On the Delta Variant

In​ Arthur Conan Doyle’s The Lost World, Professor Challenger takes a party of adventurers to South America, where they discover a plateau filled with dinosaurs. The book’s lesser-known sequel, The Poison Belt, isn’t quite so thrilling, and not only because of the disappointing lack of dinosaurs. This time Challenger summons the same cast to Sussex, where he tells them that the Earth is passing through a ‘poison belt’ of ether. But – unlike the rest of humanity, which is doomed to immediate extinction – he assures them they will be able to live a brief while longer by sealing themselves up in his wife’s boudoir with the supplies of oxygen he has instructed them to bring. The heroes pull through, of course, and take a melancholy wander through an apparently dead world, but then everyone wakes up, happily without any memory of what happened.

The Poison Belt was published in 1913. By then oxygen cylinders were not new technology: they were first developed in 1868 for use in anaesthesia. J.S. Haldane had taken oxygen cylinders on an expedition to Pikes Peak, Colorado in 1911, where he made observations on the physiology of hypoxia at high altitude. (The research continues to this day at the US Army Pikes Peak Research Laboratory.) The only female member of Haldane’s party was Mabel Fitzgerald. After the trip, she published a paper in the Philosophical Transactions of the Royal Society demonstrating the close relationship between altitude and increased concentration of haemoglobin. The scientists who finally explained the molecular basis for the oxygen-sensing pathway that underlies this phenomenon won the Nobel Prize in Physiology or Medicine in 2019. One of them was Peter Ratcliffe, the clinical research director of the Francis Crick Institute, where I work. He began his Nobel lecture by presenting some of the data Fitzgerald collected.

There is much still to discover about the physiology of oxygen sensing, but there’s no mystery at all about what happens when you deprive someone of oxygen entirely. Infected, inflamed lungs can’t take in enough oxygen from the air. That’s why severely affected Covid patients die. Supplementary oxygen can prevent this – it’s that simple. If hospital oxygen supplies become overwhelmed, many people who would otherwise have lived – thanks to a straightforward treatment that’s been around for 150 years – will die. That’s what we’ve seen in India, and are now seeing in Uganda and many other parts of the world where public health measures have been insufficient to prevent uncontrolled transmission of the virus. This too is no mystery.

An infectious agent capable of infecting one new person for each currently infected person is said to have a reproductive rate (the famous R) of 1. If R is below 1, the number of infections will exponentially decay; if it is above 1, the disease will exponentially spread until the population develops sufficient immunity to bring R below 1. Exponential growth is easy to understand in mathematical terms, but challenges human intuition. The story of the rice and the chessboard is the classic illustration of this. A supposedly wise ruler asks a servant what he would like as a reward for an act of bravery. The servant says he would like one grain of rice for the first of the 64 squares on the emperor’s chessboard, two for the second, four for the third and so on, doubling the number of grains for each square. The emperor is happy to grant this apparently humble request. After all, the first ten squares provide hardly more than a few mouthfuls, and by the 15th square he’d have about a kilogram of rice. But by the 64th square he’d have accumulated 2⁶⁴-1 grains (18,446,744,073,709,551,615), more than a thousand years’ worth of the world’s rice production – or, more likely, fatally angered an unwise ruler. Trying to get politicians to understand the concept of exponential growth seems to be a perennial problem.

In India, where both chess and variant Delta originated, the unwise ruler had all but declared victory over Covid. In an address to the World Economic Forum at Davos on 28 January, Narendra Modi brought a ‘message of confidence, positivity and hope’. The hubris continued: ‘It would not be advisable to judge India’s success with that of another country. In a country which is home to 18 per cent of the world population, that country has saved humanity from a big disaster by containing corona effectively.’ Oxygen shortages and mass cremations followed in April. Contrary to Modi’s ridiculous claim, variant Delta has been exported to the world and is outpacing the previously all-conquering Alpha.

Alpha, Beta, Gamma, Delta: the variants of Sars-CoV-2 have been given new names by the World Health Organisation. Variant Alpha is otherwise known as B.1.1.7, and was previously known as ‘Kent’ in the UK, and ‘UK’ everywhere else. Beta is B.1.351, or ‘South Africa’, Gamma is the ‘Brazilian’ P.1, and Delta is B.1.617.2, or ‘India 2’, since B.1.617.1 and B.1.617.3 also looked potentially nasty. Alpha, Beta, Gamma and Delta are easy enough to remember, but we have reached variant Lamda at the time of writing, and I’m not looking forward to Omicron, Pi etc. There are no plans for what happens after we reach Omega. It’s at some level easier to understand variants by their country of origin, but this can give rise to unfortunate and occasionally racist interpretations. Greek colleagues complain, but it’s easier to call a variant of concern by a Greek letter than a Roman letter followed by a string of dots and numbers that don’t even really make sense to bioinformaticians.

Viruses evolve to optimise transmission. It’s often said that this means they become less deadly over time, because a virus that kills its host will not then transmit to others. This may be true of a very deadly virus that kills its host early, but with Sars-CoV-2, most infected people have mild symptoms, and most deaths are two weeks or more after infection, while the virus is most transmissible much earlier, peaking approximately between the fourth and seventh day. Variants are designated ‘of concern’ if they are more transmissible, if they cause more serious disease or if they evade immunity. So far, the variants of concern that have been identified exhibit all three phenomena to a greater or lesser extent.

Alpha is more transmissible, and a bit more pathogenic, but only marginally more likely to evade immunity. Beta and Gamma have mutations that confer reduced antibody binding, and hence are expected to evade immunity. While Beta and Gamma haven’t taken hold in the UK, Alpha caused a devastating wave of infections in December, when there was very little immunity. I have written previously about the infuriating blunders that caused this (LRB, 4 March). The government’s subsequent response – a very effective vaccination campaign and a relaxation of restrictions gradual enough to keep R below 1 – have nearly extinguished Alpha. Were it not for Delta, we would be applauding their success and readying ourselves for a moderate amount of hedonism on 21 June.

When the surge in India first made headlines, it wasn’t clear whether it was really caused by a new variant. Unlike the UK, where the sequencing of viruses enables rapid detection of variants, there wasn’t much data to go on in India. Alpha was present, but so were three ‘Indian’ variants. One of these, B.1.617.1, now designated Variant Under Investigation Kappa, exhibits a mutation seen in Beta (informally known by scientists as ‘Eeek!’) that’s associated with evasion of the body’s immune response. Early speculation suggested it might be Kappa that was responsible for the disaster unfolding in India.

On 2 April, the Department for Transport announced that the Philippines, Pakistan, Kenya and Bangladesh would be added to the ‘red list’ of countries, triggering travel bans and a ten-day quarantine for returning citizens. This was ‘to protect the country against new variants … at a critical time for the vaccine programme’. A good decision, but why was India not put on the red list too? Modi and Johnson were due to meet later in the month. In the end, their photo opportunity had to be cancelled anyway, but India didn’t join the red list until 23 April, and without this delay we probably wouldn’t now be postponing the end of restrictions to 19 July. The Delta variant was seeded, and outpaced Alpha to the point where it’s now responsible for more than 90 per cent of new infections.

The UK’s vaccination programme has been among the best in the world. The latest data from Public Health England show that 80 per cent of adults have antibodies to Sars-CoV-2. Vaccine uptake has been above 90 per cent in the oldest age groups, and it may eventually be similarly high for younger cohorts. This success is down to excellent communication of the (tiny) risks and (massive) benefits of vaccination. The very small risk of a particular form of clotting engendered by the AstraZeneca vaccine was carefully contextualised, and a wise decision to offer those under forty an alternative was explained with clarity and honesty. Despite this near flawless vaccine roll-out, Delta is still spreading, with R well above 1.

Why should Delta succeed where Beta, Gamma and the other variants of concern have so far failed? Part of the explanation is increased transmissibility. Alpha has a particular mutation in the viral protein known as Spike which is associated with increased transmissibility. Delta has a different mutation at the same place that seems to confer even more ability to infect. What about immune evasion? A group of my colleagues at the Crick used the live virus neutralisation assay we’ve been developing to investigate whether Delta was able to escape antibody responses. We expected a modest drop in the ability of immune serums to prevent the virus from entering cells. Instead, we saw that variant Delta was as bad as variant Beta from this point of view, and most people who had received only one dose of a vaccine didn’t have sufficient antibody levels to prevent infection. This result is confirmed by other laboratory studies and real-world data. Variant Alpha was largely stopped by one vaccine dose, but for Delta it’s really necessary to get both doses for adequate protection. This is why the timetable for second doses has been accelerated in the UK.

We will soon reach a point where the threat of Covid in the UK is substantially diminished by widespread immunity. The 19 July opening is unlikely to cause another devastating wave of hospitalisations – though there may still be significant pressure in some areas. In parts of the world where vaccine coverage is nothing like as high, I fear a catastrophe. Delta is much more transmissible than the original variant, partially evades immunity and probably causes more severe disease. India’s healthcare infrastructure is relatively well-developed compared to many countries, and has still suffered greatly.

At the G7 summit in Cornwall, one billion doses of vaccine were pledged for the poorest countries by the end of next year – enough to protect 500 million people. That number has already slipped, and even if it were not too little, it’s definitely too late. We could and should be doing more – with more urgency – to support poorer countries: not just vaccines, but testing, sequencing, oxygen. The more we delay, the closer we get to the second half of the chessboard. To talk in abstract terms of ‘reducing global health inequalities’ doesn’t convey enough moral force. If altruism isn’t enough motivation, enlightened self-interest will do. More transmission of this virus anywhere in the world increases the chance of a more deadly, more transmissible, more likely-to-evade-vaccination variant emerging. Do we really want to sit like Professor Challenger in his sealed room, watching while the world suffocates?

18 June