Rapid epidemic expansion of the SARS-CoV-2 Omicron variant in southern Africa

Since late 2019, repeated emergence of SARS-CoV-2 variants has driven asynchronous waves globally. Early variants of concern (Alpha, Beta, Gamma) spread widely, but Delta displaced them across most regions. Contrary to expectations that the next VOC would derive from Delta, a genetically distinct lineage (B.1.1.529, Omicron) emerged in southern Africa as a large Delta wave subsided. This study aims to characterize Omicron’s genomic features, evolutionary origins, and early transmission dynamics, and to assess its growth advantage relative to Delta and the potential roles of intrinsic transmissibility versus immune evasion. Understanding these factors is critical for forecasting epidemic trajectories and informing public health responses.

Background references establish that prior waves in southern Africa were driven by Beta and then Delta, with Delta evolving multiple sublineages and becoming globally dominant. Prior work mapped mutations impacting ACE2 binding and antibody escape in the receptor-binding domain and N-terminal domain, and documented T cell epitope conservation. Studies on furin cleavage site–proximal mutations (e.g., P681R/H) and nucleocapsid mutations (R203K/G204R) suggest effects on transmissibility and infectivity. Serosurveys indicated substantial prior exposure (40–60% before Delta; >70% in Gauteng by late 2021), and vaccination coverage was around 40% of adults, implying high population immunity against Delta. These findings provide context for evaluating whether Omicron’s rapid spread is due to immune evasion, increased transmissibility, or both.

Surveillance and detection: Investigators monitored SARS-CoV-2 incidence, test positivity, and S-gene target failure (SGTF) on the TaqPath PCR assay as a proxy for Δ69–70. Targeted whole-genome sequencing prioritized SGTF samples and routine diagnostic specimens across South African provinces and Botswana. Genomic datasets: As of early December 2021, analyses incorporated 686 Omicron genomes (248 southern Africa, 438 global) within a global reference of 12,609 SARS-CoV-2 genomes sampled Dec 2019–Nov 2021. Subsequent phylogenetic and phylodynamic analyses focused primarily on BA.1, with BA.2 and BA.3 assessed as data allowed. Phylogenetics: Preliminary maximum-likelihood phylogenies positioned BA.1 as a monophyletic clade within B.1.1 (Nextstrain 20B), phylogenetically distant from known VOCs/VOIs and regional lineages (e.g., C.1.2). Phylodynamics: Time-calibrated Bayesian analysis (southern Africa BA.1, n=553, as of 11 Dec 2021) estimated TMRCA and exponential growth rates/doubling times. Birth–death skyline (BDSKY) models, accounting for variable sampling through time, estimated doubling times and effective reproduction numbers (R), including province-specific analyses (e.g., Gauteng) with multi-epoch models. Phylogeography: Spatiotemporal reconstruction inferred spread among South African provinces and into Botswana regions, with uncertainty quantified via 80% HPD regions. Molecular characterization: Comparative analysis of spike mutations (RBD/NTD), furin cleavage site–adjacent changes, and mutations/deletions outside spike (nsp6 del105–107; N R203K/G204R). Structural mapping highlighted epitopes and potential impacts on monoclonal antibody binding and T/B cell epitopes. Recombination analyses: Applied GARD, 3SEQ, and RDP5 to detect recombination signals among BA.1/BA.2/BA.3 and diverse background lineages, localizing potential breakpoints and evaluating phylogenetic support across genome partitions. Selection analyses: Implemented a selection pipeline (e.g., BUSTED for gene-wide positive selection; MEME for episodic selection at sites; RELAX/FADE where applicable) on downsampled BA.1/BA.2/BA.3 alignments alongside background sequences. Assessed intensification factors (K) relative to reference lineages and identified clade-defining and post-MRCA selected sites. Growth advantage modelling: Multinomial logistic regression estimated relative growth of Omicron vs Delta in Gauteng. Scenario analyses explored combinations of intrinsic transmissibility increases and immune evasion across assumed levels of population immunity against Delta (Ω = 0.4, 0.6, 0.8), informed by seroprevalence and vaccination coverage.

Epidemic dynamics and detection: Following a low-transmission period (mid-Oct to mid-Nov 2021; 100–200 daily cases; <2% positivity), Gauteng experienced a rapid resurgence mid-November. SGTF frequency rose sharply, and sequencing of SGTF and routine samples revealed a new lineage widespread in Gauteng by the second week of November. By late November to early December, weekly test positivity in Gauteng increased from <1% (week of 31 Oct) to 16% (week of 21 Nov) and 35% (week of 28 Nov). National daily cases exceeded 22,000 by 9 Dec 2021 (84% of prior peak), and SGTF reached ~90% nationally by the week beginning 21 Nov, consistent with Omicron driving the fourth wave. Botswana cases doubled every 2–3 days, rising from <10 to >25 per 100,000 within ~10 days. By 16 Dec 2021, Omicron was detected in 87 countries; by 1 Jan 2022, >100,000 Omicron genomes from >100 countries were available. Evolutionary origin and spread: BA.1 forms a distinct clade within B.1.1 with no clear basal progenitor and is phylogenetically distant from other VOCs/VOIs. Related sublineages BA.2 and BA.3 share many but not all BA.1 mutations and have unique changes, evolving independently from the same B.1.1 node. Time-calibrated analyses dated the BA.1 TMRCA to 9 Oct 2021 (95% HPD: 30 Sep–20 Oct) with exponential growth rate 0.137/day (95% HPD: 0.099–0.175), implying a doubling time of 5.1 days (95% HPD: 4.0–7.0). Gauteng-only analyses indicated faster doubling (2.8 days; 95% HPD: 2.1–4.2). BDSKY estimated a doubling time of 3.9 days (95% HPD: 3.5–4.3) and R=2.79 (95% HPD: 2.60–2.97) across South Africa/Botswana; Gauteng R was 3.86 (95% HPD: 3.43–4.29) or 3.61 (95% HPD: 3.20–4.02) depending on epoch model. Phylogeography suggested seeding from Gauteng to most South African provinces and into Botswana in late Oct–late Nov 2021, with ongoing interprovincial transmissions. Molecular profile: BA.1 carries 15 RBD mutations; five (G339D, N440K, S477N, T478K, N501Y) individually enhance hACE2 binding. Seven RBD changes (K417N, G446S, E484A, Q493R, G496S, Q498R, N501Y) are predicted to reduce binding of multiple classes of neutralizing antibodies. NTD substitutions (A67V, T95I, G142D, L212I), deletions (69–70, 143–145, 211), and an insertion (EPE at 214–215) further underpin antibody escape. Despite spike alterations, >70% of T and B cell epitopes are predicted to remain unaffected. Mutations near the furin cleavage site (H655Y, N679K, P681H) may enhance cleavage/fusogenicity; nsp6 del105–107 may aid innate immune evasion; N R203K/G204R is associated with enhanced infectivity. Recombination: Weak evidence supported a single recombination event among BA.1/BA.2/BA.3 within the spike NTD region (breakpoints ~21690 and ~22198), but methods could not identify the recombinant definitively; no evidence that the MRCA of BA.1/BA.2/BA.3 was recombinant. Selection: BA.1 exhibited strong gene-wide positive selection in 11 genes/ORFs (e.g., S, exonuclease, RdRp, helicase, ORF3a/6/7a, M, N, nsp3), with dN/dS>5 and burst-like patterns. Selection was intensified relative to background in S (K=2.1), exonuclease (K=3.50), nsp6 (K=2.4), RdRp (K=1.14), and M (K=4.6). Among 1,546 polymorphic BA.1 sites, 45 showed episodic positive selection (MEME P≤0.01), including 23 in S; three post-MRCA S sites (346, 452, 701) converged on mutations seen in other VOCs/VOIs (R346K, L452R, A701V). BA.2 showed strong selection on S (P<0.0001; K=6.25) with selected S sites at 371, 376, 405, 477, 505 and M sites at 19, 63. BA.3 showed selection on S sites 67, 371, 477, 505 and N sites 13, 413. Growth advantage and immune evasion: In Gauteng, Omicron had a growth advantage of 0.24/day (95% CI: 0.16–0.33) over Delta, corresponding to a 5.4-fold (95% CI: 3.1–10.1) weekly increase relative to Delta. Scenario modelling indicates that, given likely high population immunity against Delta (≥60%), partial immune evasion by Omicron can largely explain the observed growth advantage without requiring substantial intrinsic transmissibility increases. Nevertheless, changes in intrinsic transmissibility relative to Delta cannot be excluded.

The rapid dominance of Omicron in settings with substantial prior infection and moderate vaccination coverage suggests immune evasion is a principal driver of its growth advantage over Delta. Molecular evidence (numerous RBD/NTD mutations impacting neutralization, with conservation of most T cell epitopes) aligns with reduced neutralizing antibody sensitivity while preserving cellular immunity, helping explain increased reinfections and breakthrough infections with comparatively constrained severe disease protection from vaccines. Phylodynamic estimates (short doubling times and elevated R) and phylogeographic patterns underscore swift dissemination from Gauteng to other regions. Although recombination does not appear to underlie the BA.1/BA.2/BA.3 common ancestor, adaptive evolution and potential limited recombination events within spike may have contributed to Omicron’s mutational constellation. The findings address the core question by quantifying Omicron’s growth advantage, timing, and spread, and by attributing its epidemiological success primarily to immune evasion, with possible contributions from changes in intrinsic transmissibility due to spike and cleavage-site–proximal mutations.

Rapid genomic surveillance in South Africa and Botswana enabled near-immediate identification of Omicron and global notification, affording time for preparedness. Omicron has driven a fourth epidemic wave in southern Africa and spread globally. Genotypic and early phenotypic evidence supports substantial escape from neutralizing antibodies, and modelling indicates immune evasion is a major driver of its observed growth. Continued monitoring across diverse epidemiological contexts is required to refine estimates of transmissibility and immune escape. Because cellular immunity is predicted to be less affected, vaccination remains critical to protect against severe outcomes. The emergence and rapid spread of Omicron pose a global threat, especially in regions with low vaccination coverage, underscoring the need for robust surveillance, rapid data sharing, and equitable vaccine access. Future research should clarify Omicron sublineage differences (BA.1/BA.2/BA.3), quantify immune escape across immune histories, and track ongoing adaptive changes.

Estimates of Omicron’s growth advantage are based on early sequence data that may be biased by targeted sequencing of SGTF samples and stochastic effects (e.g., superspreading) in a low-incidence context, potentially inflating growth advantage and inferred transmissibility or immune evasion. Uncertain and heterogeneous levels of protective immunity against Delta in the population preclude precise separation of intrinsic transmissibility changes from immune escape effects. Phylogeographic inferences may change with additional genomic data from under-sampled regions. Recombination detection methods have limited power to identify small or complex recombination events, constraining definitive conclusions about recombination’s role.