Pathogens
Norovirus
This ~7kb RNA virus causes a large proportion of viral gastro-enteritis worldwide and leads to seasonal outbreaks of vomiting and diarrhoea in hospitals. In closed communities such as cruise ships, large numbers of cases close in space and time are likely to be true outbreaks – however, in hospitals it is less clear whether such persisting clusters are caused by ongoing transmission (from a "super-spreader", environmental contamination or failure to contain successive transmissions within the hospital) or, given ward movements and visitors, whether they arise from repeated re-introductions of new strains from the wider community (a series of independent, limited outbreaks). The difference is critical, since the respective most appropriate control strategies differ: eg, closure and extensive decontamination of infected wards versus local decontamination and simple cohorting of cases.
Local population-based sampling and increased resolution from sequencing the whole of the viral genome will increase the ability to distinguish ongoing point source transmission from re-introductions. Whole genome sequencing has the potential to improve outcomes for individual patients admitted to hospital, by determining whether a persistent Norovirus outbreak is actually occurring (and therefore more stringent isolation procedures are required) and whether multiple outbreaks are in fact caused by the same strain (in which case merging cohorts could free beds for other patients and prevent specialist wards such as stroke units being closed to new admissions).
C. difficile infection has emerged as a major public health threat in multiple countries, with evidence of epidemic spread of some strains. However, the relative contribution of person-to-person versus environmental contamination to transmission is poorly understood. Most knowledge about sequence diversity comes from restriction fragment length comparisons, the most established being ribotyping and pulse field gel electrophoresis. An existing MLST scheme, that was critically limited in its typing resolution by its choice of relatively non-variable and difficult-to sequence target genes, has recently been modified in Oxford. This novel extended scheme hosted on pubmlst, currently achieves resolution at least equal to ribotyping and can be refined to achieve greater resolution.
Whole genome sequencing will aid in diversity discovery to identify suitably informative genome fragments for improving discrimination. Samples will be whole genome sequenced from an ongoing collections of C. difficile stool samples and isolates from hospitals, to identify transmission events. This has the potential to provide more rapid and discriminatory typing and to achieve individualised infection control strategies for patients (by identifying whether C. difficile infection is likely due to recent acquisition or activation of a carriage strain) and targeted hospital-wide approaches to C. difficile control (through mapping of transmission routes and identifying key points for direct interventions, eg nursing or residential homes, antibiotic usage).
The carriage of S. aureus varies widely - about 30% of people are known to remain free of carriage, whilst the remainder continuously or intermittently carry S. aureus, but only rarely develop disease. Furthermore, S. aureus is associated with a wide range of disease manifestations ranging from skin and soft tissue to severe rapidly fatal invasive disease eg infective endocarditis.
Despite extensive study of S. aureus, there are no current means of consistently differentiating and determining transmission within the common MLST clones including MRSA, among UK hospitals precisely because almost all new infection events are with one of two clones. Rapid typing using well established techniques with higher resolution than MLST alone (eg MLST combined with Spa) may be able to detect some transmission routes of dominant clones. However, particularly for this pathogen, whole genome sequencing of representative strains for diversity discovery is needed to identify additional targets for sequence typing to provide sufficient resolution for any clinical utility.
The strains for investigation will be collected retrospectively and prospectively from hospitals. Samples will be taken from patients admitted to general medicine/surgery, renal and ITU on admission and discharge and any S. aureus from screening or clinical samples in these areas typed and whole genome sequenced. This research has the potential to improve hospital management; by understanding more precisely the routes of transmission within hospitals, eg environmental versus healthcare contact, and therefore providing potential "breakpoints" at which mini-outbreaks could be halted.
Mycobacterium tuberculosis
Whole genome sequencing of M. tuberculosis isolates can provide inference of genome-wide relationships, such as detection of outbreaks, national routes of transmission and likely countries of origin. Since many of the point mutations associated with anti-mycobacterial drug resistance have already been determined, then ‘resisto-typing’ of isolates should also be feasible, speeding up susceptibility testing and optimising drug treatment decisions.
Genome sequencing has the potential to change management of individual tuberculosis cases (both through earlier and more certain identification of donors for new acquisitions and substantially reduced time to knowledge of susceptibility patterns); but also the public health approach to tuberculosis control (through improved knowledge of transmission networks and contact tracing). Effective translation will be verified by (i) comparison of turn-around-time and drugs prescribed based on resisto-typing versus traditional susceptibility testing methods and (ii) comparison of clinician's beliefs about the tuberculosis source and diagnosis before/after sequence comparison.
