Multiple summer events, including large indoor gatherings, in Provincetown, Massachusetts (MA), in July 2021 contributed to an outbreak of over one thousand COVID-19 cases among residents and visitors. Most cases were fully vaccinated, many of whom were also symptomatic, prompting a comprehensive public health response, motivating changes to national masking recommendations, and raising questions about infection and transmission among vaccinated individuals.
In order to characterize the outbreak and the viral population underlying it, the study team combined genomic and epidemiological data from 467 individuals, including 40% of known outbreak-associated cases.
The Delta variant accounted for 99% of sequenced outbreak-associated cases.
Phylogenetic analysis suggests over 40 sources of Delta in the dataset, with one sub-variant AY.25 responsible for a single cluster containing 83% of outbreak-associated genomes.
This cluster was likely not the result of extensive spread at a single site, but rather transmission from a common source across multiple settings over a short time.
Together, genomic epidemiology provides a high-resolution picture of the Provincetown outbreak, revealing multiple cases of transmission of Delta from fully vaccinated individuals.
The study findings were published on a preprint server and are currently being peer reviewed.
https://www.medrxiv.org/content/10.1101/2021.10.20.21265137v1
Genomic analysis of SARS-CoV-2 viruses from the outbreak
The Provincetown outbreak was driven by the Delta lineage: 99% (462/467) of genomes were Delta, with the remainder being Gamma (Figure 2A).
Of the Delta genomes, 84% (394/467) were from the lineage AY.25, a lineage circulating throughout the United States but rarely observed outside North America.
The outbreak occurred during a broader rise of the Delta lineage in MA: first detected on March 28th, its prevalence increased sharply in the weeks preceding the July 4th weekend, from 16% on June 15th to 77% at the start of our sampling period on July 9th, and reached 97% by the end of our sampling period on August 3rd (Figure S3).
We found no evidence of genetic differences of known functional consequence between outbreak-associated and other publicly available Delta genomes, or between outbreak-associated genomes from vaccinated and unvaccinated individuals. The Delta genomes in this dataset did not have novel consensus-level variants in the spike protein, nor did they have an increased frequency of any amino acid change of known or suspected functional impact (Figure S4) 5,6.
Vaccinated individuals also did not have significantly different numbers of intrahost single-nucleotide variants (iSNVs) compared to unvaccinated individuals (p=0.721), suggesting little difference in viral genetic diversity with vaccination (Figure S5A and S5B).
DISCUSSION
The outbreak of SARS-CoV-2 in Provincetown during and after the July 4th weekend was the first large outbreak of the Delta variant in a highly vaccinated population in the US. The robust public health response permitted extensive epidemiological and genomic characterization of the outbreak, the structure of transmission within it, and the role of vaccinated individuals, and offers generalizable insights for containing future outbreaks of Delta and other highly transmissible lineages of SARS-CoV-2.
The Provincetown outbreak raised public health concern and attracted international attention primarily due to the prevalence of symptomatic breakthrough infections and the potential occurrence of transmission from vaccinated individuals. Consistent with other recent reports13,14, breakthrough infections with Delta, while often symptomatic and with moderate to high viral loads, were typically mild.
Confidently assigning transmission links between individuals was unusually challenging: conventional contact tracing was difficult because of the many locations with dense potential contacts involved, while genomic inference of transmission was hindered by the low overall genetic diversity and large fraction of identical genomes.
Nonetheless, using genomic data to prioritize plausible connections between samples followed by more detailed epidemiological investigation identified several likely instances of transmission between fully vaccinated individuals and may serve as a model for future investigations of large outbreaks.
The size of the Provincetown outbreak – over one thousand cases – and its rapid early growth demonstrate that in densely crowded events and indoor conditions the SARS-CoV-2 Delta variant can cause a large outbreak even in a mostly vaccinated population. However, the Provincetown outbreak did not contribute substantially to the increase in Delta cases in Massachusetts.
The high rates of vaccination and the swift public health response1, which included deployment of mobile testing, a local indoor masking mandate, and an extensive outreach campaign, likely contributed to the short duration and restricted impact of the outbreak.
Additionally, the active engagement of the affected community in the epidemiological response, possibly influenced by historical public health outreach in the gay community, may have helped mitigate the impact of the outbreak15. The rapid decline and limited impact of the outbreak suggest that while Delta-driven outbreaks are not eliminated by high vaccination rates, they can be controlled with well-understood public health measures.
