Last month, a small group of international scientists met to decide an issue critical to the health of millions of people all over the planet. For once, it wasn’t about coronavirus, although these experts know a lot about that, too. The task in hand was to save us from a bug we’ve been fighting since before the days of Hippocrates, the first doctor to describe it in 400BC. It’s an adversary potentially as much of a threat as Covid. These scientists are the flu hunters – heads of a handful of international institutions who track this old foe as it evolves and disperses in its own fight for survival.

This crucial annual meeting was held, for the first time since Covid struck, in the plateglass anonymity of the World Health Organization building in Geneva, with a few participants on Zoom. The scientists came armed with mountains of data and decades of experience to decide which four strains of the flu virus circling the globe should be in the next flu vaccine, to protect us from illness and our healthcare systems from buckling. Each has their own opinion and there can be wrangling and even some political positioning. Experts from China, the US, Australia, Japan and Russia are involved. It’s not always easy. But on 25 February, with what surely ought to have been a white puff of pontifical smoke, they made their recommendation public. Many thousands of lives will be saved – if they have got it right.

Chairing the meeting was British virologist Dr John McCauley, who has attended these meetings for more than a decade – usually in person. “It’s hard not on Zoom, but it’s awful on Zoom,” he says with a grimace. McCauley, who is in his 60s, grey-haired and jocular, has spent a lifetime watching the devious twists and turns of the shapeshifting virus. Since 2009, he has been director of the Worldwide Influenza Centre in London, based at the Francis Crick Institute next to St Pancras station. And in the months leading up to the meeting, he was busy taking delivery of large numbers of boxes of carefully packaged, potentially lethal human flu virus samples from 90 centres, mostly in Europe. In a normal year – and the last two have been far from normal – the Crick receives 120 or 130 shipments, containing about 5,000 samples. Other boxes stacked ready to be shipped out contain viruses destined for labs in other countries, where scientists will compare them with their own samples. The labels are a story of friendship, colonialism and cooperation: Cyprus, Valladolid, a pre-war St Petersburg.

“There are periods in which we’re very busy and periods in which we’re extremely busy,” McCauley says. He patiently explains the complexities of the virus, with just an occasional flash of suppressed irritation at this interruption to his lifesaving work. He strides briskly in his white coat between his small glass office overlooking the institute’s atrium and the dark, closed-off labs behind it. Entry to the labs is restricted whenever there is live flu virus around.

In January, the trickle of flu virus samples becomes a flood. Cardboard boxes containing the evidence of its latest mutations are brought in and transferred to freezers at -80C, where they will last for a couple of weeks while lab staff get to work. For what they have to do, the word “testing” is inadequate, McCauley’s frown makes clear. It’s about “isolating the virus, propagating the virus, doing the genetic analysis, doing the antigenic analysis, doing the drug resistance analysis, on all of these,” he says. McCauley must work out how the virus is changing so that he can inform the February meeting.

***

This year something unusual has been happening. There has hardly been any flu for the past two winters. There’s even speculation that some strains have vanished completely. But the flu hunters know better than to underestimate their enemy. If you can’t see it just now, it’s probably only hiding. Flu is a killer. It’s not just a bad cold. Man flu is – well, not flu. If you’re at work and think you’ve got flu, you probably haven’t. A proper bout will keep you in bed for days, unable to do much more than moan. Without vaccines, it would regularly scythe through the population, and occasionally cause pandemics, as it did in 1918, when it killed at least 20 million people (Covid has so far killed an estimated 6 million). And it’s a complicated, slippery foe, one that is very good at evolving or mutating to escape vaccines.

There are even two types: B, a historically human variety, and A, the animal sort, which can jump species and cause pandemics. Flu B, McCauley says, “has been with us for 10,000 years”. That’s the one Hippocrates knew about. Flu A, on the other hand, originates in animals and birds. Sometimes people in daily contact with domesticated animals, such as pigs and chickens, pick up the virus. Just before Christmas 2021, 79-year-old Alan Gosling caught bird flu from his Muscovy ducks. About 100 wandered his land in Buckfastleigh, Devon; he fed them and 20 lived in the house. He was the first person in Britain to test positive for an A strain called H5N1. The tragedy for Gosling was that all the birds had to be killed and he was banned from adopting more. The good news for everyone else is that the strain was incapable of spreading from him to any other human.

There has been a lot of bird flu about, in the UK and across Europe, known as “highly pathogenic avian influenza” or HPAI. In England, it started in October in wild swans in Worcester. Wild birds carry the virus; domestic birds pick it up. When it gets into poultry, whole flocks have to be slaughtered. There is always the risk that farm workers in contact with chickens, geese or ducks will contract the virus, as Gosling did.

Usually, this bird strain of flu cannot be transmitted from the farm worker to any other human. The nightmare scenario is where the virus evolves again and is able to be passed on. Then you’re in potential pandemic territory. In 1918, the “Spanish flu” came from birds. This was a lethal strain of Flu A called H1N1 that killed scores of soldiers and young people at the end of the first world war – previous infections meant older generations had more immunity to it. The strain stuck around until 1957 or so, but by then it was no longer causing devastation because so many people had been exposed to it.

Another strain of H1N1 caused the swine flu pandemic in 2009, which took off in Mexico and the southern US. It caused a massive scare, but far fewer deaths than feared: approximately 400 in the UK and up to half a million worldwide. Bad, but not on the scale of 1918.

Pandemic flu is so deadly because it is caused by a bird or animal virus that is totally unknown to the human immune system. As with Sars-CoV2, the cause of Covid-19, viruses kill when we are not prepared for them. Seasonal flus surge in the winter months, when people are huddled together indoors and breathing each other’s air. In a normal year, up to 646,000 people worldwide die of flu.

A vaccine against one flu strain won’t always defend against another, which means the shot has to contain a cocktail. The two strains of Flu A in the current vaccine are H1N1 and H3N2. The latter arrived in the 1968 pandemic that began in birds in Hong Kong and killed millions. Each of these strains has evolved different versions, which have complicated code numbers and may also be named after the place they were first spotted, such as Victoria, Yamagata, Texas or Bangladesh. Flu isn’t fussy; it will happily mutate anywhere.

So how do you outwit and outflank a virus this adaptable and devilish? How can you stop it in its tracks?

***

Global flu surveillance started in the UK. Back in 1947, the year before the WHO was launched, British virologist Christopher Andrewes, arguably the first flu detective, was involved in talks about the nascent UN organisation’s work. The Spanish flu was a horror story in living memory and tracking influenza a priority. Andrewes was asked to set up a Worldwide Influenza Centre, which he did in 1948 under the auspices of the National Institute for Medical Research in London. In 2016, the NIMR became part of the Francis Crick Institute.

Obituaries of Andrewes, who died in 1988, paint him as the archetypal eccentric but brilliant British boffin. “Christopher Andrewes was characteristically seen in a tweed jacket, white hair rather dishevelled, his ruddy face expressing interest, puzzlement, concentration, or fun – never boredom,” wrote fellow virologist, David Tyrrell, who in 1965 discovered the first human coronavirus (a cause of the common cold).

In 1933, Andrewes helped discover the influenza A virus. Three years later, 16-year-old Dennis Busby, who had joined the NIMR straight from school as a “lab boy”, received the first ever flu vaccination. He was not paraded for the cameras like Margaret Keenan, the first in the world to receive an approved Covid vaccine. In those days, scientific breakthroughs happened more discreetly, and a vital global breakthrough went unmarked.

The stories have become folklore among the virologists who have followed Andrewes down the years – a small group you can’t help thinking were cultivated in the same petri dish. “I knew Dennis Busby,” McCauley says. “He was one of the best virological, technical guys.” Busby went on to become head technician in the bacteriology and virus division of the NIMR, which moved to Mill Hill, London, in 1950. Was it ethical to jab him with an experimental vaccine at an age when most boys are at school? “He prepared it himself,” McCauley says, almost proudly.

McCauley’s predecessor-but-one as head of the Worldwide Influenza Centre was Sir John Skehel, an eminent virologist who has shown, among other things, how the influenza virus enters human cells and how antibodies can block it. Skehel ran the centre from 1975 to 1993 and was director of the NIMR until 2006.

“Would you like to meet him?” McCauley asks. And there he is, the 81-year-old recipient of a string of awards and honorary degrees, across the atrium from McCauley at the Crick, working in a similar small office, cheerful and welcoming. The two chat about the difficulties in tracking flu caused by the coronavirus pandemic.

“We’re trying to get some virus in from Hong Kong, but there are no flights,” McCauley says.

“It’s too bad,” Skehel says. “Hong Kong is an important laboratory. They picked up H5N1 bird flu in 1997.”

In the past, Hong Kong has been a window on to what flu is doing in mainland China. It has excellent scientists, often trained in the UK and well known to their British counterparts, and some of the samples they collect are from southern China. As China takes an ever firmer grip of Hong Kong, it’s uncertain whether the UK will continue to get those sorts of insights.

Seven WHO global reference centres – the Crick, Atlanta and Memphis in the US, Beijing, Tokyo, Melbourne and Novosibirsk – lead flu surveillance around the world. Five are devoted to human flu; two watch for animal viruses that might jump to humans. They are on the alert for flu outbreaks in chickens in Egypt and pigs in Central America, seeing whether farm workers or families with animals get infected and start to infect others.

Feeding into those WHO hubs are national influenza centres. In England, this is run by revered virologist Prof Maria Zambon at the UK Health Security Agency (UKHSA). Zambon was one of the key members of the government’s Sage advisory committee during the Covid pandemic, alongside her Imperial College colleague Prof Neil Ferguson. She is also an adviser to the WHO. And she never stops working.

Decades ago, Zambon was part of the core influenza group at Mill Hill. There is a “strong historical thread” running from Andrewes to the present day, she says. And she’s not surprised so many of the people she knew a quarter of a century ago, such as Skehel, are still tackling flu. “You fall in love with working on viruses,” she says. “It was a very good start for me because I was surrounded by top-class scientists in a wonderful, collaborative environment. And you take that passion with you.”

Even when the world is engulfed in a pandemic caused by a different virus, flu really matters. “Whichever way you look at it, it’s a major public health threat. And we still haven’t quite solved the problem,” Zambon says.

She is heavily involved in the response to Covid at UKHSA and Imperial College, where she is co-director of the NIHR health protection research unit. Was she surprised the pandemic that shut down the world was caused not by flu, as scientists had for so long predicted, but coronavirus? “The answer to that is yes,” she says, “and no. Coronaviruses have demonstrated their potential for causing havoc over a number of years.”

Zambon helped identify the first Sars virus in 2003, as one of an international team of virologists analysing samples from 436 patients in six countries. In 2012, she led the UK’s efforts to understand another coronavirus, Mers (Middle East respiratory syndrome), that turned up in two people who had travelled to the UK and spread to two others. She advised the WHO on it. “With both of those episodes, globally we dodged a bullet,” she says.

It’s thanks to all that historical UK influenza effort that the world is in a better place to tackle the coronavirus pandemic. Not only is surveillance strong, but so is the UK’s ability to assess how well vaccines are working through the matching of data on immunisation and illness. “We have a vaccine effectiveness system for Covid vaccines that is virtually second to none,” she says.

Zambon’s team at the national influenza centre sends the Crick “one or two examples of things that are absolutely representative. And then if we see anything unusual, or a bit weird, we’ll give it to them as well. So they are getting a kind of digest of viruses from us.”

Each of the devolved UK administrations does its own collection and analysis of flu samples, and sends McCauley those they consider interesting. There can be variations in what you find even within regions, says Dr Catherine Moore, who heads flu surveillance at Public Health Wales. Her field is molecular diagnostics, investigating changes in human, viral and microbial genomes to diagnose diseases. She swung into action at the start of the Covid pandemic, developing a molecular assay diagnostics test (PCR) to detect its genetic material (RNA) within a month of the coronavirus being recognised.

Moore knows Zambon and McCauley well, and shares their enthusiasm. She is “the current custodian” of a system that has been running for 40 years, she says. It’s “an absolute joy” to be even a small cog in the Global Influenza Surveillance and Response System (GISRS).

What matters is not the number of flu samples they collect, she says, but gathering high-quality data that can indicate the drift of virus strains, potentially into something different and troublesome. They operate within the hospital system in Wales: “What we want to look at is the severe end of the spectrum.” That’s all the way from a one-night stay to intensive care.

Moore has 44 GP practices also feeding in test results. In England, there are 245, organised through a network run by the Royal College of General Practitioners. Since 1967, GPs have collated reports of flu diagnoses from selected practices around the country. In 1994, Zambon was involved in helping with the transition to a system where patients’ flu samples are dispatched to centralised public health laboratories for virological testing. GPs used to be at the sharp end of all of this, swabbing their patients’ throats for flu virus in the way we all now know how to do. Flu is easy to catch, so it was a high-risk occupation. Covid has changed things dramatically, and now people in low-risk groups with flu-like symptoms are able to request home test kits and submit their own samples.

“One of the things the pandemic has done is ensure that much less explanation is needed now on how to take a swab,” says Prof Simon de Lusignan, an academic GP at the University of Oxford who runs the network organising sample collection in England. A charity called TakeATestUK dispatches the flu kit. The patient does the throat swab and posts it off to Zambon’s lab teams for analysis. GPs still feel they need to see some patients in person – such as older people who may have a chest infection, and small babies – but others will be told they are best staying warm, drinking fluids and getting over it at home.

***

The vaccine strain selection meetings at the WHO started in the early 1970s and take place twice a year. In February, they recommend the virus cocktail to go into the northern hemisphere vaccine, with a meeting six months later for the southern hemisphere.

The meeting used to recommend three strains. Now, manufacturing technology has improved to accommodate four. It is difficult stuff. Not every virus strain can be grown successfully – traditionally in chicken eggs – in the quantity needed (it is then deactivated). Usually, the committee gets it right, pretty much. “We normally say it’s right something like 70 to 80% of the time, because what you’re trying to do is to match a vaccine strain selection with a moving target,” Zambon says.

But it’s an educated bet. You can see the runners and riders – the strains that are most prevalent in the world – but they can change within six months. “That’s what you need the crystal ball for,” she says. “Some of the time, there’ll be a runner that pops up by August or September that was not present in January or February and if that runner emerges, and is successful against the other riders, then you have a mismatch. It does happen, and it can be painful when it does.”

A mismatch is the nightmare. The experts got it wrong in 2014-15. The vaccine protected against Flu B and the H1N1 strain of Flu A, but not against H3N2. In February 2014, the Geneva meeting had opted for the Texas strain of H3N2, kickstarting manufacture of the vaccine across Europe and the US. But as early as March, the flu watchers could see that a different strain of H3N2, from Switzerland, was spreading faster. “It dominated in the winter of 2014-15,” McCauley says. It was too late to switch horses. If vaccine production had been stopped to make a change in the component strains, there would not have been enough of it. That year, there were an estimated 43,900 excess winter deaths in England and Wales – the largest number since 1999-2000.

Something odd is happening with Flu B, too. For some time, the two B strains in the vaccine have been Victoria and Yamagata. Now Yamagata seems to have disappeared. “At the start of 2018, half the intensive care unit beds in the UK were filled with people with B Yamagata lineage viruses,” McCauley says. “Now it might have shot its bolt. We haven’t seen a B Yamagata lineage virus for over a year in these laboratories.”

What’s going on? The experts don’t know, exactly. The flu hunters trail influenza but can never get ahead of it. And right now, predicting what it will do next is difficult. Strange things are happening. Not only has Yamagata vanished, but China has reported only Flu B and no Flu A, the animal variety – so the speculation is that it’s not circulating. And in the US, neither Yamagata nor one of the substrains of H3N2 (which doesn’t have a place name) has been spotted since March 2020.

The answer must have to do with the Covid pandemic. McCauley produces a slide that dramatically demonstrates what happened in the winter of 2020-21, the first of the pandemic. Flu went into lockdown; there wasn’t any to speak of. Social distancing and mask-wearing thwarted not just coronavirus but flu, too. At the Crick, they were able to clear a backlog of work. They hadn’t been able to do that since 2008. “Last year, we did have periods when we didn’t have stuff to do,” McCauley says with a grin.

But if there are fewer reports of flu, is that really because of all the social distancing and mask-wearing? Or is it that people are not going to the GP or hospital and instead are staying at home in bed with a raging fever unnoticed?

Moore in Wales says it’s been hard to read. “Because of the pandemic, a lot of our surveillance systems have been disrupted, and particularly in the community.” But, on the whole, it seems there genuinely is less flu. “It’s been quite remarkable. I don’t think I’ve seen anything like this in my entire career,” she says.

In 2019, before the pandemic hit, things were looking dodgy. “We were in a really brisk season,” she says. There were four H3N2 substrains circulating in Wales, at least two of which were not great matches for the strain in the vaccine. Moore was expecting a tricky winter. “Then suddenly Sars [Covid-19] arrived,” she says. Because of mitigations such as social distancing, masks and working from home, “flu stopped circulating. We’ve had two years of basically nothing. No signal at all for influenza.” It’s no different in the rest of Europe.

Flu can vanish – and flu can come roaring back. McCauley and the committee have chosen their strains for next winter. It includes a B Yamagata-type strain (Phuket) in spite of its apparent disappearance, along with H1N1 (Victoria), H3N2 (Darwin) and B (Austria). “It is far too early to conclude it [Yamagata] is extinct,” he says. And not much flu means there is less natural immunity against it. People don’t have the antibodies from getting ill; at least, not the antibodies from recent strains.

All of us have some baseline immunity to flu. People born before about 1957 have some resilience to H1N1. It lurked for 40 or so years after the 1918 pandemic, altered in substrains, but older people have some general protection against it all the same. That’s why, in 2009 – the swine flu pandemic, which was also H1N1 – it was working people, aged 25 to 35, who died. Have I heard of “original antigenic sin”, Moore asks. The first flu strain you came into contact with is what your immune system remembers best, she explains. “That’s your immune system’s memory. If you encountered H1N1 as a child, you will be better protected against strains of that than against those of H3N2. And those born in the few years before the 1968-69 Hong Kong flu pandemic are likely to have more immunity to H3N2, which caused it. So H3N2 hits elderly people harder.”

Flu travels the world. It may start in Hong Kong or in Bangladesh, but there are no border controls. Global cooperation is essential, and that includes sharing samples of virus – carefully. McCauley won’t send flu virus to countries whose labs are anything less than highly secure. He mentions a country that asked for samples from the UK for comparison purposes. He wasn’t impressed with their safety protocols. “I said, ‘Would you accept Ebola? Would you do this with Ebola?’ And they went, ‘No, no, we wouldn’t do that.’ I said, ‘OK, so Ebola kills 70% of people. And this virus kills 40% of people that it infects. What’s the difference?’”

***

The flu hunters have decades of experience but there are always anxious moments. Moore is concerned about Alan Gosling and his Muscovy ducks. The ducks were infected because, while human flu is at low levels, bird flu in the UK is not. “We’re right in the middle of a massive epidemic in our birds in the UK,” she says.

She cannot understand the lack of interest. “We have a human infection from it and birds are dying. Nobody seems to care, but it’s part of the same family of viruses. I find it remarkable that flu has been denigrated almost to just a cold. It’s not. It kills people every year, and it’s distressing.” After the coronavirus pandemic is over, she says, people need to be reminded to take flu seriously.

In some respects, we are getting on top of flu. Conall Watson, a UKHSA epidemiologist, talks of “really positive developments in flu vaccinations”. He means the introduction of the adjuvant influenza vaccine, which gives older people’s immune systems a boost to help them respond and produce antibodies to the strains in the jab. “We’ve got evidence that it is preventing hospitalisations, but one of the really big steps the UK took was the introduction of a school-age and toddler vaccination programme,” he says, “with the idea of using vaccines to reduce community transmission.”

And the mRNA technology that has proved so successful in the Covid pandemic, used in the Pfizer/BioNTech and Moderna vaccines, is likely to be the future for flu vaccines as well. The companies are now looking to see how they can adapt it. We know mRNA vaccines can be fairly swiftly altered if a serious new coronavirus variant turns up. We could see a scenario where both the Covid and flu vaccines are redesigned every year, depending on the prevailing strains.

Will we ever get to a point where we don’t have to worry about flu? “No,” Zambon says, without hesitation, “because the natural reservoir for influenza A is wild birds.” They take the virus around the globe and there is nothing we can do about it. “So it will always be a threat to the human population.”

Which means that our best hope will continue to be vaccines. We are in the hands of the flu hunters.