The aim of the “Sentinel” national surveillance program is to monitor the circulating respiratory viruses, including SARS-CoV-2 variants, and hence underpin public health actions.
In week 22/2021, the overall frequency of the SARS-CoV-2 B.1.1.7 variant in all sequenced specimens decreased to 47,4% (CI 40,4% – 54,4%, p<0,05). No case was detected for the SARS-CoV-2 B.1.351 and P.1 variant, while the frequency the variant B.1.617.2 doubled to 30,9% (CI 24,4% – 37,4%, p<0,05).
The representative sample was estimated, based on the number of positive cases in Luxembourg for week 22 (324). The minimum sample size required to detect prevalence of B.1.1.7 (66%) reported in week 21, with an error margin of 5%, was estimated to be 168 specimens. This number corresponds to a coverage of 51,8 %, which exceeds the minimum coverage recommended by ECDC (10%). The sequencing results of week 22 are not representative of the circulating variants in Luxembourg with a margin of error of 5%.
The total number of sequences performed this week was 371, with 194 specimens having been collected in the time frame of week 22/2021. The sequencing coverage – based on sequenced samples collected in week 22 and corresponding to Luxembourg residents and to non-targeted collections – was 45,1% of all positive cases in Luxembourg.
The “Sentinel” surveillance network reported 222 consultations in week 22 (31/MAY/2021 – 06/June/2021). There was 1 case of ILI1, corresponding to 0,5% of the consultations, as shown in Figure 1. The number of consultations for ARI2 was 18, which represents 8,1% of the consultations.
Covid Consultation Centres have been closed by 16/May/2021. We are currently working on an alternative in partnership with private laboratories.
The National Reference Laboratory for Acute Respiratory Infections at LNS continues to improve the representativeness of the pool of sequenced specimens to reach real-time epidemiology, by implementing the following weekly sequencing activities:
The representative sequencing sample was based on the minimum number of specimens required to extrapolate prevalence of VOC variants with error rate of 5%. The representative sample was estimated based on the number of positive cases in Luxembourg in week 22 (324). The minimum sample size required to detect prevalence of B.1.1.7 (66%) reported in week 20 with an error margin of 5% was estimated to be 168 specimens. The calculation was based on a sample size calculation tool that uses the expected prevalence of the variant in the total population. (Population Proportion – Sample Size – Select Statistical Consultants (select-statistics.co.uk). This number represented a coverage of 51,8% which exceeds the minimum coverage recommended by ECDC (10%). The number of non-targeted specimens from Luxembourgish residents sequenced this week was 146. Therefore, our sequencing results this week are not representative of the circulating variants in Luxembourg.
The starting material used for sequencing is respiratory specimens (nasopharyngeal or oropharyngeal swabs) that have already been tested positive by RT PCR.
In parallel, we continue to evaluate commercial PCR screening, using the same starting material, to detect the Variants of Concern (VOCs), with specific PCR. This will enable a faster investigation time of any outbreak and allow us to screen 100% of positive cases referred to LNS, including those that do not pass the quality criteria for sequencing. Specimens that are not eligible for sequencing can thus be used in a PCR to detect the presence of variants of concern.
The LNS sequencing data sharing strategy includes sharing of the sequencing data with GISAID EpiCov database (www.gisaid.org ) on a periodic basis.
Last week the microbial genomics platform at the LNS sequenced 371 specimens, with 194 collected in week 22/2021. The sequencing pool referring to Luxembourgish residents represents 52,8% of new infections reported in Luxembourg in week 22/2021. Among these 194 specimens, 40 specimens were reported to be part of a cluster or outbreak investigation, and 23 specimens were from non-residents (15 specimen overlapping). This leads to 146 specimens, collected in week 22, and being the representative population sequencing sample. In the population representative sample of residents, the frequencies of B.1.1.7 and B.1.617.2 were 46,6% and 30,8% respectively. No cases have been detected for the B.1.351 and P.1 variants.
The population sequencing coverage in week 22/2021 was 45,1% (Figure 2). Based on statistical inference, the frequency of the reported variants in week 22/2021 is not representative of the circulating variants in Luxembourg with a margin of error of 5%.
Lineages (variants) have been assigned based on Rambaut et al by means of Phylogenetic Assignment of Named Global Outbeak LINeages (pangolin) software (v3.0.3, pangoLEARN version 2021-05-27).
The lineage nomenclature system that we use is the one proposed by Rambaut et al. that focuses on actively circulating virus lineages (https://cov-lineages.org ).
In the week 22/2021, in the population representative sample, after removal of cluster specimens, and excluding specimens collected from non-residents, there were 13 circulating SARS-CoV-2 variants, with the main three variants being B.1.1.7 (46,6%, CI 38,5% – 54,7%), B.1.617.2 (30,8%, CI 23,3% – 38,3%) and B.1.1 (5,5%, CI 1,8% – 9,2%), as shown in Figure 3.
Among specimens collected within the week 22/2021, 92 cases of the B.1.1.7 variant have been detected, representing 47,4% of the specimens in the week’s sequencing pool (by comparison, the week 21/2021 pool had shown a frequency of 65,2% of this variant, including additional specimens having been sequenced from previous weeks). The total case count of sequenced variant B.1.1.7 was 6230 by week 22/2021.
In the collection period of week 22/2021, no case of the South African variant B.1.351 has been detected (by comparison, the week 20/2021 pool had shown 3,1% of this variant, including additional specimens having been sequenced from previous weeks). The total case count of sequenced variant B.1.351 was 1048 by week 22/2021. The earliest collection date for this variant remains 11/JAN/2021 and the latest is 29/MAY/2021.
In week 22/2021, no new case of B.1.525 and P.1 has been detected. The case count by week 22 for B.1.525 remains 45 (latest sampling date 17/MAY/2021) and for P.1 the case count increases due to retrospective sequencing to 112 (latest sampling date 30/MAY/2021).
In week 22/2021, 60 additional cases of the Delta variant B.1.617.2 have been detected (latest sampling date 06/June/2021). The case count by week 22 for B.1.617.1 remains 6, and for B.1.617.2 it raises to 149. Since May 6th B.1.617.2 has been escalated by Public Health England from a variant under investigation to a variant of concern (Figure 4).
Lineage B.1.1.7 is characterized by several spike protein mutations, including N501Y, H69/V70del and P861H. The variant seems to have a considerable epidemiological impact, as it has a higher transmissibility rate.
Lineage B.1.351 holds numerous spike protein mutations, of which three are located in the receptor binding domain (K417N, E484K and N501Y), and are therefore relevant for antibody binding. As for B.1.1.7, a higher transmissibility rate and viral loads seem to be associated with this variant. Due to the K417N and E484K mutations, an impact on vaccination efficacy and possibility of reinfection is subject to scientific investigation.
Lineage P.1 (descendent of B.1.1.28), initially found in the Amazon region, has a similar mutation profile as the South African variant, including E484K and N501Y. Concerns are, as for the South African variant, higher transmissibility and a decreased protection by neutralizing antibodies.
Lineage B.1.525 carries several mutations of biological significance, including E484K, Q677H and F888L. It does not carry N501Y, but a set of deletions similar to the B.1.1.7 variant.
Lineage B.1.617 is a variant first detected in India and was designated “Under Investigation” on 1st April 2021 by Public Health England. It contains a number of spike mutations associated with antigenic escape or found in other variants of concern, including L452R, E484Q and P681R. Subtype B.1.617.2 does not carry S:E484Q and seems to more transmissible as B.1.1.7 (increasing confidence). Neutralization studies show reductions in cross-neutralizing activity between B.1.1.7 and B.1.351.
By week 22/2021, 78,3 % of the variants detected in the sequenced specimen pool are declared as either Variants of Concern. These include B.1.1.7, which is continuously decreasing, apparently being displaced by B.1.617.2.
The ReViLux will continue to use the “Pango” nomenclature system to allow easier links to any evolving variants with their ancestor (Figure 5).
Currently the LNS genomic surveillance program – independently from lineage calling – notes the occurrence of 13 different known SARS-CoV-2 mutations, assumed to have clinical and epidemiological relevance. The list of observed mutations is being updated continually, based on the appearance and prevalence of SARS-CoV-2 variants.
The following table provides the overall frequencies of these mutations, detected in the lineage-assignable genome sequences, analyzed between 01/SEP/2020 and 30/MAY/2021 (N=12080), as well as the frequencies in week 22/2021.
Genomic sequencing of SARS-CoV-2. A guide to implementation for maximum impact on public health. WHO, 8 January 2021.
COVID-19 data portal. 2020 (https://www.covid19dataportal.org/sequences )
J Hadfield et al. Nextstrain: real-time tracking of pathogen evolution. Bioinformatics 2018;34:4121-4123
A Rambaut et al. A dynamic nomenclature proposal for SARS-CoV-2 lineages to assist genomic epidemiology. Nat Microbiol 2020;5:1403-1407