--- title: "The rise of the Delta Covid variant" full_title: "Changing composition of SARS-CoV-2 lineages and rise of Delta variant in England" authors: - Swapnil Mishra - Sören Mindermann - Mrinank Sharma - Charles Whittaker - Thomas Mellan - Thomas Wilton - Dimitra Klapsa - Ryan Mate - Martin Fritzsche - Maria Zambon - Janvi Ahuja - Adam Howes - Xenia Miscouridou - Guy P. Nason - Oliver Ratmann - Gavin Leech - Julia Fabienne Sandkühler - Charlie Rogers-Smith - Michaela Vollmer - Juliette T. Unwin - Yarin Gal - Meera Chand - Axel Gandy - Javier Martin - Erik Volz - Neil M. Ferguson - Samir Bhatt - Jan M. Brauner - Seth Flaxman gleech_role: co-author year: 2021 venue: "The Lancet — eClinicalMedicine" type: journal doi: 10.1016/j.eclinm.2021.101064 url: https://authors.elsevier.com/sd/article/S2589537021003448 code: https://github.com/ImperialCollegeLondon/SARS_CoV_2_variants_uk links: explainer: https://twitter.com/JanMBrauner/status/1421549366926614534 discussion: https://www.imperial.ac.uk/news/221873/uk-covid-19-transmission-shifted-towards-emerging/ contribution_hours: 10 topics: [covid-19, genomics, variants, delta, surveillance] --- ## Abstract Using tests and sewage data from early 2021, we saw that "the English variant" of COVID (B.1.1.7), which took over England in December, was itself being displaced by other variants. The main worry was that one of the new variants would be resistant to the vaccines. ## Full text > Converted from the open-access Lancet eClinicalMedicine full text (Europe PMC, > PMC8349999). Markup removed; figures/supplement and the reference list omitted. **Background.** Since its emergence in Autumn 2020, the SARS-CoV-2 Variant of Concern (VOC) B.1.1.7 (WHO label Alpha) rapidly became the dominant lineage across much of Europe. Simultaneously, several other VOCs were identified globally. Unlike B.1.1.7, some of these VOCs possess mutations thought to confer partial immune escape. Understanding when and how these additional VOCs pose a threat in settings where B.1.1.7 is currently dominant is vital. **Methods.** We examine trends in the prevalence of non-B.1.1.7 lineages in London and other English regions using passive-case-detection PCR data, cross-sectional community infection surveys, genomic surveillance, and wastewater monitoring. The study period spans 31 January 2021 to 15 May 2021. **Findings.** Across data sources, the percentage of non-B.1.1.7 variants increased since late March 2021, initially driven by a variety of lineages with immune escape. From mid-April, B.1.617.2 (Delta) spread rapidly, becoming the dominant variant in England by late May. **Interpretation.** The outcome of competition between variants depends on intrinsic transmissibility, evasion of prior immunity, demographics, and interactions with NPIs. The presence and rise of non-B.1.1.7 variants in March was likely driven by importations and some community transmission; competition between them resulted in B.1.617.2 becoming dominant in April–May with considerable community transmission. Early detection of new variants requires a diverse array of community-surveillance data sources. ### Introduction B.1.1.7 (epidemiologically 50–80% more transmissible, more severe) rose from ~0% to >98% of sequenced English samples, necessitating a third national lockdown in January 2021. Its 69–70 Spike deletion causes S-gene target failure (SGTF) in PCR, a fast proxy for genomic surveillance (SGTF frequency rose from ~0% in October 2020 to 98.8% in March 2021). Other VOCs — B.1.351 (South Africa), P.1 (Brazil), B.1.617.2 (India) — lack the 69–70 deletion (so are S-gene positive, "S+") and carry immune-escape mutations (E484K or T478K). By early April 2021 ~55% of the English population were seropositive (infection or vaccination); such immunity is a selection pressure that could give partially immune-escaping VOCs a fitness advantage, especially as control measures relaxed and the rollout relied heavily on AstraZeneca (highly protective against B.1.1.7, possibly reduced against other VOCs). ### Methods (four data sources) - **Pillar 2 symptomatic community testing.** ~30% of Lighthouse-lab samples use the ThermoFisher TaqPath assay including Spike; SGTF is an accurate B.1.1.7 proxy at non-Spike Ct ≤ 30. Over 31 Jan – 15 May 2021 (symptomatic, non-lateral-flow): 72,881 S+ and 586,854 S− cases in England (4,246 S+ / 79,207 S− in London). - **ONS infection survey.** A fortnightly random-household PCR survey (139,948 participants in the fortnight before 16 April 2021); we use the "UK variant compatible" vs "not compatible" gene patterns. - **Sewage-water monitoring.** Fortnightly samples from Beckton Sewage Treatment Works (~4 million people, North London) — low sampling bias, capturing everyone in the catchment. - **COG-UK genomic sequencing.** 10,324 sequences from Greater London Pillar 2 testing after 1 March 2021; 2,957 non-B.1.1.7 (including B.1.617.2 n = 2,225, B.1.351 n = 254, P.1 n = 81). Lineage frequencies over time via a Gaussian-process multinomial GAM; analysis repeated excluding travel-linked cases. ### Results Since early March 2021, S+ case incidence (Pillar 2) increased against falling-then-stable overall case numbers, earliest in London but in every region. ONS S+ prevalence rose slightly in March, dipped, then rose strongly in early May (lagging Pillar 2, with sampling variability at low incidence). **Ct values:** until March, S− (mostly B.1.1.7) samples had lower Ct (higher viral load) than S+; since end-March, mean S+ Ct decreased to comparable levels, suggesting increased S+ transmission. **Sewage:** B.1.1.7-defining mutations (HV69–70del, Y144del, A570D), stable >95% from January to mid-March, fell to 67–75% by 13 April and 28–49% by 11 May; E484K rose above 30% by mid-April; B.1.617.2 mutations rose to 41–62% by mid-May (≈half the viral population). **COG-UK sequencing:** non-B.1.1.7 in March was chiefly B.1.351 and B.1.525, then B.1.617.2 rose rapidly over April to >75% of London sequences by late May — a similar pattern when excluding travel/surge-testing cases, indicating community transmission. ### Discussion Using four independent data sources, we show recent increases in the proportion of S+ infections — first a variety of immune-evading lineages, then overwhelmingly B.1.617.2. A key question when a new variant's proportion rises is whether it reflects local transmission or imported infections against low overall incidence; by the time of writing, B.1.617.2 was dominant (>75% of London sequences even excluding travel), indicating sustained community transmission. A considerable rise in non-B.1.1.7 was apparent across sources by early/mid-April, giving an early warning — though the community-transmission signal was then ambiguous (only ~10% of non-travel-linked cases were non-B.1.1.7 in mid-April; Ct and sewage signals could partly reflect importation given London's airports). Variant competition depends on intrinsic transmissibility, immune escape, and targeted NPIs; even if less intrinsically transmissible than B.1.1.7, immune-escaping VOCs may gain an advantage in a highly immunised population (cf. rapid resurgences in Manaus with P.1 and Delhi with B.1.617.2 despite high prior immunity). The results underscore the value of a diverse array of community-surveillance data sources and timely genomic surveillance for real-time lineage tracking — critical to immediate control and future vaccine development. Data/code: github.com/ImperialCollegeLondon/SARS_CoV_2_variants_uk.