Severe Acute Respiratory Syndrome Coronavirus 2 Outbreak Related to a Nightclub, Germany, 2020

We report an outbreak of coronavirus disease with 74 cases related to a nightclub in Germany in March 2020. Staff members were particularly affected (attack rate 56%) and likely caused sustained viral transmission after an event at the club. This outbreak illustrates the potential for superspreader events and corroborates current club closures.

We conducted semistructured telephone interviews with first-generation cases to gather information related to their exposure in the club, prior travel history, and characteristics of clinical symptoms. Among all first-generation cases linked to the outbreak, contact information was available for 44 cases and the study team interviewed them. We performed analysis by time, generation, symptoms, sex, and age. For analysis by time, we stratified cases by guests, staff members, and generation. For continuous variables, if not normally distributed, we calculated medians and interquartile ranges (IQR). In addition, members of the outbreak investigation team performed a site visit of club X to gain insight into the outbreak setting (Appendix Figure 1).

SARS-CoV-2 Antibody Screening of Staff Members
For laboratory-confirmation of cases, qualitative real-time RT-PCR for SARS-CoV-2 was performed on purified RNA from swabs as described (2), or the Cobas SARS-CoV-2 test (Roche, https://www.roche.com), both of which target the SARS-CoV-2 E gene.
For nightclub staff members who had negative PCR tests for SARS-CoV-2 or who were not tested after the exposure, we performed SARS-CoV-2 antibody screening during June 2-24, 2020, approximately 3 months after the outbreak, by using a 2-step approach. First, we screened samples by using Anti-SARS-CoV-2 S1 IgG and IgA ELISAs (Euroimmun, https://www.euroimmun.com) according to the manufacturer's protocol. Second, we performed a plaque reduction neutralization test (PRNT), as previously described (3,4). In the PRNT, we tested all dilutions in duplicate. Only serum samples showing an optical density ratio >0.8 in the IgA or IgG ELISA were considered reactive and tested in the PRNT.

Whole-Genome Sequencing
To investigate the sequence diversity of the outbreak, we performed whole-genome sequencing (WGS) on available samples from initial diagnostic testing that had sufficient sample material. For WGS we followed two approaches. First, we performed direct sequencing of native samples with a high viral load (cycle threshold [Ct] value <25); then, for samples with lower SARS-CoV-2 concentration, we used a PCR amplicon-based sequencing approach.
For sequencing of native samples with a high viral load (Ct value <25), we used <100 ng in 5 µL of extracted RNA for library preparation by using the KAPA  For amplicon-based complete genome sequencing of samples with a lower viral load (Ct value >25) we followed 2 approaches. First, we used 108 SARS-CoV-2 whole genomes, available in early February 2020 to design 48 overlapping heminested PCR fragment primers.
Fragment size ranged between 507 bp and 950 bp for first-round products and 414-877 bp for second-round products. Primer names including "i" were modified versions (Appendix Table 2).
For the first-round PCR, a 25 μL reaction was performed by using the SuperScript III One-Step First-round RT-PCRs were carried out by using a thermocycling protocol with reverse transcription at 55°C for 20 min and subsequent PCR at 95°C for 3 min, followed by 45 cycles of 95°C for 15 s, 56°C for 15 s, and 72°C for 55 s, followed by a final 2-min extension step at 72°C. Second-round reactions used the same cycling protocol but without the RT step. Second, for amplicon-based WGS we used random hexamers and the SuperScript III Reverse transcription kit (Invitrogen) according to manufacturer's instructions, then amplified the SARS-CoV-2 genome by using the primer sets (V1) published by the Artic Network  Table 3).

Bioinformatic Sequence Analysis
For each sample, data from the individual sequencing runs were mapped to a reference sequence (GISAID accession no. EPI_ISL_402125, GenBank accession no. NC_045512.2) by using bowtie2 version 2.3.5.1 and the sensitive-local option (5). Duplicates were removed using GATK MarkDuplicatesSpark version 4.1.4.1 (6), and the consensus reads were called at positions with coverage >3 reads by using bcftools version 1.10.2-31-gffa7016 and bcftools call-ploidy 1-mv-Oz-o (https://samtools.github.io/bcftools/bcftools.html). We included the following sequences in the phylogenetic tree: 1 sequence from each clade assigned by Pangolin; all sequences from Germany sampled before April 16, 2020 and available in GISAID on July 22, 2020; and representative sequences from GISAID clade G sampled by April 15, 2020 and available in GISAID on July 22, 2020. Sequences from each country were clustered by using CD-HIT version 4.8.1 by using a sequence identity threshold of 0.99 (7) and we picked 1 sequence from each cluster. Then we included 4 sequences from the U.S. and 1 from Canada that have the same additional SNP as sequences ChVir-W1248-16 and ChVir-D715-D799-17 from this outbreak. The phylogenetic tree was inferred by using RAxML-ng version 0.7.0 BETA (8) and an HKY substitution model, with gamma distribution rate heterogeneity among sites and invariant sites. We performed 100 bootstrap replicates and created a phylogenic tree by using baltic (9) (Appendix Figure 2).

Ethics Approval
The outbreak investigation was conducted within the framework of the German Infection Protection Act (10) as part of an outbreak response and public health practice. Mandatory regulations were respected, and thus review by an ethics committee was not required. Support by the Robert Koch Institute was provided after official request. Participation in the questionnaire and blood specimen collection for antibody testing was voluntary, for which verbal consent was obtained. For antibody testing, additional written informed consent was obtained.

Description of the Outbreak Setting
Club X is located in a basement. The area accessible to guests is ≈150 m 2 with a height of ≈3 m (Appendix Figure 1). Ventilation of the space is ensured by a mechanical air exhaust and supply system and maintenance was performed according to the manufacturer's instructions. To avoid noise pollution in the surrounding neighborhood, windows are usually closed during events.

Clinical Symptoms of Cases
Among a total of 74 cases linked to the outbreak, dates of symptom onset were available for 64 cases. Of those, 44 cases could be interviewed on clinical symptoms during their COVID-19 infection. All 44 cases reported having >1 symptom. The most common symptoms experienced were dysgeusia (65%), cough (61%), headache (58%), and dysosmia (58%) (Appendix Table 4).