Containing an infectious viral disease depends heavily on the ability to identify cases of secondary infection linked to the index case. A new study published on the preprint server medRxiv* in October 2020 reports a method to assess transmission potential for any event in the current pandemic, using what is known about the virus's spread, data on the mobility of individuals, and new ways of processing genomic sequencing data. The outcome is to determine whether cases are linked or not so as to take action to prevent further spread.
This news article was a review of a preliminary scientific report that had not undergone peer-review at the time of publication. Since its initial publication, the scientific report has now been peer reviewed and accepted for publication in a Scientific Journal. Links to the preliminary and peer-reviewed reports are available in the Sources section at the bottom of this article. View Sources
Methods of Identifying Linked Cases
Viral sequences have often been used to identify clusters of linked cases, as may occur in healthcare facilities, for instance, differentiating them from those occurring in the community. This is based on the spontaneous accumulation of gene variants in different virus strains, helping to distinguish them.
Recent studies of this type have successfully linked hospital-based cases caused by identical strains, allowing epidemiological follow-up to control potential onward transmission. However, this approach's utility is outweighed by the fact that relatively few variants have appeared so far since this virus only appeared less than a year ago. Thus, finding identical strains in different individuals need not mean they were infected by a common source.
Secondly, the effect of time from symptom onset and viral transmission, and from infection to symptom onset in the next case in the transmission chain, has been described. However, asymptomatic infection continues to pose a huge challenge to effective containment efforts.
Again, information on the physical proximity of individuals susceptible to infection can help to predict or rule out transmission. Finally, errors in measuring the frequencies of genetic variants, and emerging mutations, must be accounted for.
Another option has been short-read viral sequencing, which reflects viral diversity within a given host and enables transmission events to be evaluated. However, nanopore-based technologies have made it easy to collect data inexpensively, compared to the former technology by Illumina, albeit the latter is much more precise.
Novel Rapid Method to Detect Linkage
The current study uses data from patients in two wards of Cambridge University Hospitals with different linkage profiles for COVID-19 cases. One was designated for COVID-19 patients (Ward Y), the other non-COVID (Ward X).
The researchers developed a tool that uses patient location data, infection dynamics, and genome sequences to examine gene variation by location. This has the advantage of speedy analysis and straightforward interpretation. This was used in the hospital-based sequencing data from the above wards
The method was used on 136 cases where more than one strain of the virus was found. There were ~340 samples altogether, with 2-9 samples per individual. The researchers first developed a consensus sequence and then used only those which had 90% or higher coverage of the genome.
The novel method incorporates the error rate in measuring gene variants, at an estimated 0.2 nucleotide errors per genome sequence. They found that ~6 days elapsed from symptom onset to transmission, within which period the variation in sequence due to evolution and to noise are comparable.
By accounting for noise, they found that when consecutive transmissions are farther apart, for rapidly evolving viruses, the virus in a single host is likely to change more rapidly than can be attributed to noise in the sequencing process. This may not be the case for rapid consecutive transmissions within days of each other, in which case noise is relevant to the final analysis.
Their model arrived at the likelihood that any two individuals transmitted the infection in either direction. This was compared with a predefined scale to show if the transmission was likely, borderline, or unlikely.
HCWs May Drive Linked Infections
Overall, the study showed two clusters of linked infection. On COVID-19 Ward Y, healthcare workers (HCWs) appear to be involved in two outbreaks, with other unrelated cases occurring among the patients as well. One HCW was potentially responsible for each cluster of infection, having developed symptoms earlier than others in the same cluster. This was supported by phylogenetic reconstruction as well.
Despite this, they found that these HCWs were not recorded as having been present on the ward Y simultaneously with any of the other cluster cases, which could mean unrecorded contacts occurred. This is much more likely with HCWs compared to patients.
In the other ward, this was absent, and the pattern of linkage less complex. In Y, cases were closer together than in X.
The researchers explain, "A coherent cluster of infection, potentially indicating a single introduction of the virus into the ward, was responsible for all of the observed cases [in X]. By contrast, the nature of ward Y as a designated COVID-19 ward led to multiple introductions of the virus onto the ward, perhaps two of those cases leading to ongoing nosocomial transmission."
The Importance of Location Data
They repeated the analysis without reference to whether the patients were present or absent on the ward at the point of infection and found that it impacted linkage detection accuracy between cases.
Implications and Future Directions
This method is chiefly designed for analyzing nosocomial transmission, given the need to distinguish the cases introduced from outside in a busy hospital from those spreading via the HCWs. Providing a rapid way to untangle the two facilitates further investigation and containment to reduce cases of new transmission by targeted intervention where it is most useful.
A deeper study of viral diversity within hosts is required to increase detected linkage accuracy between cases.
The researchers comment, "Our software may provide valuable insights, but does not replace the need for full epidemiological investigation."
Using phylogenetic analysis as a foundation, the method goes further to map potentially linked infections. Thus, it can help identify preliminary possibilities in nosocomial transmission to investigate an outbreak and allocate containment resources optimally.
This news article was a review of a preliminary scientific report that had not undergone peer-review at the time of publication. Since its initial publication, the scientific report has now been peer reviewed and accepted for publication in a Scientific Journal. Links to the preliminary and peer-reviewed reports are available in the Sources section at the bottom of this article. View Sources
Article Revisions
- May 18 2023 - The preprint preliminary research paper that this article was based upon was accepted for publication in a peer-reviewed Scientific Journal. This article was edited accordingly to include a link to the final peer-reviewed paper, now shown in the sources section.
SARS-CoV-2 hijacks host proteins to escape immune clearance