There has been a lot of excitement about the prospects of whole genome sequencing (WGS) to support infection prevention and control in a really meaningful way over the past decade or two. But to me this potential seems largely unfulfilled. WGS remains largely the domain of reference and research laboratories, and has not transitioned effectively to support IPC daily decision making. A recent review highlights the potential of WGS to support IPC, and identifies some of the barriers to be overcome if WGS really is to be a big part of our future in IPC.

Many years ago, before the flood, I did a PhD in the molecular epidemiology of S. aureus, including quite a bit of genotyping (you can even read my thesis if you suffer from insomnia). Whilst WGS was just about being done when I did my PhD, it cost a lot of money and so I used some of the older methods. Back then, it was clear that WGS would be a leap forward, and a sense that we’d look back at the “partial” genome sequencing were doing (e.g. spa typing or MLST) and laugh at their puny lack of resolution. Whilst that is true, and WGS offers far more resolution, the ‘older’ approaches have not be phased out as rapidly as I’d expected.

So what does WGS offer, at least in theory, to IPC?

• Transmission insights. Perhaps the most compelling benefit of WGS for IPC is the ability to either ‘rule in’ or ‘rule out’ transmission events. This would affect our decision making and allocation of resources (e.g. there’s no point in trying to prevent further cross-transmission when WGS has ruled out an initial cross-transmission event). WGS also has the potential to identify cross-transmission events that we wouldn’t otherwise have known about.
• Source insights. WGS can provide us with unprecedented ability to understand the role of various sources of transmission. For example, manufacturing plants as global source for Mycobacterium chimaera in water heater-cooler units, temperature probes for Candida auris, and staff for MRSA. WGS can also refute commonly suspected sources, suggesting that cross-transmission is less frequent than assumed for some organisms in some settings.  
• Diagnostics. WGS can be used to support diagnostic decisions.
• Surveillance of AMR and emerging pathogens. The interrogation of large and often national datasets can help us to understand transmission trends, for example clonal vs. plasmid spread of CPE.

So why, given all of these potential benefits, is WGS still ‘the future’?

• Cost and cost effectiveness. Whilst the cost of WGS has plummeted, it’s still not free. And if you are going to sequencing everything, the incremental costs builds up. There has been some work on the cost effectiveness of WGS, and it’s looking quite positive.
• Technology. Next-generation sequencing (taking a “patchwork and reassembly” rather than “linear” approach to sequencing) is a leap forward, but WGS can still advance further.
• Infrastructure. Whilst the incremental cost of sequencing a single genome isn’t great, the cost of establishing local infrastructure for WGS is substantial.
• Education and expertise to interpret. We need bioinformaticians and scientists with unique skillsets around to make sense of WGS data.
• Debate over interpretation. Even if we have the right human resources infrastructure available, there are still questions over interpretation of WGS (e.g. “how many SNPs am I ‘allowed’ when calling cross-transmission”). Related to this, lab standards for interpretation of WGS are a work in progress.
• Timeliness. One of the major barriers is that feedback of information from WGS is too slow to be useful in any operational IPC sense.

One approach that seems to be gathering pace is prospective WGS, whereby all significant organisms (or more commonly all organisms that meet certain criteria e.g. all CPE) are sequenced prospectively. This means that when a new one appears, all you need to do is sequence that one, match it with local and international databases, and then the benefits of WGS outlined about should begin to flow.

I firmly believe that WGS is a big part of the near future of IPC, and keen to see how things play out over the next couple of years. My crystal ball tells me that over the next 5 years most IPC teams will be making decisions informed by information from WGS on a regular basis. But we do need to break down some of the barriers above to bring WGS to the IPC frontline.