Guidance

SARS-CoV-2 lateral flow antigen tests: evaluation of VUI-202012/01

Published 23 December 2020

1. Working with academic collaborators at the University of Oxford, PHE Porton Down has been evaluating the performance of lateral flow devices (LFDs) since mid-August 2020. Over 60 have been considered to date, approximately 30% of which have progressed to extended evaluation on clinical samples at Phase 3.

2. As part of the most recent round of Phase 3 testing, the performance of five LFDs has been evaluated by the scientific team here against a panel of recent clinical samples from the Lighthouse Laboratory at Milton Keynes that contain the new variant. They have also been evaluated using material cultured in recent days from the variant that was sent here by colleagues at PHE Colindale on 14 December. The following LFDs were evaluated, which target nucleocapsid protein:

  • Abbott Panbio
  • Fortress
  • Innova
  • Roche / SD Biosensor nasal swab
  • Surescreen

The Innova LFD is now being widely rolled out across the country and the Abbott LFD is being field-tested in a small number of settings. The other three LFDs above will be field-tested in the New Year to complete their Phase 3 evaluation.

3. DHSC provided us with data on the Milton Keynes samples that identified samples with the variant on the basis that the PCR on the ThermoFisher platforms did not detect the S gene. As part of our evaluation, these samples were subsequently run on the Roche cobas PCR platform that has been used throughout the validation programme and have e-gene target CT values ranging from 16 to 36. The results for the samples with S gene drop out and CT <27, which from our LFD evaluation programme correlates to infectiousness, is summarised in the table below:

LFD Abbott Fortress Innova Roche / SD Biosensor swab Surescreen
Number positive / number tested 25/27 101/112 25/27 98/112 100/110
Kit failures 0 0 0 0 2

The LFDs successfully detected samples containing the new variant.

4. The amino acid substitutions in the 5 variant samples provided from PHE Colindale were identical and are set out in the table below:

nsp3: T181I; T183I; A890D; A1305V; I1412T
nsp6 deletion: codons 106-108
nsp12: P323L
nsp13: K460R
nsp14: E347G
S: N501Y; A570D; D614G; P681H; T716I; S982A; D1118H
S deletion: nucleotides 21765-21770
S deletion: nucleotides 21991-21993
ORF 8: Q27*; R52I; Y73C
N: D3L; R203K; G204R; S235F

One of the 5 identical samples was run on the LFDs, the results of which are set out in the table below:

200ul diluted VUI-202012/01 cultured virus in saliva tested according to LFD manufacturers’ instructions for use.

Sample dilution Abbott (LFD) Fortress (LFD) Innova (LFD) Roche / SD Biosensor swab (LFD) Surescreen (LFD)
10 4 pfu/ml 3/3 3/3 3/3 3/3 3/3
10 3 pfu/ml 3/3 3/3 3/3 3/3 3/3

The LFDs successfully detected the cultured samples containing the new variant.

5. In summary, the LFDs evaluated, all of which target the nucleocapsid protein, have detected the new variant that contains four amino acid changes from the original viral sequence. This does not affect their performance and we will monitor further variant changes as they arise as part of our ongoing evaluation programme.