Low vs. Moderate to High Dose Vitamin D for Prevention of COVID-19

The purpose of this study is to compare the risks of COVID-19 in individuals from Chicagoland communities randomized to low (400 IU/day) vs. moderate (4,000 IU/day) or high (10,000 IU/day) dose vitamin D.

This study will recruit 2,000 subjects from the Chicagoland area to participate in the study at University of Chicago Medicine (UCM) and at Rush University (RU). Subjects will be invited to volunteer to participate in the study through a variety of forms of community outreach (flyers, presentations, emails, etc.). The sample size was chosen to have at least 80% power to detect a 25% or larger decrease in the hazard of developing COVID-19 between study arms at p<0.05 assuming a 30% baseline COVID-19 incidence rate over the study period, in line with a 3% per month incidence. Subjects will take a daily dose of vitamin D, attend laboratory appointments for testing, and answer surveys over 9 months after enrollment. Our overall aim is to compare the risks of COVID-19 in adults in the Chicagoland area at increased risk of COVID-19 randomized to low (400 IU/day) vs. moderate (4,000 IU/day) or high (10,000 IU/day) dose vitamin D. Our specific aims are: Aim 1: To compare the risk of developing COVID-19 in adults in the Chicagoland area at increased risk of COVID-19 randomized to low vs. moderate or high dose vitamin D. Subjects will specify their preference for moderate versus high dose vitamin D and then be randomized between low dose vitamin D and their preferred moderate or high dose alternative. We hypothesize that moderate and high dose therapy will reduce rates of COVID-19 compared to low dose therapy because vitamin D will reduce symptomatic infection that prompts testing for COVID-19. Our primary analysis will pool subjects randomized to the moderate or high dose and compare them to low dose subjects, with secondary analyses comparing low to moderate and low to high. Aim 2: To compare COVID-19 seroconversion in adults in the Chicagoland area at increased risk of COVID-19 randomized to low vs. moderate or high dose vitamin D. As we suspect the main effect of vitamin D may be to decrease symptomatic disease we hypothesize that there may be little or no difference in seroconversion. Aim 3: To compare COVID-19 outcomes (hospitalization, ICU stay, ventilator use, death) in adults in the Chicagoland area at increased risk of COVID-19 randomized to low vs. moderate or high dose vitamin D. Consistent with several recent observational analyses, we expect higher doses to improve COVID-19 outcomes. Additional aims examine the relationship between vitamin D and other markers of immune response: Aim 4: Create a biobank of PBMCs from a panel of 500 individuals with well characterized levels of vitamin A and D. We propose to cryopreserve PBMCs derived from ~20 ml of blood to test the impact of vitamin A and D levels in the regulation of immune responses. Aim 5: Evaluate the impact of vitamin A and D levels on immune function. We will perform immunophenotyping of PBMC using a panel of 30 antibodies that allow characterizing the different immune cell populations found in circulation, as well as their functional status. Using these data we will ask if there is an association between the prevalence of certain immune cell population or functional potential and vitamin A and D levels. Aim 6: Evaluate the impact of vitamin A and D levels on PBMC gene expression levels. We will perform transcriptional profiling of PBMC from a panel of 100 individuals enriched for individuals on the two extremes of the distributions of vitamin A and D levels. These data will be used to identify changes in gene expression levels that differ between individuals showing low- vs high-vitamin A/D levels. Aim 7: To test the hypothesis that oral vitamin D3 intervention will inhibit the renin- angiotensin system and reduce the incidence and/or severity of COVID-19 infections Blood labs will be taken from subjects at intake, 3 months, 6 months and 9 months. Intake labs will measure baseline calcium and PTH levels, COVID-19 antibodies, and vitamin D. Intake labs for a subset of University of Chicago Medicine (UCM) subjects will additionally measure vitamin A, vitamin D-binding protein (DBP), RAS components (Renin, Ang II, soluble ACE and ACE2), bradykinin, inflammation markers (IL-6, TNF-a, IL-1b and CRP), and SARS-CoV-2 antibodies against S-protein and N-protein. 3-month labs for all subjects will measure PTH, calcium, and vitamin D for monitoring of the intervention's safety and outcomes, and COVID-19 antibodies for monitoring of the outcome. 6-month and 9-month labs for all subjects will measure PTH, calcium, and vitamin D for monitoring of the intervention's safety and COVID-19 antibodies for monitoring of the outcome. Vitamin A will be measured for a subset of UCM subjects who were high at baseline. Throughout the study period we will tell subjects about calcium, PTH and vitamin D lab results only if they suggest a safety concern for which we would like them to seek care. Subjects will be told if they test positive for COVID-19 antibodies. Except in the instance of a safety concern, subjects will be blinded to their vitamin A & D levels, calcium and PTH throughout the study to ensure protocol compliance. At the end of the study, we will inform subjects of their vitamin D levels. Electronic Medical Records will be reviewed for predictors and indicators of COVID-19 and for severity of infection if subject has had a positive COVID-19 test result. Subjects will be asked to complete a web-based survey at intake, 3, 6 and 9 months. The intake survey will collect baseline data on subjects' current medications and supplements, sun exposure, Fitzpatrick skin type, exercise, diet and sleep habits, possible COVID-19 symptoms, and other influenza-like symptoms. Subjects will also be asked to report suspected exposure to the virus, occupation, and demographic information. Subjects will also be asked to evaluate risks of exposure. Follow-up surveys at months 3, 6, and 9 will ask the participant (or proxy) whether the participant has had a clinically confirmed diagnosis of COVID-19, and the date if so, ask about rates of study medication adherence, and assess for changes in the intake questions about use of other supplements, sun exposure, diet, exercise, and COVID-19 exposures. If the patient is reported to have had COVID-19, we ask about severity, including hospitalization and duration, ICU-use and duration, and need for mechanical ventilation and duration, death and COVID-19 symptom severity.

2,000-person two-arm, double-blinded randomized controlled trial, with half the subjects randomized to low dose vitamin D therapy (400 IU/day), which will serve as the control group, and half to moderate (4,000 IU/day) or high (10,000 IU/day).

Subjects in this arm will be randomized to receive low dose vitamin D therapy (oral, 400 IU/day) for 9 continuous months.

Subjects in this arm will be randomized to receive low dose vitamin D therapy (oral, 4,000 IU/day) for 9 continuous months.

Subjects in this arm will be randomized to receive low dose vitamin D therapy (oral, 10,000 IU/day) for 9 continuous months.

Half the subjects will be randomized to the low dose vitamin D therapy (400 IU/day), which will serve as the control group, and half to moderate (4,000 IU/day) or high (10,000 IU/day). Study participants will have the option between being randomized to the low versus moderate or the low versus high dose arms.

SARS-CoV-2 infection as measured by patient report of clinically confirmed COVID-19 (or viral PCR when available)

For this outcome, hazard models will be employed to assess the effect of each vitamin D dosing strategy on the outcome. We will first develop hazard ratios for between-group analyses on the primary outcome using log-rank tests, and then develop Cox proportional hazard models to model the hazard function on a set of covariates including but not limited to, moderate or high vitamin D dose, baseline vitamin D levels, age, gender, race, ethnicity, sun exposure, sleep habits, exposure of cohabitants, job type, and study site. We will also control for randomization date to adjust the underlying hazard function for COVID-19 prevalence over time. While our primary analysis will pool subjects randomized to either the moderate or high dose and compare them to low dose subjects, we will also perform secondary analyses comparing low to moderate and low to high and additional analyses that use post randomization vitamin D levels as time varying covariates.

We will use longitudinal mixed effects models to analyze this data that will provide measures only at the time of blood draws (3, 6 and 9 months). Covariates and secondary analyses will mirror those in the primary outcome measure.

This will be measured by a binary indicator of whether the patient is hospitalized after COVID-19 in the first quarterly survey after COVID-19 is reported and will be analyzed with a generalized linear model with covariates as in the secondary outcome measure.

This will be measured by a binary indicator of whether the patient is admitted to the ICU after COVID-19 in the first quarterly survey after COVID-19 is reported and will be analyzed with a generalized linear model with covariates as in the secondary outcome measure.

This will be measured by a binary indicator of whether the patient receives mechanical ventilation after COVID-19 in the first quarterly survey after COVID-19 is reported and will be analyzed with a generalized linear model with covariates as in the secondary outcome measure.

This will be measured by a binary indicator of whether the patient dies after COVID-19 in the first quarterly survey up to 9 months after COVID-19 is reported and will be analyzed with a generalized linear model with covariates as in the secondary outcome measure.

Subjects are able to participate if they: Are 18 years or older. Live or work in the Chicagloland area (Illinois counties: Cook, Lake, McHenry, DuPage, Kane, Kendall, Grundy, Will, and Kankakee; Indiana counties: Lake and Porter). Are interested in vitamin D as a potential preventive measure against COVID-19 in which they self-administer a daily dose of vitamin D during the 9-month study period. Are willing to attend the laboratory for drop-in appointments at UChicago Medicine or Rush University Medical Center every 3 months at 4 time points over a 9-month period for blood draws measuring COVID-19 antibodies, calcium, vitamin D and PTH levels. Are willing to complete self-report measures at 4 time points over the course of 9 months by completing a 15-minute survey at intake by telephone or via web and 10-minute web-based follow-up surveys. Subjects are excluded from study participation if they: Report ever having a positive COVID-19 PCR test result Report being pregnant, planning to become pregnant, and/or report breastfeeding during the study period. Report a history of chronic kidney disease, including a history of abnormal GFR and/or creatinine. Report a history of hyperparathyroidism. Report a history of increased falls. Report a history of hypercalcemia. Report a history of gastrointestinal absorptive disorders, including having undergone bariatric surgery. Report a history of kidney stones (1 in past year or 2 in lifetime). Report already taking more than 400 IU of vitamin D daily as recommended by their health care provider, excluding multivitamins and excluding supplements that include vitamin D and calcium together. Report taking D2. Report a history of sarcoidosis. Screen positive for hypercalcemia during the initial blood test or follow-up blood tests. Screen positive for primary hyperparathyroidism during the initial blood test. Screen positive for COVID-19 antibodies during the initial blood test. Have vitamin D levels of >100ng/mL at study start, or >250ng/mL during follow-up labs. Are unwilling to provide blood samples during quarterly blood tests. Are unwilling to take daily vitamin D.

Grant WB, Lahore H, McDonnell SL, Baggerly CA, French CB, Aliano JL, Bhattoa HP. Evidence that Vitamin D Supplementation Could Reduce Risk of Influenza and COVID-19 Infections and Deaths. Nutrients. 2020 Apr 2;12(4):988. doi: 10.3390/nu12040988.

Martineau AR, Jolliffe DA, Hooper RL, Greenberg L, Aloia JF, Bergman P, Dubnov-Raz G, Esposito S, Ganmaa D, Ginde AA, Goodall EC, Grant CC, Griffiths CJ, Janssens W, Laaksi I, Manaseki-Holland S, Mauger D, Murdoch DR, Neale R, Rees JR, Simpson S Jr, Stelmach I, Kumar GT, Urashima M, Camargo CA Jr. Vitamin D supplementation to prevent acute respiratory tract infections: systematic review and meta-analysis of individual participant data. BMJ. 2017 Feb 15;356:i6583. doi: 10.1136/bmj.i6583.

Meltzer DO, et al. "Association of Vitamin D Deficiency and Treatment with COVID-19 Incidence. medRxiv. 2020 May 8.

McCullough PJ, Lehrer DS, Amend J. Daily oral dosing of vitamin D3 using 5000 TO 50,000 international units a day in long-term hospitalized patients: Insights from a seven year experience. J Steroid Biochem Mol Biol. 2019 May;189:228-239. doi: 10.1016/j.jsbmb.2018.12.010. Epub 2019 Jan 4.