UKH

The big concern with this new variant is the number of mutations on the spike protein, if I have correctly understood? Presumably if it changes enough it will no longer be the same virus, what point is that reached? 

And with the current mutations, will the PCR and LFT tests have the same accuracy? What level of mutation would be needed for the tests to stop detecting the virus?

Just my Sunday evening musings!

At a guess:

When people decide to call it a new one. I'm sure there is a guideline somewhere in virology but it seems likely that it is arbitrary. For something like a bird or mammal, you can say it is a new species when it can't breed successfully with an entity from the species it split off from. But viruses don't reproduce that way so it can't be applied to them.

I think the stats on the accuracy of these tests should be available with a little googling, no?

Interesting question, there are very few people who would be qualified to answer this question (and even for them I would imagine the answer might be "not sure. When we have someone who definitely has covid but consistently tests negative, we'll know")

The location of the changes matter as well; the spike protein is big, there are several key areas on it in terms of escaping the most potent ("neutralising" or "sterilising") antibody response and in terms of how well it can connect to the human cells.  

Omicron has a lot of changes in key areas.  I think that's why people are being so precautionary.  

When cross-immunity fades to a minor level seems like a good criterion.  Still a wide open question to answer mind you.

Depends what you mean by "accuracy"...

• False positive rate for PCR should remain essentially zero,
• False negative rate for PCR is going to be affected by
  • The change in viral genetics- (we know it's dropped compatibility with one of the primers (targets) used by the UK testing, which may reduce the detection efficiency - certainly won't increase it
  • The change in viral load produced by infected persons (which translates to some degree in to the "Ct" value used in RT-qPCR so data should be forthcoming soon).
• LFTs - have't looked at what they target.

I suspect the grey area around virus being alive (or not) in the first place, and their reproduction not requiring sex gives speciation criteria normally used to define a species as somewhat difficult. IE generating viable offspring. Virus just facilitate the production of mutated clones. 

Therefore it's all just a point of view, an opinion.

Deeper down, nature is rarely binary.  

Whatever happens. I think it's a long game. 

Nice muse for a Sunday though. Made me consider a few 'bigger questions'. 

K

Disagree.  For PCR testing it’s down to the short regions of viral code used to trigger amplification, known as “primers”.  If the virus has a change in the code in all primer regions, it will not be detected by PCR testing.  There are half a dozen active posters who could answer this question I reckon, and many more passive ones I expect.  I know a dozen people I could ask tomorrow, and you could walk in to any university biology department or hundreds of businesses and find many people in each who could answer that question.

I haven’t looked to see if the primers the UK uses are public knowledge; I imagine they are.  If they are, anyone reading this could learn to tell if a particular viral genome will escape the primers or not.  Assuming they can play “spot the difference”.

Is there a possibility of covid turning into a disease of the young?

Genuine hypothetical question. Could a variant change it's lethality to younger people? I'm sure it would change the dynamics in behaviour within the UK.

My knowledge of infectious diseases is minimal (limited to main stream media and UKH / UKC forums) so go easy on me if it's a stupid question!

Surely it depends on what protein/part of the virion is used as the antigen for the PCR. It may well not be the spike protein, but a common fairly fixed part of the viral particle across variants. In which case, then the PCR test will remain functional and viable. I do not know if the PCR test looks for a core protein, or some other part of the virion.

When WHO run out of names based on the Greek alphabet.

"Genuine hypothetical question. Could a variant change it's lethality to younger people? I'm sure it would change the dynamics in behaviour within the UK."

That's basically what happened with the spanish flu (1918 to 1920).  

I'm hoping for a mutation meltdown christmas! (virus mutates itself to be less harmful and fades)

Possibly some good news regarding LFT? Saw this yesterday.

https://twitter.com/bealelab/status/1464970713488576521

The primers don't target the spike though I don't think?? And given that's where most of the (sequenced) mutations are it should be as-is for PCR based tests.  LFTs target antibodies anyway so will (as before) be very similar in terms of accuracy/reliability.

I think the OPs question was a generic one inspired by this new mutant rather than tied to it.

The UK PCR test uses 3 primers targeting, each targeting one gene:  S (spike), N (nucleocapsid) and ORF1ab (bunch of stuff) - see page 59 in [1].

So, for a variant to fully evade our PCR testing it would need mutations in the relevant parts of all there genes.

Omicron has changed in the primer matching part of the S-gene causing an "S-gene target failure"  (SGTF) - the presence of two other genes still gives a robust detection methodology so long as the threshold criterion are adjusted accordingly.

It's remarkably handy that the S-gene target failure has been [0 > 1 > 0 > 1] with the sequence of variants [ᴡɪʟᴅ-ᴛʏᴘᴇ > ᴀʟᴘʜᴀ > ᴅᴇʟᴛᴀ > ᴏᴍɪᴄʀᴏɴ] as it means each time a new one sweeps in, it can be tracked in detail without sequencing, by proxy of the SGTF.  

https://assets.publishing.service.gov.uk/government/uploads/system/uploads/...

I haven't looked at the sequence matching for the LFTs.  

Neither had I for the primers in use in the UK, so thanks for the links. 

Is there any indication ofhow much sequencing they're doing?

I read through the PCR method many months ago, but I've not had a look at the sequence method.

.... around the beginning of next year.

You could watch the counter on here....

https://www.cogconsortium.uk

It's at about 1.5 million sequences today, press releases have it at 500,000 in June 2021 and 1,000,000 in October 20201.  So, about 250,000 sequences a month or about 4,000 sequences a day with the rate apparently rising.  So, we should be able to sequence about 10% of daily cases.  

I think the UK is about 25% of global sequencing activity.  A staggering level of activity.

Pretty much that. The species concept as applies to animals or plants indeed does not work for viruses. However the species definition for animals is also arbitrary and much more complex than testing whetherb two types of animal interbreed in the wild.

The best know examples are from ornithology, where many species pairs form hybrids, but hybridization as restricted to well defined areas without introgression of gene versions from one species ino the wider gene pool of the other (e.g. hooded and carrion crows in central Europe, where you will find plenty of hybrid pairs and their offspring along Elbe, but almost none just a few km East or West).

CB

Thanks folks  

I share the sentiment of those saying that they sincerely hope this level of mutation becomes less dangerous. That would be the ideal, more transimissable and less dangerous! Fingers crossed, but mask still firmly on and windows open until we know it to be the case!!

I thought the prevalence in the young in the Spanish Flu was due to the fact that many young people had just returned from war so massively impacted by the toll the war took on their bodies. They would have, presumably, had far worse immune systems having gone through so much stress. 

It seems that many of those who died were born around 1889 when there was an earlier less deadly flu pandemic

Interesting article here about it:

Exposure to an influenza virus at a young age increases resistance to a subsequent infection with the same or a similar virus. On the flip side, a person who is a child around the time of a pandemic may not be resistant to other, dissimilar viruses. Flu viruses fall into groups that are related evolutionarily. The virus that circulated when Adolfo was a baby was likely in what is called “Group 2,” whereas the 1918 virus was in “Group 1.” Adolfo would therefore not be expected to have a good ability to respond to this “Group 1” virus. In fact, exposure to the “Group 2” virus as a young child may have resulted in a dysfunctional response to the “Group 1” virus in 1918, exacerbating his condition.

It also mentions secondary infection with pneumonia

https://www.smithsonianmag.com/history/why-did-1918-flu-kill-so-many-otherw...

Interesting, thanks!

Except that the "Russian Flu" of 1889/90 may not even have been an influenza pandemic, but in fact a Coronavirus! There are good arguments suggesting that one of the current coronaviruses causing seasonal colds (OC43) started its career as an agent causing a lethal pandemic after jumping to humans from (most likely) mice again most likely via cattle.

I assume that this is what the endgame for the current pandemic may also look like, with everyone who is unable to mount proper immunity having died by some point, and everybody else being immune to the extent that they can be reinfected once immunity from the vaccination or infection drops low enough but still be protected from severe illness.

For current adults that will be achieved by immunization, for the old virus this worked by infection in early childhood: Everybody got a cold from the virus while the immune system worked nicely and did not overreact, which is enough to keeps you from getting severely ill later in life.

I guess this is probably overly optimistic, but could the current nasty cold which is going around actually be another variant of covid which has made enough of a transition that it is no longer covid but an endemic and less dangerous version? Could it be the start of us reaching that point? 

Probably just wild fantasy that we could be reaching this point, but reading your answer did make me think of the fact that there has been a sudden rush of a bad cold which is negative on covid tests. 

Obviously, from explanations further up, there is going to have to be some major mutations to reach that point, so it is unlikely, but would be good if it was the case!! 

I very much doubt it.  More likely people are getting more severe cold symptoms because immunity has faded a bit from a lack of exposure while things were more restricted.

Happy to hear you in particular say that.  

Or receiving a not inconsiderable amount of MAB preparations and/or anti-virals every winter.

Indeed.  I see immunisation with the current vaccines as the start of a bootstrapping process though; hopefully future years' infections won't be as clobbering as post vaccination delta is for many as immunity approaches whatever asymptote is achievable for first exposure well in to adulthood.   I'd hoped Valneva was going to feature in closing the gap to my personal asymptote, doesn't look like that's going to happen...

Conversely, I have read that the Spanish Flu was more deadly to the young paradoxically because they had stronger immune systems.  It provoked a 'cytokine storm', a kind of overreaction of the immune system that did more damage than the virus itself.

Eg:
"Most influenza outbreaks disproportionately kill the young and old, with a higher survival rate in-between, but this pandemic had unusually high mortality for young adults. Scientists offer several explanations for the high mortality, including a six-year climate anomaly affecting migration of disease vectors with increased likelihood of spread through bodies of water. The virus was particularly deadly because it triggered a cytokine storm, ravaging the stronger immune system of young adults, although the viral infection was apparently no more aggressive than previous influenza strains. Malnourishment, overcrowded medical camps and hospitals, and poor hygiene, exacerbated by the war, promoted bacterial superinfection, killing most of the victims after a typically prolonged death bed."
https://en.wikipedia.org/wiki/Spanish_flu

If I were still in the virology game I would have immediately applied for a grant to travel to Svalbard or Greenland to dig up some permafrost graves from 1890 and search for coronavirus RNA! Win / win even if you find nothing, merely for touristic reasons!

This has of course been done before (both for the "Spanish" and the "Russian" pandemics), but then people were specifically looking for influenza RNA, which of course was also present to some extent, even if only from seasonal influenza, and I am sure that other people immediately went looking once the Covid pandemic started.

The "Russian flu" was very unusual clinically (with reports of loss of taste, common "long covid" like fatigue, etc.), based on molecular clock estimates OC43 jumped to humans exactly around that time, and turning into a seasonal agent that recycles its hosts by allowing acute immunity (to ensure host survival) but interfering with long term memory (thus guaranteeing that hosts are available again after a few years) is a well known coronavirus strategy.

So if I had to make an educated guess, OC43 should be a good model for the long term outcome of the current pandemic:  Once everybody is infected with SARS-CoV2 as a young child we should see mainly mild disease in the years to follow.

It is worth keeping in mind that the "Russian flu" also caused at least 250.000 recorded deaths across Europe alone, so in reality more likely millions. This is what we need vaccines for, and why a herd immunity approach fails!

CB

My grandmother used to still talk emotionally to us about the death of her youngest sister, Dora, from the 1918-19 flu, describing how she came home as a thirteen year old to be greeted by her mother outside the door telling her they'd lost Dora.

It did, unfortunately, take the young and fit.

Shame, it would have been convenient if it was! 

I suspect there's a long way to go until it would need to be reclassified. I briefly worked with flu virus and there were lots of different types with varying H and N numbers depending on their hemagglutinin and neuraminidase proteins. 

This is very likely to happen at some point, if not with Omicron then with another letter of the Greek alphabet. From an evolutionary perspective increased host survivability is in the  virus's interest and this has happened in the past with other coronaviruses as well. 

However, what 'more transmissible and less dangerous' means in the short run (without restrictions) is more deaths. I read a column yesterday written by a doctor crunching the numbers and the short story is even if Omicron is 5x more transmissible and 10x less lethal than Delta that still means massive suffering if you go into big number territory. It's impossible to say for sure at this point what the stats will actually be but even with a best-case scenario it's still edge-of-our-seats stuff. 

I’ve seen this said a lot, but I’ve seen no actual proof.  Always interested if someone has a link to a strong scientific case for this…

Increased survivability comes with repeat immune exposure, particularly at younger ages; the virus doesn’t become less lethal but the survivors of the pandemic go on to be more immunologically familiar with it and so more robust to it.

So as I see it, the decreased lethality over time is more down to the changes in the surviving hosts than to changes in the virus.

The fatality rate of this virus on adults without immune exposure doesn’t seem enough to limit its spread nor to pose much selective pressure.

Flu is very different due to the different way it changes genetics regularly.

It makes most sense to rename a virus when there are markedly different clinical signs (although in veterinary science, they have just amended names on occasion e.g. high path feline Calicivirus or when there is no tangible link to previous viruses on phylogenetic analysis (i.e. like alpha when it was suspected to have emerged from bats rather from than a previous human coronavirus. I cannot see a name change whilst there is a link to the previous Wuhan (alpha) virus. 

The following other variants are being monitored at the moment. A new name for each would be very confusing.

• Alpha (B.1.1.7 and Q lineages)
• Beta (B.1.351 and descendent lineages)
• Gamma (P.1 and descendent lineages)
• Epsilon (B.1.427 and B.1.429)
• Eta (B.1.525)
• Iota (B.1.526)
• Kappa (B.1.617.1)
• 1.617.3
• Mu (B.1.621, B.1.621.1)
• Zeta (P.2)

You may be right. 

https://theconversation.com/will-coronavirus-really-evolve-to-become-less-d...

TLDR; it could swing both ways and it's sort of a lottery.

Can we say this yet? If the new variant emerged in South Africa (which we presume) the number of people vaccinated is low (24%) and so this is likely a change in virus through passage in a mixture of a few vaccinated, some naturally immune, but mainly non-immune people rather than a response to the immune response from the vaccine (vaccine escape mutant). That change through selection and chance is clear but I think the lethality or otherwise needs yet to be determined.

I am not sure this is fair comment, whilst mechanisms can be different the rate and consequences of change in the two viruses are not dissimilar [unless I am missing your point]. We are seeing drift and the emergence of the SARS-Cov2 is very much a shift event.

If the young are unexposed and there is little/low maternal antibody, this is not a silly thought. It would probably remain circulating in the adult population with potentially nasty consequences for the old and immunocompromised.

I was talking about the previous posters comment on what's happened with "other coronaviruses in the past"; from what I can tell increased population immunity in the survivors is likely the main factor in reduced lethality, not the virus evolving to become less lethal.

Hard to know either way, but I've yet to see any evidence in support of decreased intrinsic lethality.

For sure, and it takes a long time to get a decent measurement of the lethality.  

If I had to make a guess.... We do know that this nCov is towards the fluffy-bunny end of the nCov family in lethality terms, and it seems that effective viral load > lethalithy, ongoing rapid host adaption in the RBD and antibody evasion are all in part tied up together so I'm leaning one way on this if omicron turns out to be able to out-compete delta in highly vaccinated populations.

But it's a long way to go yet.

My understanding is they're different kinds of shift, with influenza reassorting by swapping entire RNA segments between different strains, where-as coronaviruses engage in recombination at a finer grained level than whole segments.

The way I see it, I think this means that the shift events have different consequences in terms of immune responses as the changes fall at very different levels of granularity.  If the viruses only circulated in humans, this means that coronaviruses would have far more potential to evade immunity than influenza, as they can change their sequences in a much finer-grained way through shifts.  So, Covid has more potential access to an "evade" button, whilst influenza could swap away as much as it liked, but immune memory to the individual RNA segments would only be degraded by drift after people had eventually been exposed to all segments.  In effect there are a limited pool of RNA segments available to flu but an effectively unlimited pool to Covid as its mutation mechanism is more like a chop-shop.

The wild card (and where my point breaks down I think) is animal reservoirs - a shift event of either sort that brings in previously unseen material from an animal reservoir presses a very big "evade" button.  I'm kind of hoping Covid doesn't hit on some recombinant magic in one of the many animal species it now seems to be in...  The distinction between shift mechanisms doesn't seem to important here.

Perhaps I'm mis- or over-interpreting here!  

Flu has always held bird/pig recombination (think small holding) as a major concern, and that does not seem to be an issue in coronavirus. I can see your point about the more refined shifts, but that may leave T-cell immunity (likely directed against some core proteins rather than just spike) as useful. 

As for interpretation over or otherwise – we will see
[Must do some work now!]

We would normally keep the optimised primer sequences for our tests as a trade secret, but for our SARS-CoV-2 tests we decided to file a patent application so that (perhaps a bit paradoxically) we could disclose them to people who needed to know.   The patent application will publish fairly shortly and then anyone can see them.

I don't know - I've never actually seen any evidence of a coronavirus becoming less intrinsically lethal through mutation, but I've seen a lot of evidence that repeat immune exposure lowers the risk.

Evolving to become less deadly is something of a trope that gets trotted out by some scientists in discussion and runs in a bunch of discussion sites, but never seems to boil down to anything concrete...  The article you linked is an example of this - the actual science it links to is entirely about loss of lethality through building immunity in the host population, not an evolutionary dialling down of intrinsic lethality by the virus.

Funnily enough I spent an hour last night looking for the TaqPath primer sequences and was going to partially retract my reply to henwardian - whilst I maintain that a lot of people are "qualified" to answer this question, that turns out to be irrelevant basically nobody is able to, because the primer sequences themselves appear to be trade secrets - as you now confirm... 

I did eventually find what I think are the forward and reverse sequences for the S protein in some technical documentation on sequencing the S-gene drop out but nary a sight of the other.  I'd expected to find it in a regulatory filing with the FDA at least, but no.  Nothing ever actually spelling out "trade secret"...  Sort of a black hole in the documentation.

Very interesting to see the trade secret route taken rather than patent protection in this case, because you appear to walk a very fine line between giving away the sequences to others whom you have no chance of enforcing against vs giving away material that can be reverse engineered and that is not patent protected.  Bet there's been some long and difficult discussions behind the strategy there...?  I suppose there's a reason these can't be ordered through the web portal and I imagine there's a certain pickiness over who goes on the customer list...

"Start your photocopiers".

It was interesting diving in to the primer sequences to see how much variation there is internationally in terms of what's chosen.  

You're probably right. Although I do hope that we catch a lucky break here. 

It never makes sense to look at changes in pathogenicity from the perspective  of host or pathogen. Evolution always is coevolution.

I guess it will be very hard work to cleanly disentangle the effects of changes in host and pathogen even if in this pandemic we are for the first time able to track those coevolutionary changes in real time by molecular means.

There are good prehistoric examples for "weeding out" susceptible haplotypes, which is the most extreme evolutionary response on the host side. E.g. the large scale stone age population displacement in Europe was presumably driven by higher plague resistance in the herders that genetically replaced the previous hunter gatherers.

Similarly, adpatation to the evolving human measles virus after it jumped over from cattle certainly involved a gigantic selection pressure on genetic resistance!

In contrast, clear, mechanistically understood examples of pathogens weakening over the course of a pandemic are much harder to find and are typically less convincing.

I very much doubt that either elimination of susceptible genotypes from the host gene pool or evolution of the virus towards lower pathogenicity played a major role in the evolution of seasonal cold coronaviruses, simply because even the precursors likely were not lethal enough.

Instead, I guess that the gradual establishment of a new equilibrium between population level immunity and infections following mass immunizations and widespread childhood infections will dominate our "learning to live" with SARS-CoV2.

CB

Not really its just being able to spread between hosts. So ideally it needs a decent period where you are highly infectious and are capable of interacting with others.

Something that kills rapidly and in a visibly unpleasant manner (eg many filoviruses) is likely to burn out fast but so is something which has a zero percent lethality but has low transmissability and puts you in bed before you can pass it on.

You then also have the issue of who is the intended host. It might just be killing us by chance whilst being perfectly optimised for transmission between bats in a nonlethal manner.

I think mutation into whatever lethal new form, and our sustained ability to test and vaccinate for them, is proportionate to the number of libertarians who are so unburdened by empathy and intelligence as to go maskless into crowded places.

Today's Guardian website featured quite a harrowing piece about a relatively fit and young unvaccinated man dying of Covid. 

I only have one person (for as far as I know) in my immediate surroundings who is willingly unvaccinated and that piece gave me the chills. 

And yet every A-level biology student will have been told about the journey from parasitism to commensalism. Perhaps a half truth that needs addressing at that level, as it seems to have set a trend of thought that is unjustified. 

Surely this depends on the timescales...

The virus has perhaps 500 to 1500 generations in the space of one human generation.

•  Over longer timescales than a human generation it's coevolution
•  Over shorter ones its pure viral evolution.

The claims that routinely get made are that the virus is likely to evolve to be less lethal on a timescale of years, which puts it in the viral evolution bucket, not the coevolution one.

At least, this is the case where most people are not being killed by the virus.  Your measles example is a counterpoint where the selective effects on the humans are writ large immediately.  Covid's IFR is very low in adults of child making age, so it's going to have very little drive selective pressure indeed.

Covid19 itself became a "new virus" only when people started getting admitted to hospital in Wuhan.

There's an element of "when it causes illness" in "When does a variant become a whole new virus?".

Yes, time scales matter as much as the degree of selective pressure. We Europeans all are descendants people who were able to deal with plague, measles, a whole zoo of diarrhea agents, typhoid, etc. than the populations our ancestors replaced (the modern humans that largely replaced Neanderthals in Europe had dark skin, smooth, black hair, and screaming blue eyes, a combination that is pretty much absent today!). Malaria? Less of an issue here.

Measles evolved from Rinderpest virus in almost historic times, so we even have written indications of the kind of selective pressure it put us under when it was a relatively young zoonosis: There are written reports from Mesopotamia where the governor of some city writes to his superior not to worry, after all their local outbreak is only smallpox not the measles....

The much bigger difficulty is to distinguish the effect of viral evolution from changes in population immunity, which are just as quick. What are you going to do, infect a few naive control populations over time to see whether the virus circulating elsewhere becomes weaker? 

CB

edit: of course there are clear examples of short term pathogen evolution, e.g. selection for antibiotic resitance in bacteria or immunity evading mutants in viruses. However, I cannot come up with one example where the genetic evolution is obvious (i.e., one variant becoming much more prevalent in the population) and this is associated with reduced pathogenicity.

It's easy to forget just how cushty modern life is compared to those times...  

You could draw down the immune memory cells from some volunteers, then isolate them until any antibodies fade away, then expose some to OC43 and some to Covid?

I expect this is totally and utterly unethical by any sane legislation, so is off the cards.  But...  following the effects of accidental infection in survivors of adult measles infection might be interesting, both to pre-existing coronaviruses and to new ones.  I suppose people with severe immune deficiency aren't so relevant as immune dysregualtion is so central to the typical pathology.

That idea sounds as if there is a deus ex machina at work!

Not really. Ebola is a virus that has a long history of living happily ever after with its co-evolved fruitbat host. THIS is where any ongoing selection pressure for the virus is relevant. Wasting a few virus genomes on a quickly burnt out, dead end chain of human infections is irrelvant from the virus' perspective.

Successful zoonoses are rare, but quite a few of those that did manage to gain a foothold in humans have large effects: HIV, several pig or swine derived flu strains, or indeed multiple coronaviruses all made the jump in our lifetimes, but not all of these led to significant and stable human transmission.

CB