The new variant of the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) detected in England replicates more than two times faster than the earlier strains according to new research. This replication advantage promises to make it the dominant global variant very shortly. This finding comes from a startling preprint that appeared on the medRxiv* server in January 2021.
It was on September 20, 2020, that scientists first detected the first two genomes that are now labeled as the novel SARS‐CoV‐2 Variant Under Investigation (VUI)‐202012/01, also known as the B.1.1.7 lineage. The next 30 days saw the exponential growth of the lineage, which comprised over 1/3 of all SARS-CoV-2 sequences in England in the first half of December.
Characteristics of B.1.1.7
The new variant has nine mutations in the spike protein, all of which are not unique to it. One of them is a deletion, del 6-70HV, which makes the spike gene invisible to the dPCR probe used in England. This is called spike gene target failure, SGTF.
This strain has become the most common strain to carry this deletion due to its high multiplication rate. It has been found, by scientists at Public Health England (PHE), to have a reproduction number that is almost 1.5 times higher than earlier strains, while others showed its estimated transmissibility to be 56% higher.
Exponential growth rate
The current study uses a new approach. The scientists gathered data from all over England, from the Global Initiative for Sharing All Influenza Data (GISAID) database, to find the ratio of genomes from VUI‐202012/01 to non‐VUI‐202012/01. This data covers about 5% of genomes from England.
The researchers tracked this ratio and calculated the growth rate of the number of sequenced genomes in these two categories from week to week, as over the five-week period of the whole study. They found that the growth of VUI‐202012/01 was virtually exponential over weeks 43-47 of 2020. The ratio of the replication numbers of the two types of genomes was estimated.
Over the four weeks from week 38 to week 42, the researchers found that there were less than five genomes from the VUI‐202012/01 being collected each week. However, the number shot up to 28 in week 43 and continued to rise exponentially until week 47. In week 48, this growth seems to come to an end. But this may be more apparent than real, perhaps because the data coming in was incomplete.
In the five-week period, the growth rate fraction of VUI‐202012/01 to non‐VUI‐202012/01 strains was at over 2 per week – in fact, it was 2.31. The ratio of reproduction numbers of these two types of variants was estimated on the assumption that both types of strain have the same serial interval (interval from symptom onset of the primary case to that of secondary cases), namely, 6.73 days. This ratio came to 2.316.73/7= 2.24.
The ratio of VUI-202012/01 to non-VUI-202012/01 genomes collected in weeks 38–50, 2020, based on GISAID data submitted till Dec 26, 2020. The trend line in (A) is fitted to data points from weeks 43–47 (blue), in which the number of collected VUI-202012/01 genomes exceeds 27 and the total number of genomes exceeds 4500 per week. The weekly growth rate is 2.31 [95% CI: 2.08–2.57]. The number of positive cases in (B) is given after https://coronavirus.data.gov.uk/details/cases.
Excluding other factors
A smaller subset of sequenced variants was studied by using an earlier cut-off date for sequence submission. Setting the cut-off at December 16 rather than December 26 reduced the number of VUI‐202012/01 genomes by 200. This work resulted in a calculated reproduction number advantage of 2.3 times in favor of VUI‐202012/01 strains.
Again, since there were only 19 cases of this lineage occurring outside England during weeks 43-48, the researchers used data from England rather than the UK. However, using whole-UK values instead still resulted in a ratio of 2.3.
To rule out the effect of a chance aggregation of cases caused by the new variant, the researchers looked at the confidence intervals again. They found that the confidence intervals calculated on the assumption that the genomes sequenced from the new variants were drawn at random from a binomial distribution were narrower than their own results, calculated from the standard error of the slope. This supports the accuracy of the data on which the current study is based.
The serial interval was calculated as equal for both strains. However, this could be wrong, as the serial interval is not a number but has a sub-exponential distribution. The researchers comment, “Although no current data indicate this, it may happen that the faster spread of VUI‐202012/01 strain partially results from the shorter generation time.”
What are the implications?
These findings indicate that the reproduction number of VUI‐202012/01 strains is 2.24 times higher than that of non‐VUI‐202012/01 strains of SARS‐CoV‐2. This difference in replication number is not the result of or affected by public health interventions such as protective measures because the growth rates were measured prospectively.
With London being a central hub connecting most of the transportation routes the world over, the new lineage may have already covered Europe and the USA to become globally dominant. In fact, the day after the study ended, there were 12 of these genomes from Italy, nine from Denmark, 7 from Ireland, 6 from Japan, and several from other European and Asian countries.
“If confirmed, spread of the faster‐replicating VUI‐202012/01 strain may hinder the efforts to contain the COVID‐19 epidemics prior to massive vaccinations.”
medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.