Subscribe to the OSS Weekly Newsletter!

The COVID Science Express: Therapeutic Blood Transfusions and Tuberculosis Vaccine Protection

Hypotheses are quickly put to the test as countries gear up for clinical trials on the use of convalescent plasma and a repurposed tuberculosis vaccine

A weekly explanation of the emerging science behind COVID-19 and its infectious agent, SARS-CoV-2.

Blood from people who have recovered from COVID-19 could theoretically help those struggling with the infection

Our blood might become an even hotter commodity than usual if an extensive clinical trial that is about to begin delivers a positive answer. The trial hinges on a promising, all-natural substance: our antibodies.

Antibodies are Y-shaped molecules produced by a subset of our white blood cells. They are fitted with tips that will specifically bind to a particular virus or bacterium, the way that a key fits a lock. When antibodies bind to a virus, they can kill the virus directly or recruit other blood cells to come in and finish off the invader. We start producing antibodies when our body detects an invader and they stick around for a while after the infection is over. So if you have recovered from COVID-19, could we take some of your antibodies and put them in the body of someone who is struggling with the infection, as a way to reinforce their troops? Maybe. It may sound novel but it’s actually a fairly old medical intervention.

The concept is called “passive immunity” and it was used for a long time as an attempt to curb outbreaks of measles, mumps, and polio. Blood from people who had survived the 1918 influenza pandemic (commonly known as the Spanish flu) was given to people who were suffering from pneumonia due to the disease, and there is modern evidence that these transfusions might have lessened the number of deaths. This rather crude therapy was eventually relegated to an occasional supporting role as we developed vaccines, potent drugs, and even therapeutics based on very specific antibodies. But this idea of therapeutic blood transfusions is brought back to life each time we deal with a nasty epidemic and we need a quick and dirty solution.

Ebola, Lassa fever, Bolivian and Argentinian hemorrhagic fevers, as well as a number of severe respiratory viral infections like SARS all urgently pushed doctors to study passive immunity to see if they could improve the odds of their patients. Unfortunately, while many of these studies report encouraging results (fewer deaths, faster recovery), they are usually poorly done and uninformative. Many study participants ended up receiving better care and many different drugs, making it hard to evaluate the contribution of the antibodies themselves, if any. A systematic review of these blood transfusions in the context of respiratory viral infections concluded that the “studies were commonly of low or very low quality, lacked control groups, and at moderate or high risk of bias.” That’s not good.

Will transfusions of blood products from survivors of COVID-19 improve the lives of people struggling with it? We don’t know but a fairly massive clinical trial is about to begin in Canada to address this question properly. Forty Canadian hospitals have already jumped aboard to participate, and the trial is aiming to test this approach on 1,000 patients, two-thirds of whom will receive standard of care coupled with an antibody-rich blood fraction from survivors while a third will serve as a control group and only receive standard of care. The part of the blood that will be administered is called plasma: it is a yellow-tinted liquid full of antibodies. Plasma from people who have recovered from the infection is known as “convalescent plasma.” If this therapy is shown to work, we can foresee a future in which these powerful antibodies are purified by pharmaceutical companies and given in higher doses, which would make them safer and more potent. For now, we must wait six to ten months before the definitive results of this trial are known.

A live vaccine against tuberculosis will be tested in healthcare workers following a contentious analysis that reports an unexpected link

Because of the severity of this pandemic, medical doctors and scientists are looking for any crumb of evidence that might put us on stronger footing, any interesting observation that could reveal a chink in the coronavirus’ armour or a reinforcement in ours. This is, after all, how science works. We notice something and think, “How strange… I wonder if…”, and we test that hypothesis with the necessary rigour to see if we might learn something useful. The gravity of the pandemic has led to a compression of this process, where the “how strange…” moment is immediately followed by a boots-on-the-ground trial, and this is how we get to the BCG vaccine story.

The BCG vaccine (short for Bacille Calmette-Guérin) was developed in the early 1900s by two French bacteriologists to protect people from tuberculosis. The disease itself is caused by a bacterium, M. tuberculosis, which infects the lungs. The vaccine is made from a related bacterium, M. bovis, which normally infects cattle and which was essentially “defanged” by growing it in a lab for a period of 13 years. The twist is that the BCG vaccine is not automatically given to children in every country. As tuberculosis became less and less common, many developed countries limited or stopped the use of the vaccine. For example, this vaccine was phased out in Canada in the 1960s and 1970s and is not routinely recommended here except for First Nations and Inuit people, in whose communities tuberculosis is more common. So what does a vaccine against a bacterial infection have to do with COVID-19, a viral infection?

An assistant professor with a background in electrical engineering and who studies animal brains wrote a manuscript with a technician and a team of students reporting an unexpected association. In countries where everyone is immunized with the BCG vaccine, the team found that there were fewer cases of COVID-19 and fewer deaths related to the infection. This manuscript has not been officially reviewed by other scientists and has not yet been accepted for publication by a journal; rather, as is being done more and more to speed up knowledge transfer in this pandemic, it was uploaded by its authors onto a “preprint” site, the equivalent of publishing a blog post on a platform read by scientists. It’s easier to get the word out in this way but you are also temporarily bypassing most of the quality control measures useful to a healthy research environment. Nonetheless, this glimmer of hope, shimmering faintly around the BCG vaccine, was enough to fast-track a number of clinical trials to test this in healthcare workers. Certain countries like the Netherlands and Australia are now giving this vaccine to their healthcare workers and comparing it to a placebo to see if the real deal might have a protective effect.

It may all sound implausible, but there is evidence that this particular vaccine may gift us with a broader protection against all sorts of viruses and bacteria. Still, the paper showing a link between the BCG vaccine and COVID-19 has received its fair share of criticism (as all scientific studies must), including from Ph.D. students at the McGill International Tuberculosis Centre. The reported association is based on information downloaded March 21st, but anyone who follows the news is acutely aware that the number of confirmed cases changes every day. Early data can make a country look like it was spared, but check in a week later and it will look quite different. Moreover, the number of confirmed cases is not an accurate measure of who has the infection. Not every person with COVID-19 gets counted because not everyone gets tested. Different countries have different testing strategies and different access to testing kits, which make comparisons between countries tentative. Finally, it’s important to bring up that old chestnut, that a correlation between two things does not necessarily mean that one causes the other. It could be that the other causes the one, that both are caused by a third thing, or that the association exists purely by chance. The authors should be aware of this and yet they have used language that implies causation in their manuscript (“Our data suggests that BCG vaccination seem [sic] to significantly reduce mortality associated with COVID-19”). An association can generate a hypothesis which can be tested, but it would be foolish to report that the BCG vaccine itself reduces anyone’s chance of dying from COVID-19. We simply do not know.

A crisis requires swift action and the smallest seed of evidence will be spotlit in the hope that it can quickly germinate into a solution. It’s not unlike lottery tickets. You invest in a whole bunch of them, knowing full well that most of them will be duds, yet hoping that one of them will change the course of your life. Will the BCG vaccine be a winning ticket? The odds are not in its favour, but let’s wait and see.

Take-home message:
- Blood from people who have recovered from COVID-19 could theoretically help those struggling with the infection, a hypothesis that will be tested in a large Canadian clinical trial
- A live vaccine against tuberculosis (the BCG vaccine) will be tested in healthcare workers following a contentious analysis that reports an unexpected link between countries where the vaccine is universally administered and fewer cases of and deaths from COVID-19


Leave a comment!

Back to top