UK Variant B.1.1.7 becoming the dominant strain in the USA -- what does it mean for Spring/Summer?

Key points:

• The UK variant B.1.1.7 has become the dominant strain in the USA, going from 0.1% prevalence on January 1 to up near 50% prevalence by March 23.

• Bad News: This variant has been shown to produce higher viral loads, leading to ~50% higher transmission rates and ~50% higher death rates than wild type variants

• Good News: Fortunately, this variant does not have "immune escape" capabilities, so previously infected and vaccinated individuals should retain their full protection.

• The increasing prevalence of this variant could produce a spring surge of cases, but this potential surge is partially mitigated by the following factors:

  • An estimated 30-40% of USA residents have been exposed and infected with the virus and recent studies suggest a vast majority retain protection vs. reinfection for at least 6 months, and this protection should confer to this variant.

  • A total of 25% of USA residents have received at least 1 dose of vaccine and 15% fully vaccinated, with 2.5 million/day currently vaccinated and suggesting that another 5% are receiving vaccination doses every week, with 45-50% fully vaccinated, and 55-60% receiving at least one dose, by the end of May, and the vaccine protection should confer to this variant.

  • Between these factors, it might be 40-55% of USA residents have some protection vs. the UK B.1.1.7 and this % will continue to increase as vaccinations are done.

• However, given that >1/2 of USA residents are still currently susceptible to this variant, it is important to remain vigilant to do the small things to limit spread until a higher proportion are vaccinated to prevent a spring uptick that could threaten school and business openings.

• With the current pace of vaccination and pending durable immunity, population level protection vs. the wild type and UK B.1.1.7 variants may be widespread enough to prevent any further exponential spread by the time we reach the summertime.

• However, the "immune escape" variants like the South African B.1.351, Brazilian P.1, and New York B.1.526 could become more dominant in the summertime and produce more upticks in viral cases, but since it appears that the current vaccines maintain at least 2/3 efficacy vs. these escape variants, plus boosters are being developed that could restore full efficacy. Thus, the alarm and fear conveyed by some about these "immune escape" variants may be overblown, and it is still reasonable to expect that with broad vaccination we can have a summer and fall that more closely resembles normalcy.

UK Variant B.1.1.7 becomes dominant in USA

Tom Wenseleers has done a great job of tracking the spread of the UK B.1.1.7 variant -- here is a tweet with plots showing how it has become dominant in many parts of the USA, with the proportion of sequenced cases increasing exponentially based on a predictable pattern that has been seen in the UK and Belgium, places where it has taken over as the predominant strain.

The linear plot on this log scale indicates exponential growth, and in many states we see exponential increase from near 0 prevalence on January 1 up near or above 50%. It is clearly taking over in the USA as well based on these data.

Helix, one of the companies doing widespread PCR testing, is also presenting data on variants on its dashboard. They have sequenced a limited number of their samples, but importantly the B.1.1.7 has a specific mutation -- deletion of location 69/70 on the genome -- that makes the "S gene" part of their SARS-CoV-2 PCR test fail. However, the other components of the test work so the cases are still accurately detected, and so "S gene target failure" or SGTF is a surrogate for this hallmark 69/70 deletion for the B.1.1.7 variant. As can be seen by the lower right plot below, over time, nearly 100% of the sequenced samples that had SGTF were found to be the B.1.1.7 variant, thus making prevalence of SGTF a reliable surrogate for prevalence of B.1.1.7 in the population. This is very useful since sequencing is expensive and not done for many samples, and the sample selection for sequencing is typically not representative for the population, but SGTF can easily be obtained from ALL samples submitted for PCR testing so making it a reliable, inexpensive and more representative measure.

From the upper right plot, you can see that as of last week (March 14, 2021), the proportion of tests with SGTF is up to 50% nationally, and increasing exponentially. You can see where the Helix tests are used -- a lot in California, Florida, Texas, Pennsylvania, and the midwest.

So, this variant is becoming predominant -- what does this mean for us? First the bad news, and then the good news, and then I will discuss what I think this means for what we can expect in the summer and fall.

The Bad News: UK B.1.1.7 variant has increased transmission and severity.

The warranted concern about this variant comes from its increased transmissibility and risk for severe disease. In early January, I did a blog post breaking down in detail the population based studies done in the UK that I believed made a compelling causal link between the strain and increased transmissibility. These studies and subsequent studies that have followed have validated that this variant is 30-60% more transmissible, which makes a huge difference in the population level spread, taking circumstances for which current mitigation strategies are keeping the virus at bay (e.g. Rt~1.0 indicating no growth in cases over time) and producing instead sharp exponential growth (e.g. Rt~1.3-1.6) of the level that would double the daily cases every two weeks or so. This is a big deal and makes it much more difficult for mitigation strategies to keep the viral spread under control.

I also mentioned a medRXiv paper from late December that demonstrated that this B.1.1.7 variant (estimated again by the surrogate SGTF which they refer to as "S negative") appears to produce infections with much higher viral loads, with the median viral loads 10 to 100 times higher than for wild type strains. The plot below shows the PCR Ct value which is on the log scale, with lower numbers indicating higher viral loads, and a difference of 23 to 18 in Ct corresponding to a 2^5=32 fold increase in viral load.

This type of higher viral load may provide explanation for the increased transmissibility of the virus, but would also suggest that those infected with this variant are more susceptible to severe disease and death. A recent study showed that that the B.1.1.7 variant increases the risk of death by >50%, verifying that this variant indeed appears to produce more severe disease and higher risk of death.

For this reason, the increase of the B.1.1.7 variant poses significant risk for viral surge and increased hospitalizations and deaths from COVID-19.

The Good News: The UK B.1.1.7 variant is not an "immune escape" variant Fortunately, this variant appears to not be an "immune escape" variant, as studies have suggested that individuals recovered from previous infection maintain immune protection and are not more prone to reinfection, and that individuals vaccinated with the various vaccines appear to be fully protected relative to wild type variants. This is encouraging and will mitigate the effects of the B.1.1.7 variant in the USA given the very large number of people who have been exposed and infected in the USA to date and the relatively large and growing number vaccinated. There are currently 30 million PCR-confirmed cases in the USA, which is ~10% of the country, with likely 2-3x as many unconfirmed infections that have not been validated PCR providing another 60-90 million more. This means 90-120 million Americans have been exposed and infected with the virus to date, which is 30-40% of the country. Many of these people retain protection vs. reinfection, as confirmed in a recent Lancet study showed that reinfections were very rare, with individuals with previous infection protected 80% vs. reinfection at 6 months, and seniors (>65) protected nearly 50% vs. reinfection at 6 months, and in the Johnson and Johnson study those in the placebo group who were previously infected were protected >90% vs. infection relative to those in the placebo group without previous infection, as I mentioned in a previous blog post. Most of these people should maintain this level of protection against reinfection with the UK B.1.1.7 variant, and thus the high infection rate in the USA should provide some degree of protection against a variant-induced surge this Spring. Also, it appears that the current vaccines used in the USA retain near full efficacy vs. the UK variant. According to the CDC COVID data tracker, to date >82 million (24.9%) have received at least one dose and >44 million (13.5%) have been fully vaccinated, with the current vaccination rate over 2.5 million per day (0.75% of population per day, >5% per week). At this rate, we could arrive at 50-55% of the USA population receiving first dose and 40-45% by the end of May, and with vaccination increasing and Novavax and possibly AstraZeneca coming online in the next month, these numbers could be higher.

Further, a recent study of essential workers demonstrated the mRNA vaccines provided substantial protection vs. infection (symptomatic or asymptomatic) after the first dose (80% efficacy), so expect that all those two weeks past first dose are substantially protected.

Based on data to date, it appears the vaccinated population should be significantly protected against infection vs. this UK B.1.1.7 variant.

The race between vaccination and UK B.1.1.7 variant spread!

In a sense we are in a race to vaccinate quickly enough to prevent this variant from producing a major springtime viral surge in the USA. Had this variant hit in early November when the USA had none vaccinated and a MUCH lower proportion of the population at least partially protected (i.e. on 11/1/20 only 10m confirmed cases and likely 40m-60m more unconfirmed cases, totaling only 15-20% of the population previously exposed and infected and none vaccinated), its higher transmissibility and severity would have likely made our winter surge much more devastating in terms of cases, hospitalizations and deaths.

Currently, we are much more protected, with perhaps 40-50% of the population having at least partial protection vs. the virus (30-40% previously exposed and infected and maybe 10-15% more who were not previously infected but have received at least first dose of the vaccine). This should take the edge off any propensity of the emergence of the UK B.1.1.7 variant to produce a spring surge. However, this still leaves at least half of the population susceptible to infection with this variant, so if we are not careful there are still plenty susceptible to fuel an uptick in cases this month that could endanger the reopening of schools and businesses that have recently occurred in many places. If those of us who are still not protected can maintain vigilance a little longer, the increasing vaccination rate should get us through the Spring and into the Summer without a major surge. It is likely that if we can maintain the current rate of vaccination distribution, we can realistically expect that by June we will have sufficient protection that the wild type and UK variant B.1.1.7 virus will not be able to sustain exponential growth, and the major danger posed by this variant will be mostly behind us.

At that point, our major concern is the vaccine durability and effectiveness vs. the "immune escape" variants like the South African B.1.351 and New York B.1.526 for which vaccine efficacy is less and susceptibility to reinfection is higher. There is evidence, however, that the vaccines maintain at least 2/3 of their protective efficacy vs. these variants, so I think it is plausible that with the current projected vaccination rates we can look forward to a good summer, and perhaps a return to close-to-normalcy by the fall.