Talking Vitamin D and COVID-19 With Professor Adrian Martineau

Vitamin D – "the sunshine vitamin" – has received increasing attention throughout the course of the COVID-19 global pandemic, as questions mount relating to its potential protective role against respiratory infections.

The majority of our vitamin D intake is obtained via sun exposure. But in winter months sunlight is sparse in the UK, and consequently vitamin D levels are generally lower. In parallel, respiratory infections are more common.

In the context of COVID-19, there is suggestive evidence that vitamin D could serve a protective role. Populations that are more likely to develop the most severe forms of the disease, including elderly people, overweight people, or individuals of Black or Asian ethnicity, are also populations that are most often vitamin D deficient. Whether a direct link exists between the two factors remains to be definitively answered.

The CORONAVIT trial is an ongoing, prospective, Phase III randomized clinical trial in the UK that has recruited over 6,000 participants. It aims to determine whether a "test-and-treat" method to identifying and correcting vitamin D deficiency in individuals reduces their risk of developing COVID-19 and other respiratory infections. Recruitment is now complete, and an interim analysis is to be conducted in the near future.

Professor Adrian Martineau at Queen Mary University of London is the lead researcher on the trial. In this interview, we ask Martineau about the rationale behind this study, what impact the UK COVID-19 vaccine rollout may have on the power of the trial and the potential impact of the results. 

Molly Campbell (MC): Could you please talk to me about the rationale behind the CORONAVIT trial? 

Adrian Martineau (AM): Laboratory studies have demonstrated a role for vitamin D in responses to viruses other than SARS-CoV-2, specifically boosting a range of innate immune responses to viruses, such as rhinovirus and influenza, but simultaneously downregulating potentially harmful inflammatory responses. This is quite a good combination, potentially, for SARS-CoV-2, because we know that patients tend to do badly due to poorly regulated inflammation, as illustrated by the fact that dexamethasone helps them, which is acting primarily via an anti-inflammatory mechanism.

There is a slew of observational clinical studies that have linked lower vitamin D levels, both to susceptibility to COVID-19 and also to increased severity of COVID-19. However, their results have to be interpreted with caution because the associations demonstrated might be due to confounding or to reverse causality for those studies which are not prospective. In other words, it is possible in studies, which are cross sectional, that the disease itself could cause low vitamin D levels, as opposed to the other way around. Therefore, it is important to go forward to clinical trials.

In terms of clinical trials, there are now over 40 clinical trials that have investigated the effectiveness of vitamin D to prevent the key respiratory infections other than COVID-19. These have yielded variable results, with some showing strong protection, and others showing no effect whatsoever. It has been quite a conundrum in the field as to why that variability should exist.

We did a meta-analysis published in the BMJ back in 2017, which suggested that vitamin D work best when it was given in daily doses to people who had a low vitamin D level to start with. That was based on ~11,000 patients in 25 trials. We recently updated that work to include over 30,000 patients in 43 trials and we continue to see a signal for daily dosing as opposed to dosing that was more spread out, but we did not replicate our original finding that the benefit was stronger in those with lower baseline levels.

The bottom line is that there is a suggestive story from the basic science, a suggestive story from the observational epidemiology, but rather inconclusive data from trials in viruses other than SARS-CoV-2.

Essentially, we are in a situation of equipoise, where nobody is quite sure whether it works or not, and this has caused two different reactions. Some individuals say, 'If there is a fighting chance that it [vitamin D supplementation] will work, what harm will it cause? Let's roll it out because we're in a public health emergency', whereas others are keen to avoid recommending something if we do not know that it works. I used to belong to the former camp, but when I arranged to provide free vitamin D to colleagues on the frontline back in March of last year, the offer was turned down, on the grounds that the evidence base was too weak. This persuaded me that trials would be needed – so we set one up.

MC: Could you please discuss the structure of the trial?

AM: The CORONAVIT trial has recruited 6200 participants who are randomized to receive either the standard government recommendation of 400 units of vitamin D per day, or vitamin D testing, with an offer of one of two higher dose regimens if levels are found to be low: 800 units per day, or 3200 units per day. We want to know, if it [vitamin D supplementation] does cause benefit, which dose is better?

We sent out the first doses in November and we are following people up. It is a six-month study due to report in the summer. We are conducting an interim analysis in March, and if there is a strong protective effect then we would be able to terminate the trial early and publicize that data.

MC: Could the COVID-19 vaccine roll-out in the UK impact the trial?

AM: We collect information on vaccination. You are right that it could potentially decrease the power of the trial by reducing the event rate. However, like the cat that jumped out the seventh-floor window, we are trying to land on our feet by adapting the trial to determine whether vitamin D supplementation could boost the effectiveness of vaccination.

Recently I was one of a team that published a paper showing that vitamin D supplementation boosts immune responses to varicella zoster virus (VZV) vaccination. We have shown that vitamin D supplementation in older adults (who have a more dysregulated immune system) reduces inflammation, and the reduction in inflammation enables the immune system to produce stronger antigen-specific responses to vaccination. We are hoping that vitamin D might improve the immunogenicity of SARS-CoV-2 vaccination, and we are investigating that as a sub-study within the trial.

MC: Interestingly, the populations that are developing more severe COVID-19 are also those at higher risk of vitamin D deficiency. Please can you talk about this? 

AM: It is quite striking that the people who get the most severe cases of COVID-19 are those that are at the highest risk of D deficiency, namely elderly people, individuals who are not able to get out into the sun, overweight people and people from Black and Asian ethnic minority backgrounds.

The mechanisms by which they are vitamin D deficient are variable. In obesity, vitamin D deficiency is thought to be an issue relating to volume of distribution. People who are overweight have a larger fat reservoir, which potentially diverts vitamin D away from circulation. Black and Asian individuals are at high risk of deficiency, partly because the melanin in the skin blocks the UV to a greater extent, and so reduces the ability of the skin to make vitamin D where sunshine is limited, as it is in the UK. In some South Asian communities, there may also be dietary influences, for example vegetarian diets.

We also know that the skin of older people is less efficient at making vitamin D, for a given amount of UVB, than the skin of younger people.

This association between vitamin D deficiency and COVID-19 severity may be a coincidence. The vitamin D deficiency might be a bystander, and there are plenty of reasons to think that. For example, in obesity there is often increased inflammation, which could be the main factor contributing to the more severe COVID-19 cases in this population. That is why we really need to do a controlled trial to see if it [vitamin D] is a player or not.

MC: Are you able to shed light on whether the study population are representative of the general population?

AM: Compared to the general population, the following groups are underrepresented in the trial: males (more than 50% of the trial are female), younger people (the median age of participants is 60), and regrettably, ethnic minorities are underrepresented in our study. 14% of the UK population nationally are of ethnic minority origin, whereas our study comprises 6% ethnic minority groups, despite our best efforts to reach out equally to all communities.

Adrian Martineau was speaking to Molly Campbell, Science Writer for Technology Networks.