Molecular epidemiology of SARS-CoV-2 genome sentinel surveillance in commercial COVID-19 testing sites targeting asymptomatic individuals during Japan's seventh epidemic wave

Abstract

Eight peaks of coronavirus disease 2019 (COVID-19) outbreak occurred in Japan, each associated with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants of concern. The National Epidemiological Surveillance of Infectious Diseases (NESID) analyzed viral genome sequences from symptomatic patients and submitted the results to GISAID. Meanwhile, commercial testing services occasionally sequence samples from asymptomatic individuals. We compared a total of 1248 SARS-CoV-2 full-genome sequences obtained from the SB Coronavirus Inspection Center Corp. (SBCVIC) during Japan's seventh wave, which was dominated by Omicron variants, with 1764 sequences obtained in Japan from GISAID during the same period using chronological phylogenies and molecular transmission networks. The number of SBCVIC sequences was consistent with the number of cases reported by NESID. The SBCVIC detected a shift in the PANGO lineage from BA.2 to BA.5 earlier than that of GISAID. BA.2 lineages from the SBCVIC were distributed at different locations in the transmission network dominated by GISAID entries, whereas BA.5 lineages from SBCVIC and GISAID often formed distinct subclusters. Test-based sentinel surveillance of asymptomatic individuals may be a more manageable approach compared to notifiable disease surveillance; however, it may not necessarily capture all infection populations throughout Japan.

Keywords: Genome surveillance; SARS-CoV-2; Transmission network.

Fig. 1

The waves COVID-19 and the dynamic evolution of SARS-CoV-2 variants in Japan. (A) Trends in the number of COVID-19 cases reported in Japan from May 2020 to April 2023 relying on the NESID reports. (B) Trends in the frequency of VOC and PANGO linages in Japan from May 2020 to April 2023 relying on CoVariants.org (https://covariants.org) data. The prevalent Pango lineage for Omicron is shown, with each month of the study period represented by blue brackets between panel A and B. VOC, variants of concern; NESID, The National Epidemiological Surveillance of Infectious Diseases.

Fig. 2

Diverse genetic profiles of SARS-CoV-2 genome sequences collected using various routes across different months. The line graph depicts the average number of base substitutions per site, calculated from averaging over all sequence pairs within each sampling route per month group. Bars show the standard error estimates, which were estimated using the bootstrap method with 500 replications. Open and closed circles indicate diversity within the SB Coronavirus Inspection Center Corp. and GISAID samples from each month, respectively.

Fig. 3

Viral lineage composition at the SBCVIC and GISAID entries during the seventh wave, and the concurrent COVID-19 prevalence in Japan. (A) A stacked bar graph illustrates the frequency distribution of viral lineages at the SBCVIC (left) and GISAID entries (right), whereas a line graph represents the number of cases reported in NESID during the corresponding period. (B) The number of cases for each viral lineage at the SBCVIC (left) and GISAID entries (right) are shown as stacked bar graphs. Concurrently, a line graph indicates the number of cases reported in NESID during the same period. NESID, The National Epidemiological Surveillance of Infectious Diseases.

Fig. 4

Cluster analysis of mean genetic distance between groups based on collection route and month. The relationship between viral genome differences among the collection routes in each month is shown in Supplementary Table S2 and illustrated in a dendrogram using the Ward method.

Fig. 5

Transmission network of viruses collected during the seventh wave in Japan. (A) Estimated transmission links for cases belonging to BA.2 with genetic distances < 0.015% are shown. Each node in the network graph indicates a case infected with the BA-2 virus. Node colors in the graph denote collection routes and months. Blue and red indicate cases derived from the SBCVIC and GISAID, respectively. The darkness of the color reflects the recency of the collection month. (B) Estimated transmission links for cases belonging to BA.5 with genetic distances < 0.015% are shown. Each node in the network graph represents a case infected with the BA-5 virus. The color of the node is the same as described in (A). d indicates the graph density of each network.