Undocumented Migrants Reintroducing COVID-19, Yunnan Province, China

To limit the spread of severe acute respiratory syndrome coronavirus 2, the government of China has been monitoring infected travelers and minimizing cold-chain contamination. However, other factors might contribute to recurring outbreaks. We analyze the role of undocumented migrants as potential transmitters of severe acute respiratory syndrome coronavirus 2 in China.

(COVID-19), the illness caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), rely on rigorous quarantine measures. These measures contributed to a decline in COVID-19 cases; no new locally acquired cases were reported in China on March 18, 2020 (http://www.nhc.gov.cn/xcs/yqtb/ list_gzbd_10.shtml). As a result, the focus of epidemic control and prevention work has shifted from local to imported cases of COVID-19. Although viral spread has been contained by mandates minimizing travel and cold-chain contamination (1), recurring COV-ID-19 outbreaks might be caused by other factors and pathways. On September 14, 2020, the discovery of 2 SARS-CoV-2-infected undocumented migrants from Myanmar prompted large-scale testing of >280,000 persons in Ruili, Yunnan Province, China (Figure).
On March 31, 2020, the Yunnan Provincial Leading Group for COVID-19 Epidemic Response published Notice No. 15 (http://www.yn.gov.cn/ztgg/ yqfk/zcfk/202004/t20200401_201604.html), which outlined strict measures to prevent COVID-19 importation from land and water ports. This notice discouraged citizens of adjacent countries from entering Yunnan Province; if entry was required, then those citizens should enter Yunnan Province via 1 of 19 official land ports (Appendix Table 1, https://wwwnc. cdc.gov/EID/article/27/5/20-4944-App1.pdf). Before continuing their travels, these persons needed to undergo 14 days of quarantine and test negative for SARS-CoV-2 by nucleic acid amplification test. On September 3, 2020, two undocumented migrants with no history of rejection at any official port crossed the Ruili River from Nankan (Myanmar) to Ruili. Patient 1, who had lost her senses of smell and taste for 1 week before diagnosis, received a COV-ID-19 diagnosis on September 12, 2020, ending a 139day period in which no new cases had been reported in Yunnan Province. In Myanmar, she had not had contact with any known COVID-19 patient. Officials identified 201 close contacts of patient 1 in Yunnan Province; these contacts were then tested for SARS-CoV-2. All contacts tested negative except the person who had entered China with patient 1. Because of the 9-day delay between entry and diagnosis, whether community transmission had occurred was unknown.
To evaluate potential spread, we undertook a large-scale SARS-CoV-2 screening campaign of >280,000 citizens and legal migrants in Ruili during September 15-19, 2020. We did not detect any cases of community transmission, possibly because of patient 1's low viral load; she had a mild case of COVID-19, with normal lung physiology and an N-gene cycle threshold of 37.41, suggesting a low level of infectiousness (2) (Appendix Figure 1). Patient 1 also wore a mask in public, potentially hindering COVID-19 transmission. In addition, patient 2 did not spend much time in public, further reducing potential for transmission. Although this event did not cause community spread of SARS-CoV-2 in Ruili, it highlights the need to curb undocumented immigration to prevent recurring outbreaks of COVID-19. This need is especially relevant in Yunnan Province, which shares a 4,060-km border with Myanmar, Laos, and Vietnam. The border spans 8 cities and 25 counties of China.
Although the SARS-CoV-2-infected migrants did not cause a COVID-19 outbreak, the event illustrates a transmission pathway distinct from air travel and cold-chain food transmission (1). The International Health Regulations and World Health Organization encourage open borders and suggest that COVID-19 control measures be applied only in limited circumstances (10). In 2020, official land ports in Yunnan Province did not close for the COVID-19 pandemic. Because of the long international border, epidemic control remains challenging in this province. Governments should control illegal immigration to avoid future reintroductions of COVID-19. Regional guidelines for COVID-19 control and prevention should strengthen surveillance of undocumented movement across borders, especially from neighboring countries with high rates of infection. Ruili People's Hospital by a negative pressure ambulance for treatment and isolation.

SARS-CoV-2 RNA Detection
We extracted total nucleic acids using an automatic preparation system (NP968, Tianlong, http://www.medtl.com). We analyzed the nucleic acids by real-time quantitative RT-PCR according to the protocol recommended by China National Health Commission (Appendix Figure 1) (1).

Sequencing Data Analysis
We sequenced >4.5 Gb; >90% had a base quality score >Q30, indicating a <0.1% error rate. We trimmed the raw sequencing reads using CLC Genomics Workbench version 12 software (QIAGEN) with quality scores >0.05 and ambiguous nucleotides <2. We trimmed the 5′ sequencing adaptor by removing 5 terminal nucleotides and the 3′ by removing 18 terminal nucleotides. Then, we mapped the high-quality reads to the Wuhan-Hu-1 reference (GenBank accession no. MN908947.3) using the following mapping options: match score = 1; mismatch cost = 2 (linear gap cost); length fraction = 0.5; similarity fraction = 0.8; autodetect paired distance; nonspecific match handling: map randomly. We obtained the consensus sequences for downstream analysis.

Evolutionary and Mutational Analysis
We downloaded 110,596 high-quality reference sequences from 6 continents from the GISAID database (https://www.gisaid.org) on September 15, 2020. To ensure that analyses were robust, we randomly extracted 1 sequence per month (collection date) from each country.
Including the Wuhan-Hu-1 reference and 2 Ruili sequences, a total of 592 sequences were aligned by using MAFFT version 7 (2) using the default options (Appendix Table 2). We obtained the mutation sites using the robust alignments.