How do we know vaccines won’t have long-term safety risks?

Examine, a weekly newsletter written by national science reporter Liam Mannix, is sent every Tuesday. The latest instalment is below. Sign up to get it in your inbox here.

Whenever I talk to people about vaccines, the same questions come up over and over again: have we tested them for long enough?

And how do we really know they are going to be safe, long term?

It seems like a compelling point. The first vaccine trials started about 16 months ago. Sure, we can be confident in their short-term safety … but how do we know in a decade or two we’re not going to start turning into horrifying lizard creatures?

The answers are surprising. Vaccine trials are shorter than you might think – but time and time again, they produce safe and effective vaccines. And we probably have better data about COVID-19 vaccines at this point than almost any other vaccine we have developed.

And yes, it is scientifically impossible to know if a lizard-creature future awaits us in a decade. But basic science, evidence from every other vaccine we’ve developed and critical thinking can give us an awful lot of confidence.

How long should you trial a vaccine for, anyway?

One telling of the “we don’t know if the COVID-19 vaccines are safe” idea goes like this: typically, a vaccine is tested for seven or eight years, and that’s how we know it’s safe before we roll it out.

Pfizer’s phase-three trial of its COVID-19 vaccine went for about four months; AstraZeneca’s ran about six months. That seems short, no?

The truth is more reassuring.

The scientists I spoke to aren’t sure where the “seven or eight years” rule comes from. Most phase-three vaccine trials – that’s the big one you do before you apply to roll it out – follow up volunteers for just 12 months, Jim Buttery, head of signal investigation at Victoria’s immunisation safety service, tells me.

Alejandra Gerardo, nine, gets the first of two Pfizer vaccinations during a clinical trial for children in North Carolina, United States.Credit: Duke Health

This testing schedule has produced many safe vaccines.

Of 57 vaccines approved by the US since 1996, the median safety follow-up was 1½ months. Yet only one had to be withdrawn from the market for safety reasons; at the same time, these vaccines cut the incidence of common viruses such as polio and measles by more than 90 per cent.

It is quite amazing how safe and effective these medicines are.

And the trials often have to run longer than COVID-19 ones because they have to capture enough infections in the placebo group to be sure a vaccine is working. It might take a long time to capture enough shingles infections; that is not a problem when testing a COVID-19 vaccine in the middle of a pandemic.

Why are the COVID vaccine trials so short?

To get the answer, we need to look not at vaccines but at the thing they are acting on: our immune system. Remember, a vaccine by itself does not do anything. It’s the immune system that needs to spot it, respond, and develop immunity.

If something is going to go wrong, it will probably go wrong there. But the thing about the immune system is, it is designed for speed.

“The immune responses act very quickly. If they don’t, we wouldn’t be able to fight infections. They need to get on top of the pathogens before they replicate and kill us,” says Stephen Turner, head of microbiology at Monash University’s biomedical institute.

This explains why scientists are so confident the vast majority of side effects will be spotted within the first few months of vaccination.

There’s more good news: we are no longer in a situation where we have five months of data on 30,000 people.

We now have more than a year of data for some people, and an extraordinary 4.18 billion doses of vaccine administered worldwide (at the time of writing) in which we can look for safety signals.

“A phase-three is typically tens of thousands. We already have real-world experience in tens of millions. We have massive experience using the best surveillance system we’ve ever had,” says Robert Booy, senior professorial fellow at the National Centre for Immunisation Research and Surveillance.

OK, but what about long-term safety? Am I going to wake up one day as a lizard person?

It would be nice if I could simply say, “There is a 0 per cent chance of long-term consequences.” Due to our continued failure to invent a time-travel machine, we cannot do that.

“Logically and scientifically, you can’t talk to the long-term protection or safety for more than a year. That’s an impossibility,” says Professor Booy.

This problem, of course, applies to all new vaccines. So what gives scientists and medical regulators the confidence to recommend vaccinations?

First, experience with past vaccines. As we have seen, if something is going to go wrong with the immune system, it’s going to happen fast, probably within the first six weeks after being given a vaccine, says Professor Buttery.

We see this with other vaccines. The only vaccine to be removed from the US market for safety reasons since 1996 was the Rotashield, which was intended to prevent rotavirus gastro in children. Its extremely rare side effect occurred within two to three weeks of vaccination, and was swiftly picked up by post-market surveillance.

Without 10-year safety data, “you have to always think from the best analogy”, says Professor Booy. “The best analogy is, we have a dozen or more routine vaccinations, for which we know the long-term safety profile is excellent, and adverse events that are important happen within six minutes or six weeks of getting the vaccine.”

But doctors would never recommend a treatment that came with a risk, even a small one, if that risk outweighed the benefit.

Over time, your chances of being exposed to COVID-19 rise to close to 100 per cent. Australia is not going to remain closed forever. Therefore, the correct risk-benefit balance is the long-term risk of vaccines versus the long-term risk of COVID-19.

We cannot know what will happen in a decade to people infected with COVID-19. But, unlike vaccines, the signs are not good.

Simply being in hospital with COVID-19 is not good for you, leaving patients with impaired cognitive and physical function and mental health. Problems have been reported in the heart, lungs, and brain. In a large Lancet study, 63 per cent of patients reported fatigue or muscle weakness six months after being diagnosed. People with severe illness ended up with damaged lungs. The stories of patients trapped on ventilators because their damaged lungs cannot support them are particularly harrowing. What is the long-term prognosis for this group?

“It’s always a balance of risks,” says Tony Cunningham, director of the Centre for Virus Research at the University of Sydney and one of the country’s leading vaccine experts.

“You have to always weigh the risk of any intervention in medicine against the disease itself.”

If you appreciated this article, consider signing up to receive Examine, a free newsletter, each week in your inbox.