Genpax Estimates Significant Cost Savings, Health Benefits From WGS Infection Surveillance

NEW YORK – Bioinformatics company Genpax estimates that healthcare systems can save considerable money while preventing infection outbreaks through microbial whole-genome sequencing.

In a study published earlier this month in Microbial Genetics, the London-based company calculated that implementing a WGS surveillance strategy could prevent some 74,000 infections annually in England, saving the National Health Service approximately £480 million ($604.5 million) per year, amounting to a return on investment of £7.83 per pound invested in diagnostic WGS. Similarly, Genpax's model estimated a net saving in the US of roughly $3.2 billion, or $18.74 for every dollar invested, while preventing around 169,000 infections.

"In the NHS, we're just always patching people up," said Susie Jerwood, Genpax's chief medical officer and formerly a medical consultant for the NHS. "Why can't we do a bit more stopping things [from] happening?"

Genpax sees proactive microbial surveillance as a strong way to both prevent infectious outbreaks from spreading far in the first place and to break the chain of transmission of ongoing outbreaks.

Conducting surveillance via WGS, says Jerwood, is the best way to do this because accessing the whole genome provides a clear and comprehensive view of the genetic connections between infections.

"If you're just looking for housekeeping genes, for example," she said, "you're looking at, say, seven regions, and you might come out and say, well, these [microbes] are all the same because these regions look the same, but you've only looked at 50 percent, 70 percent of the genome. Maybe less."

In contrast, depending on the species in question, WGS provides access to closer to 90 percent or more of the genome.

"By looking at so much of the genome," Jerwood said, "you can tell quite clearly if something is related or not because if you have [even] a [single nucleotide polymorphism] difference, then you have a difference, whereas if you're looking at a smaller part [of the genome], you might not see that difference."

Based on this guiding philosophy, Jerwood and her colleagues built an economic and health impact model to estimate both the impact of WGS-based surveillance directly on hospital resources such as consumables and antibiotics, and on the costs of allocating medical professionals and hospital resources before and after surveillance by WGS.

The model included estimated infections, colonization by bugs with antimicrobial resistance (AMR) genes, surveillance and detection of clusters by WGS, expected deaths, healthcare resources and costs, bed costs and stay lengths, and the cost of infection control and other relevant specialists.

In all calculations, the Genpax team aimed to keep estimates conservative, such as avoiding the kinds of high-cost estimates that might unnecessarily concern decision-makers and being conservative with respect to calculating lengths of stay, as some patients are predisposed to long stays regardless of acquiring an infection while in the hospital.

The resulting model suggested that proactive WGS surveillance in healthcare settings may significantly reduce infections, infection-related deaths, and all related expenses.

Alexander Sundermann, an infectious disease expert at the University of Pittsburgh who specializes in microbial epidemiology, said in an email that the study provides "another piece of evidence in the recent growing literature that genomic surveillance for [healthcare-associated infections] is cost-beneficial for healthcare but also beneficial for patient safety."

Sundermann has hands-on experience implementing microbial WGS surveillance, having developed one such program for use in the University of Pittsburgh Medical Center and its affiliated facilities. Last year, that program proved instrumental in enabling Sundermann and his colleagues to identify two drug-resistant Pseudomonas aeruginosa infections that traced to contaminated eyedrops.

Sundermann praised Genpax's study for basing its assumptions on the most conservative parameters drawn from the studies that the company referenced in building its model.

"Because of this," he said, "the paper gives me confidence that they could be accurate in an applied setting."

Nonetheless, following the adage that "all models are wrong, but some are useful," he said that "there needs to be some direct evidence that genomic surveillance in a healthcare setting has a direct reduction in HAIs."

Finding such evidence can be difficult, he explained, as it essentially involves demonstrating that a future outbreak that never occurred was due to a given intervention.

This is not, Sundermann pointed out, to suggest that such a demonstration is impossible or not worthwhile.

"Many studies that have performed genomic surveillance have uncovered outbreaks and transmission that would have never been detected without it," he said. "This alone, added with the multiple studies on economic value, should give pause to healthcare institutions who have not thought of using genomic surveillance for HAI prevention."

Genpax is, in fact, now preparing a pilot study aimed at backing up its model with solid empirical data. The company has installed an Illumina sequencer in a hospital in southern England, has completed some preliminary work, and currently awaits ethics board approval.

"We're putting our money where our mouth is," Jerwood said. "Rather than just saying that this a good theory … we are going to do our best to see whether actually using this [model] live, in real time, actually does do what we say it does."

Although Jerwood said that Genpax is currently at a pre-commercial stage, the company is ready to begin taking on clients. So far, she said, the company has mainly done some pro bono work for a few clients "to show people what we can do."

The business model is to analyze sequencing data generated by clients, charging on a per-analysis basis. At the moment, the firm's technology is optimized for Illumina sequence data, but it plans to begin accepting data from other platforms, such as Oxford Nanopore.

Oxford Nanopore had, in fact, been a platform that Genpax had initially considered for its analysis, but at the time, Jerwood said that that platform appeared less stable than Illumina's and that contracts with Oxford Nanopore included a clause that placed certain restrictions on who could analyze its data.

"I know it's got a much more stable platform, and they've taken that clause out," Jerwood said. "So they're definitely going to be on our list."

Genpax's key differentiating technology lies in its ability to efficiently compare large numbers of microbial isolates, normally a computationally expensive task.

While the precise details of how Genpax accomplishes this task amount to the company's "secret sauce," Nigel Saunders, the firm's CSO, attributed it to "a combination of underlying features linked to the near-zero-error high accuracy, and our novel reference-free interrogation solution."

This solution, he said, enables all strains to be analyzed, comparing them by SNP, rather than summary information, as is more typically used.

"Exactly how we do this is part of our trade secret IP," he said.

Jerwood noted that one key aspect in WGS surveillance that can be easier to overlook in reactive testing — when seeking to identify organisms after an outbreak has occurred — is that antibiotic-sensitive organisms can do as much damage as antibiotic-resistant ones until caught and treated.

"In the UK," she said, "sequencing is only used if you've got a bug you're worried about," such as in the case that a hospital discovers two resistant bacteria and wants to know if they're related.

"But if you have two patients in next-door beds and they both have the same fully sensitive E. coli, don't you want to know about that, too?"

Beyond the healthcare setting, Genpax aims to tackle other areas where microbial infection is a concern, such as in agriculture and food production.

"Food is a huge, huge area that we're really hoping to help," Jerwood said.

An outbreak of Listeria, for example, may not go away simply by treating affected cows or recalling certain batches of cheese, but by identifying the outbreak's exact origin.

"With the high resolution that we've got," Jerwood said, "you can actually look back and say where [the infection] came from rather than just saying yes, [the bugs] are related and probably had something to do with each other."