The first molecular-typing based cluster analyses at the European level has shown “very high performance,” according to the European Centre for Disease Prevention and Control.
The eighth external quality assessment (EQA-8) scheme for the typing of Salmonella was organized for public health national reference laboratories (PH NRLs) in the center’s Food and Waterborne Diseases and Zoonoses network (FWD-Net). The objectives were to assess quality and comparability of molecular typing data produced by the labs.
The European Centre for Disease Prevention and Control (ECDC) told Food Safety News that almost half of participating labs could already perform cluster analyses using data from whole genome sequencing (WGS). The EQA also provided information about the comparison of results using different molecular methods.
Salmonellosis was the second-most commonly reported zoonotic disease in the EU in 2016. From 2012 to 2016, the annual number of cases was in the range of 92,012 to 94,597. However in 2013 87,453 cases were reported.
Surveillance data are reported by member states to the European Surveillance System (TESSy). Data from pulsed-field gel electrophoresis (PFGE) and multiple-locus variable number of tandem repeat analysis (MLVA) are collected in TESSy.
EQA-8 contained two new features: assessing performance of MLVA for Salmonella Enteritidis and ability to identify a cluster based on molecular typing by PFGE, MLVA and/or whole genome sequencing-derived data.
However, changes were only accepted by member states to “some degree” with an overall decrease in participants and only 13 of 24 labs performing cluster analysis using either PFGE, MLVA and/or WGS derived data. The number of participants may decrease further when the gel quality and analysis (PFGE part) is removed.
An ECDC spokesman said PH-NRLs can use existing typing methods for their outbreak investigations on a national level and PFGE is still a valid method for comparing human and food isolates in outbreak situations.
“ECDC and EFSA have established a joint-database based on PFGE and MLVA for investigation purposes in multi-country outbreaks. The transition from the use of traditional molecular typing methods to sequencing methods is ongoing and has progressed rapidly. WGS is prioritized by EU member states for both outbreak investigations and surveillance,” he said.
“The upgrade of the joint typing database in the coming years with WGS data is likely to cease the reporting of PFGE profiles. We expect that the number of member states using WGS will increase rapidly in the coming years and we want to support this evolution in our EQAs.”
Laboratories could participate fully or one part only (PFGE quality, PFGE analysis, MLVA Salmonella Typhimurium, MLVA Salmonella Enteritidis and/or molecular typing-based cluster analysis based on PFGE, MLVA and/or WGS-derived data).
A PFGE gel of sufficiently high quality was produced by most labs, 14 out of 17. Almost all, 10 out of 11, made the gel analysis in accordance with guidelines for producing inter-laboratory comparable gels. Three labs scored “poor” in one or three parameters having generated a gel of insufficient quality for inter-laboratory comparisons.
“In order to perform good national surveillance as well as submit profiles to EU-wide surveillance, it is important to have the capacity to properly analyse and interpret PFGE profiles at the national level and compare human and food isolates,” according to the report.
Among six labs scoring “poor” in the gel-producing part of the previous EQA, three achieved “poor” for some of the same parameters in this EQA despite recommendations for improvements. Labs were encouraged to resubmit a new PFGE gel, but none responded that they would.
“All participants get troubleshooting during the EQA when necessary. After sending their results they received an individual evaluation report with comments and suggestions for improving their performance,” said the ECDC spokesman.
For both MLVA schemes, overall performance was high, with nine out of 12 labs reporting correct allelic profiles for all test isolates in each set. In the new MLVA part for Salmonella Enteritidis, two participants had low scores, as only 40 percent and 70 percent of the test isolates were reported with correct allelic profiles.
Just over half do molecular typing-based cluster analysis
Out of the 24 labs participating, 13 performed molecular typing-based cluster analysis. The expected cluster contained seven isolates based on PFGE, five by MLVA and four if identification was based on WGS-derived data.
Four labs used PFGE for cluster analysis and two of them reported a cluster analysis based on WGS data. Only one lab did not identify the correct cluster using PFGE. Four labs used MLVA for cluster analysis and all also reported a cluster analysis based on WGS data. All labs identified the correct cluster using MLVA.
In a Salmonella Enteritidis outbreak from 2016-2018 that was linked to eggs, MLVA profiles were used in the definition of probable cases but out of 24 EQA participants only four performed cluster analysis using MLVA-derived data.
Performance was high using WGS-derived data for cluster analysis, with 10 out of 11 labs correctly identifying the cluster of four closely related isolates.
The majority of participants preferred a whole core genome multilocus sequence typing (allele-based) method, with 7/11 using cgMLST and only 4/11 using single nucleotide polymorphism (SNP-based) analysis as the main method for cluster analysis.
ECDC officials say there are difficulties in inter-laboratory comparability between member states with regard to surveillance and outbreak investigation when different methods are used.
“Despite the increasing use of WGS as a typing tool for large outbreaks, some labs still use PFGE for their primary surveillance and outbreak investigation, which complicates communication regarding outbreaks,” according to the ECDC.
“For inter-laboratory comparability and communication about cluster definitions, cgMLST using a standard scheme gives a very high degree of homogeneity in the results, whereas the use of non-standardised SNP analysis may be more challenging for comparison and communication between labs. This issue is further complicated because many labs still use other methods (PFGE or MLVA) and will probably not switch to WGS in the near future.”
ECDC also said SNP is an alternative method for cgMLST to assess the signals of multi-country salmonellosis outbreaks but comparability requires further studies and experience from outbreak investigations.
(To sign up for a free subscription to Food Safety News, click here.)