Genomic epidemiology of SARS-CoV-2 under an elimination strategy in Hong Kong

Abstract

Hong Kong employed a strategy of intermittent public health and social measures alongside increasingly stringent travel regulations to eliminate domestic SARS-CoV-2 transmission. By analyzing 1899 genome sequences (>18% of confirmed cases) from 23-January-2020 to 26-January-2021, we reveal the effects of fluctuating control measures on the evolution and epidemiology of SARS-CoV-2 lineages in Hong Kong. Despite numerous importations, only three introductions were responsible for 90% of locally-acquired cases. Community outbreaks were caused by novel introductions rather than a resurgence of circulating strains. Thus, local outbreak prevention requires strong border control and community surveillance, especially during periods of less stringent social restriction. Non-adherence to prolonged preventative measures may explain sustained local transmission observed during wave four in late 2020 and early 2021. We also found that, due to a tight transmission bottleneck, transmission of low-frequency single nucleotide variants between hosts is rare.

Fig. 1. Epidemiological summary and time-scaled phylogeny of SARS-CoV-2 in Hong Kong.

Confirmed cases (above) and sequenced genomes (below) are shown as bar charts across the four pandemic waves. Control-measure stringency applied in Hong Kong is based on the Oxford COVID-19 Government Response Tracker. Red shaded bars delineate five levels of control-measure stringency in Hong Kong (Level 1: <40; level 2 : 40-50; level 3: 50-60; level 4: 60-70; level 5: >70). Time-scaled phylogeny of representative genomes from Hong Kong (n = 610) and overseas regions (n = 1,538) shows monophyletic clades containing at least five community cases in Hong Kong. The two largest Hong Kong lineages during HK-wave3 and HK-wave4A, B.1.1.63 and B.1.36.27, were subsampled to 100 and 65 sequences, respectively. Other PANGO lineages detected during HK-wave3 and HK-wave4A are shown in Supplementary Table 4.

Fig. 2. Descriptive and temporal dynamics of SARS-CoV-2 lineages in Hong Kong.

a Time to most recent common ancestor (tMRCA) among the five earliest circulating local lineages of SARS-CoV-2 during waves 1 and 2 in Hong Kong. b Number of SARS-CoV-2 genomic samples per lineage identified over time using a maximum clade credibility phylogeny. Lineage size is ordered on a log10 scale and plotted by earliest confirmation date. c Correlation between the detection lag of locally circulating lineages and the final lineage duration with overlapping points showing uncertainty in lineage detection and duration. Detection lag over time as a function of tMRCA across three epidemic periods d waves one and two, e wave three, f wave four. Overall, a significant reduction in detection lag was observed over time and across each epidemic wave. Points in panels c–f represent a random sample of 1000 lineages from a Bayesian posterior tree distribution (n = 8000).

Fig. 3. Phylodynamics of waves three and four in Hong Kong.

Evolutionary relationships and effective reproduction number (R_e(t)) of HK-wave3 (B.1.1.63) and HK-wave4A (B.1.36.27) estimated using tree heights and sequenced incidence data. Node shapes indicate posterior probability >0.5. Histogram shows the number of genomes by collection date. Control-measure stringency applied in Hong Kong is based on the Oxford COVID-19 Government Response Tracker. Black line shows the instantaneous effective reproduction number (R_t), estimated based on infection dates inferred from reported symptom onset or confirmation dates for asymptomatic cases.

Fig. 4. Transmission bottleneck size estimates and single nucleotide variant (SNV) frequencies.

a Estimated transmission bottleneck sizes (maximum-likelihood estimates with 95% confidence intervals) for paired donor and recipient samples. Lag time defined by difference in dates of symptom onset. Bottleneck size estimates for transmission pairs fam_562 and fam_730 are not available due to a limited number of intra-host SNVs (iSNVs) in the recipients’ samples. b Jaccard distance of consensus-level SNVs and iSNVs among epidemiologically and phylogenetically different types of paired samples. Jaccard distances range from 0 to 1, with 0 indicating identical SNV profiles, and 1 indicating no SNVs in common. Violin plots show the range and distribution of Jaccard distances. Boxplots indicate median and inter-quartile ranges (IQR), and whiskers represent value ranges up to 1.5 * IQR. Between-group differences were tested by two-sided Wilcoxon tests separately for consensus-level SNVs and iSNVs. Significance was represented by **p < 0.05 (p = 0.006 between Same cluster and Identified transmission pairs for consensus-level SNVs; p = 0.034 between Identified transmission pairs and Same patient for consensus-level SNVs; p = 0.014 between Identified transmission pairs and Same patient for iSNVs) and ***p < 0.001.