point estimates and confidence intervals obtained from a missed linear model. Table 4. Changes in immunological outcomes for vitamin D vs placebo group Change vs 0 months for vitamin D vs placebo Note: 95% robust confidence intervals in brackets. Point estimates and confidence intervals were obtained from a mixed linear model. aAfter square-root transformation. DISCUSSION This is the first RCT aimed to evaluate changes in 25(OH)D, 1,25(OH)2D, and PTH levels in HIV-infected youth undergoing vitamin D3 supplementation over a long period (12 months) and, concomitantly, to test whether such supplementation may affect the T-cell phenotype. Given the risk of poor adherence to vitamin suppleme