David Crow on the complications — and consequences — of testing for coronavirus immunity.
Like countless others, I was convinced I'd already had coronavirus. I travelled a lot at the start of the year: a hellish, delayed, 30-hour flight to the UK from Australia, where I had spent Christmas and New Year, a skiing trip with friends in the Swiss Alps, and a week at the World Economic Forum in Davos, where I shook hundreds of hands in stuffy conference rooms by day and at crammed cocktail parties by night.
In February, I developed the worst chest infection of my life, so persistent that one night, while eating in a restaurant, I had to go home mid-meal because I could not stop coughing. I discounted any evidence that contradicted my hypothesis: at no point had I developed a noticeable fever, my sense of taste and smell had remained intact, and I did not seem to have passed the illness to family, friends or anyone at work.
When a test that could apparently prove I had contracted coronavirus first became commercially available in New York last month, I asked my doctor to send me for a blood draw. Rather than testing whether a person has an active infection, these tests perform a sort of biological archaeology, searching for pathogen-fighting proteins known as antibodies that show a person has had the illness in the past and possibly developed enough immunity to fend it off in the future.
• Covid 19 coronavirus: No new cases for fourth day in row - Ashley Bloomfield
• Covid 19 coronavirus: Biggest number of new cases recorded in a single day
• Covid 19 coronavirus: Superspreader infected 52 people in 2.5 hours at choir practice
• Covid 19 coronavirus: What you need to know about Thursday's big developments
The story of antibody testing thus far is a microcosm of the broader response to coronavirus itself: misplaced confidence and hype followed by premature pessimism. In March, the UK government announced it would make millions of at-home antibody tests available via Amazon and in high-street pharmacies. Those who tested positive would be given "immunity passports", enabling them to re-enter the workforce and everyday life.
The Chinese-made tests — procured at a cost of £16m ($32m) — turned out to be so unreliable that they never made it on to the market. In the US, the Food and Drug Administration allowed more than 100 different tests to go on sale, before performing a U-turn and warning that some "unscrupulous actors" had been peddling fraudulent kits.
Last month, an alarmed World Health Organisation said there was "no evidence" that such tests could "show that an individual is immune or protected from reinfection". Later, it partially walked back those comments, adding: "We expect that most people who are infected with Covid-19 will develop an antibody response that will provide some level of protection."
The allure of such a test is obvious: a positive result might be a one-way ticket to a semblance of normality. Underpinning the initial fanfare was a misguided hope that a large chunk of the population had contracted coronavirus but experienced only mild-to-moderate symptoms or none at all. If that were the case, society might be closer to "herd immunity", which generally means a combined infection and vaccination rate of between 60 and 80 per cent, and a return to the quotidian would have been just around the corner.
Such a reality has not come to pass. Even in New York state, on many measures the global epicentre of the virus, only 12.3 per cent of the general population has developed antibodies specific to this particular strand of coronavirus, Sars-CoV-2, according to a recent study of 15,000 people.
"The hope was that we'd roll out the testing, find lots of people with antibodies, infer many people are immune, and we can start going back to our regular lives," says Lloyd Minor, dean of Stanford University's school of medicine. "By anyone's measure, we're nowhere close to that."
While the early hype was misplaced, there is a risk that the pendulum has swung too far the other way. A colleague of mine recently got tested for antibodies, but her doctor said even a positive result was of little use: she was told she was still at risk of reinfection or could catch a mutation of coronavirus. However, there are very few if any documented cases of reinfection in the medical literature; and, while coronavirus is certainly mutating, most scientists say it is not yet doing so at a pace that would mean humans with the right level of antibodies are unable to fend it off.
My own test took place at a privately run centre in Harlem, operated by Quest Diagnostics, one of the biggest laboratory operators in the US. My blood sample was then shuttled 320km away to Baltimore, where the plasma was spun off and tested on a platform developed by Abbott Laboratories, one of the world's largest diagnostics companies. In the US, a dozen similar tests have been approved for emergency use, including one made by Roche, the Swiss pharmaceuticals group, as well as less well-known companies and individual hospitals.
An antibody test is quite a simple construct. In the case of coronavirus, the plasma sample is placed on a dish known as a well plate containing virus particles or "spike proteins", which Covid uses as a key to unlock human cells and infect the body. If the plasma sample contains enough antibodies, they will bind to the spike and thus to the plate and act as a kind of lock that prevents infection. That binding is then detected using a form of chemical torchlight.
My own test came back negative. "I'm sorry this isn't the result you wanted," my sympathetic doctor told me over the phone.
Dr Jay Wohlgemuth, chief medical officer of Quest Diagnostics, told me that "during a pandemic, when someone has an illness, they naturally assume the worst. But it is still far more likely they have one of the other viruses we get coughs and colds from."
I had not developed the antibodies, I had never had the virus. Case closed. Well, not quite.
The immune system is best thought of in two distinct parts. The first is known as innate immunity, a catch-all response that can fight off all sorts of illnesses. Dr Scott Boyd, an associate professor of pathology at Stanford, describes this as "built-in hardware" which can "recognise general molecular patterns that many different pathogens" use to attack the body.
In many cases, this innate system is not enough to prevent viral infections, and so the second-stage "adaptive" immune response kicks in. Within a couple of weeks, the body's B cells will have started producing a longer-lasting antibody called IgG, which is what most of the tests are searching for. No one yet knows how durable these antibodies are in the case of coronavirus, but as long as a person is producing them in sufficient quantities, most immunologists agree they have a degree of protection against reinfection.
Eventually, the B cells turn into a memory device that can quickly reproduce the antibodies if a person is exposed again. "Once the body recognises subsequent exposures, the memory cells say, 'Hey, we've seen this guy before and we need to rapidly take action,' " says Prakash Nagarkatti, a professor in the pathology, microbiology and immunology department at the University of South Carolina's school of medicine.
People who do not get sick very often sometimes boast of having a "strong" immune system, though this is something of a misnomer. Most of the people who become seriously ill with coronavirus initially mount a subpar immune response but then the body goes into overdrive.
"The body becomes so frustrated that it starts mounting a hyper-immune response," says Nagarkatti. "It's like a war zone where you use B-52 bombers to fire on a tiny village with only a hundred enemy people hiding there. You're using these huge bombs and causing devastating collateral damage."
It is possible that I had a mild case of coronavirus that was dispatched by my innate immune system before the adaptive system had to kick in. Cold comfort, however, because I still wouldn't have antibodies to protect me against future infection.
Another possibility is that I did indeed contract coronavirus and develop IgG antibodies, but not in sufficient quantities to be detected by the test. Even tests that are generally thought to be of high quality, such as the one from Abbott, have been developed quickly and in a manner that makes them more likely to generate positive results in people whose illness made them incredibly sick.
Abbott says its test is 100 per cent sensitive, meaning it will not return a false negative result. However, the tests have not been evaluated in full-blown clinical trials, which can take many months to complete. Instead, to speed up the development process, regulators have allowed Abbott and others to assess their tests using blood specimens from confirmed coronavirus patients.
"This was one of the challenges," says John Hackett, the Abbott executive responsible for developing his company's test. That difficulty arose because the main way of confirming coronavirus is not a blood test but a polymerase chain reaction (PCR) test that is conducted using a nasal swab. "So a lot of patients were never done with a blood draw unless they were hospitalised," Hackett adds.
Abbott managed to secure the requisite number of blood specimens, but the fact remains that the majority of samples will have come not from patients with no symptoms or mild-to-moderate ones, but from those who became so poorly that they ended up in a hospital where doctors were taking their blood.
"Almost all the tests on the market are reporting sensitivity based on a set of samples that are from known infected patients, but I think often there's a bias for ones that are from patients who are sick enough to be in the hospital," explains Boyd.
He adds: "That leads me to think that there may well be a decent number of individuals who did have true infections, but who might not have had an antibody response, or had one below the level of detection of most of the current [tests]."
Given all the uncertainty, it is perhaps unsurprising that many public-health experts have greeted the arrival of antibody tests with scepticism. Some fear that people who do test positive for antibodies will start behaving irresponsibly and stop adhering to blunt but effective measures such as lockdowns and social distancing that have managed to slow the spread of the disease.
"If you test positive for IgG, I don't think there is any science out there that says you don't have to follow the rules," says Dr Rajiv Shah, who ran the US Agency for International Development between 2010 and 2015 before becoming president of the Rockefeller Foundation. "If we start using the results in a way that's not backed up by science, people could end up doing some risky things. It is not a licence to do anything that you're not already allowed to do."
None of this means that people should not get tested for antibodies. The science is evolving at breakneck speed and, while there is no concrete evidence, every immunologist I have spoken to says there is good reason to believe that people who had coronavirus and developed high levels of IgG antibodies will indeed be protected against future infection. These people could also decide to donate their plasma to the sickest coronavirus patients so that their antibodies can neutralise someone else's disease and even save their life.
It is only a matter of time before there is incontrovertible proof one way or another. If antibody tests do turn out to be a reliable gauge of immunity, employers could start staffing their warehouses and factories with workers who have tested positive for antibodies, while implementing strict surveillance or even regular PCR testing for those who have not. However, they will have to tread carefully to avoid an Orwellian outcome, where employees are separated into "have antibodies" and "have not antibodies", potentially based on whether they are able to access or afford the tests.
Even a negative result is of some benefit to society. My result, for instance, will be stripped of any personal data and used by researchers conducting studies in New York and elsewhere to build a constantly evolving picture of the true rate of infection.
For now, then, think of a positive antibody test result as a young antique. It might not be worth a huge amount today, but its value could increase exponentially over time. I certainly wish I had one in my back pocket.
Written by: David Crow
© Financial Times