Tuesday, December 29, 2020

UK PHE Technical Briefing #2: Investigation of novel SARS-CoV-2 variant


 

#15,662

It is just over two-weeks since the `UK Variant' (aka B.1.1.7) was announced by UK Health Secretary, and while there is still much we don't know about the impact of this mutated virus, international reaction has been both swift and aggressive (see Japan Bans Entry To All Foreign Nationals Over COVID Variant Fears).

Early, mostly epidemiological, investigations have suggested this varian may be between 50% and 70% more transmissible than its COVID predecessors (see PrePrint: Estimated Transmissibility & Severity Of UK SARS-CoV-2 Variant - CMMID), although some of this evidence is, admittedly, anecdotal.

So far, we've seen no evidence that this produces more severe illness in humans, or would evade the current crop of vaccines, although investigations continue.  Nevertheless, even if all other aspects of the virus remained the same, a significant jump in transmissibility would be of deep concern. 

Complicating matters, many countries don't do a lot of genomic sequencing of the virus, meaning that B.1.1.7 - and other variants - are likely circulating unnoticed.  One piece of luck with the UK variant is that some PCR tests use a specific three target assay (N, ORF1ab, S), and the UK variant virus fails on the S-gene target. 

While it is possible that other variants might return this same (2 positive, 1 negative) result, early testing suggests that 97% of these S gene dropouts (dubbed S gene target failure (SGTF)) in the UK are a result of the UK variant, making these tests a reasonable (and rapid) proxy for genomic sequencing in regions where they are used. 

Yesterday the UK's PHE released their second technical briefing on the UK variant, and while much of the data is preliminary, their first case (edited-mpc) matched cohort study suggests that infection with this B.1.1.7 variant doesn't result in increased mortality or morbidity. 

The 12-page PDF is very much worth downloading and reading in its entirety, the matched cohort  study is the biggest news.  I've excerpted the relevant sections below. 

Preliminary findings of matched cohort study 

A matched cohort study was undertaken to inform a preliminary assessment of outcomes of hospitalisation and case fatality associated with VOC 202012/01. This analysis of sequenced positive SARS-CoV2 cases used confirmed VOC cases matched to confirmed wild-type comparator cases (classified as distinct to the variant sequence). To optimise comparability of the variant (VOC 202012/01) and wild-type cases and manage the impact of non-random sampling of SARS-CoV-2 cases for sequencing, variant cases were frequency matched to wild-type cases on a 1:1 basis by age group, sex, upper tier local authority (UTLA) of residence and two-week time-period for specimen date. 

Of the 2,693 variant cases identified at the time of analysis, 1,769 variant cases with specimen dates between 20 September and 15 December 2020 were matched to 1,769 wild-type comparator cases and were included in this analysis. In reflection of the matching criteria, the median age of variant cases was 36 years and 35 for the wild-type cases. 51.4% of both the variant cases wild-type comparator cases were female. 

The comparison of the 3,538 variant and wild-type cases showed that the majority of variant cases were of White ethnicity (75.2%) followed by Asian (10.1%) and Black (5.9%). The ethnic profile of wild-type comparator cases was broadly similar but a higher proportion of Asian ethnicity (13.5% v 10.1%). 

The majority of variant cases were resident in private dwellings (95.0% in variant cases and 94.3% in wild-type comparator cases). Variant cases were more likely to be part of a residential cluster (defined as all laboratory confirmed cases occurring at the same Unique Property Reference Number (UPRN) within 14 days of each other) compared to wild-type comparator cases (63.5% vs 56.1%, Chi-Squared test p=0.00).

Review of hospital admissions data from the NHS identified that of the 3,538 cases, 42 individuals had a record of hospital admission after the date of specimen. Fewer variant cases (16 cases (0.9%)) were admitted to hospital compared to wild-type comparator cases (26 cases (1.5%)) but the difference was not significant (Chi-squared test p=0.162). Due to potential time delays for receipt of hospital admissions data, the identified hospital admissions should be regarded as a minimum number of hospital admissions and further admissions data are likely to be received into this NHS dataset in the future.

The 28-day case fatality was assessed for variant cases and comparator cases. Analysis was restricted to 2,700 cases with a full 28 days elapsed since the specimen date. Among variant cases, 12 of 1,340 (0.89%) variant cases died within 28 days of their specimen date compared with 10 of 1,360 (0.73%) wild-type comparator cases; this difference was not significant (Odds ratio:1.21, p=0.65).         

(SNIP) 

Summary

The findings from the analysis of data on SGTF cases showed a roughly similar spatial distribution of cases as observed from mapping of genomic data. SGTF cases were mostly observed in the South East, London parts of the South West regions and Cumbria. This regional variation in should be interpreted in the context of limited coverage in regions like East of England of by the three Thermofisher TaqPath lighthouse labs. However, in regions with relatively high and consistent coverage, the findings can be used as a proxy for the burden of VOC 201212/01 infection. 

Preliminary results from the cohort study found no statistically significant difference in hospitalisation and 28-day case fatality between cases with the variant (VOC 201212/01) and wild-type comparator cases. There was also no significant difference in the likelihood of reinfection between variant cases and the comparator group

(Continue . . . ) 

While this case control study is somewhat reassuring, these results should be interpreted with caution, as the data is still quite limited, and these variants - which continue to evolve - are a moving target.  

What we say about them - or others like them (like the 501Y-V2 variant from South Africa) - today, may not hold true next week or next month. 

There are still open questions about relative transmissibility, and vaccine effectiveness against these variants. But for now, after months of adversity and setbacks with this pandemic, I'm grateful for any glimmer of `good news' we can get.