Two hospitals in the UK recently demonstrated that outbreaks of methicillin-resistant Staphylococcus aureus can be monitored by next-generation sequencing on the Illumina MiSeq.
Groups from the University of Oxford's John Radcliffe Hospital and the University of Cambridge published separate studies in BMJ Open and the New England Journal of Medicine, respectively, demonstrating their use of the MiSeq to track MRSA outbreaks in hospitals.
Next-gen sequencing on the MiSeq platform enables clinicians to "redefine a hospital outbreak with far more confidence now than you could as recently as a year or 18 months ago in a time frame that is clinically useful," Derrick Crook, a professor of microbiology at the John Radcliffe Hospital, told Clinical Sequencing News.
Crook added that the hospital recently received a grant to conduct a pilot experiment to test the MiSeq alongside its routine methodology technology to "get a clearer view of its utility" in monitoring outbreaks, he said.
The hospital will continue to test the MiSeq over the next two to three years, not only for its ability to monitor outbreaks, but also to gauge the economic impact.
So far, he said, the technology has generated "hugely useful information in guiding our infection control team." Next-gen sequencing "will become part of the everyday practice of infectious control," he added. "It's not a matter of if, but when."
One of the main advantages of whole-genome sequencing over current typing methods such as pulsed-field gel electrophoresis or multilocus sequence typing is that next-gen has a much higher resolution. "It can be difficult to have sufficient resolution to tell the difference between members of a clone that are being brought into the hospital independently or a transmission event that's occurred in the hospital," Julian Parkhill, the head of the pathogen genomics team at the Sanger Institute and a senior author of the NEJM study, told CSN.
Current methods "lack the resolution that allows you to track particularly common strains in hospitals such that you could work out who acquired the germ from whom," Crook added.
Eventually, Crook said, hospitals will incorporate sequencing to monitor not only MRSA outbreaks, but also Clostridium difficile, norovirus, Escherichia coli, and even tuberculosis.
A number of other researchers have also been investigating the potential of rapid next-gen sequencing to monitor outbreaks. Last year, researchers in Germany demonstrated that Life Technologies' Ion Torrent platform could sequence an outbreak strain of E. coli, help piece together its evolutionary history, and identify genes that made it especially virulent (IS 6/7/2011).
Other research teams subsequently sequenced the strain on the Pacific Biosciences RS, the 454 GS Junior, and the MiSeq (IS 7/12/2011 and IS 7/5/2011).
In this month's BMJ Open study, the researchers compared standard methods for monitoring hospital infections with next-gen sequencing, demonstrating how sequencing could shed additional light on the origin of the infections and potentially change outbreak management.
The researchers sequenced isolates from a cluster of eight MRSA carriers and an associated bacteremia case in an intensive care unit; another MRSA cluster of six cases; and 15 C. difficile isolates from two separate clusters.
All 41 isolates were sequenced and analyzed on the MiSeq within five days of being cultured. From the first cluster of MRSA isolates, obtained from eight patients, whole-genome sequence data enabled the researchers to peg the cluster as an outbreak.
The isolates came from patients in the same intensive care unit over four months and were indistinguishable by pulsed-field gel electrophoresis, a commonly used technique to identify pathogen strains. Additionally, they were all members of a rare subtype. However, the strains differed in their drug sensitivities and no common source was identified, so it was unclear whether the cases were connected.
Sequencing showed that six of the cases contained no sequence differences, while one case differed at a single site, and the eighth differed by three single nucleotide variants. This information helped clarify that the cases were indeed part of an outbreak and led to "implementation (and escalation with subsequent cases) of intensive infection control supervision of the unit with visits up to four to five times per day," the authors wrote. Additionally, the hospital retrained its medical staff in administering intravenous medication and taking blood samples.
In the second cluster, MRSA isolates were identified over a three-month period, five from patients on the same ward with overlapping stays, and one from a patient that was a relative of one of the infected patients on the ward. The cases were all of the same subtype — a relatively rare type, but one that has been observed to occur sporadically — and three of the isolates were indistinguishable by pulsed-field gel electrophoresis.
However, "given the prolonged time scale, it was unclear whether these cases reflected a genuine outbreak or background circulation of related organisms," the authors wrote.
Sequencing clarified that the pathogen had indeed been transmitted between all six patients, identifying only one single nucleotide variant between the six isolates.
While sequencing confirmed that the MRSA clusters were outbreaks, sequencing of the C. difficile isolates provided evidence that suggested the infection had not been transmitted between patients.
The first cluster included three cases that had occurred over four days among inpatients in the same ward. While clusters are considered outbreaks only if the cases share a strain type, because that information can be slow to obtain by PCR or other methods, such clusters are treated as presumptive outbreaks.
In this example, sequencing showed that the three cases had different sequence types and that their genomes differed at over 4,000 sites. However, a fourth case that occurred in an adjacent room to one of the three cases was subsequently found to be indistinguishable from that case, suggesting that the pathogen was transmitted between these two patients.
In the second C. difficile cluster, three cases occurred over a three-week period in an elective surgical unit. These cases were suspected to have been transmitted between the patients, because the most recent previous case had occurred six months before. However, sequencing revealed that the cases were genomically diverse, ruling out transmission.
Separately, a group at the University of Cambridge, the Wellcome Trust Sanger Institute, and Addenbrookes Hospital also tested the MiSeq's ability to monitor a MRSA outbreak in a neonatal setting.
According to the Sanger Institute's Parkhill, the study, published in the NEJM, showed that "if you were to sequence in real time, you would identify an outbreak much earlier than current technologies, and you could also be much more accurate in your identification of an outbreak."
While the researchers sequenced samples retrospectively, had it been done in real time, "it would have changed how the outbreak was managed," Parkhill said.
Additionally, the researchers estimated that the materials cost of using whole-genome sequencing was $150 per sample, including sample preparation, library quality control, and sequencing — a cost that is "roughly equivalent to the cost of two PCR tests used to screen for MRSA carriage," the authors wrote.
In the study, the researchers examined a putative outbreak from 2009. An infant, who was the index case, had been transferred to the neonatal intensive care unit of the Rosie Hospital of the Cambridge University Hospitals system. After admission, the infant became infected with a strain of MRSA that was different from the strain normally present in the hospital. Seven other patients were subsequently infected and three staff members also tested positive for MRSA.
To further study the outbreak, researchers used the MiSeq to sequence the MRSA strain from an index patient and from six others involved in the outbreak, as well as isolates from seven infected individuals not associated with the outbreak. Illumina performed the sequencing because the MiSeq had not yet been released commercially.
Sequencing and phylogenetic analysis showed two distinct groups separated by 102 SNPs. One group included the strains from the patients in the neonatal intensive care unit, while the other group contained the reference genome and three isolates not associated with the outbreak.
The team also constructed a "resistome" for the isolates by searching for mutations in genes associated with antimicrobial resistance, identifying mutations in genes associated with resistance to ciprofloxacin and rifampin. Additionally, they put together a "toxome" from SNPs identified in toxin genes. In both cases, the team found that these profiles were concordant with the phenotypic data from each of the strains.
Parkhill said that the Sanger Institute plans to continue its surveillance of MRSA within the hospital and will also expand to a wider region within the UK.
Sequencing will ultimately be useful for not only detecting and monitoring outbreaks, but also for providing additional, valuable information about the pathogens themselves that current typing methods do not, he said. Sequencing can "fairly accurately predict drug sensitivity and resistance, and which toxin genes are present," he said. "So, switching to whole-genome sequencing brings all this additional information."