Time for a change of pace. We usually talk about hospital acquired infections in our Lab Services Blog, but a recent blog article from Eric Topol asking why whole genome sequencing isn’t used for infectious disease diagnostics got us talking. Sequencing-based diagnostics have become almost standard of care in areas like prenatal testing and oncology in recent years, but have yet to make a dent in hospital infectious disease testing. For example, did you know there is currently no sequencing-based diagnostic for infectious disease that is cleared by the FDA?
Why has the move from traditional culture-based diagnostic methods to sequencing-based diagnostics been so slow? The article points at the lack of technical maturity, cost feasibility, and faulty financial incentives. All true. But, we at DZD believe the next 3 to 5 years will be a transformative period for sequencing-based diagnostics, and that FDA grade clinical applications will start to enter the micro lab within this window of time. Here’s why:
Rapid sequencing technologies have just hit clinically viable accuracy.
Illumina revolutionized the world by introducing high accuracy, high throughput sequencing technology. However, the problem with severe infections is a problem of time. If you are going to affect treatment, the diagnostic has to work on the scale of minutes to hours, not days. Unfortunately, traditional sequencing requires batch processing of many samples to be cost-effective, and overnight runs to get results. For most severe infections, the results take too long and the costs are too high to serve as an alternative to culture.
Rapid sequencing technologies like those from Oxford Nanopore Technologies (ONT) have always been appealing because of their speed, yet suffer from lower data yield and accuracy – but the pace of improvement has been remarkable. Over the last 5 years, raw read accuracy for the ONT platform has gone from below 85% to over 99%. At DZD, we are now seeing enough data with just an hour or two of sequencing to enable downstream analysis with accuracy rates high enough to make clinical grade calls.
Sequencing costs for single sample testing continue to fall dramatically.
You’ve all seen the charts, sequencing costs have fallen dramatically over the last decade. But those charts gloss over the fact that the majority of cost reduction comes from being able to multiplex increasingly more samples on ever larger machines. What if you just want to sequence one sample and do it quickly?
The cost for single sample sequencing is still prohibitively expensive. With some of the latest advances coming from the sequencing world, and process optimizations of the kind we are working on, we anticipate that microbial sequencing costs for single samples will drop by 90% over the next 3 to 5 years, enabling cost competitive sequencing for one patient sample at a time.
The road ahead…
Sequencing capabilities and costs aren’t the only barriers to moving away from culture-based diagnostics. To replace culture, any solution also has to think about how to isolate and retrieve low-abundance pathogen signals from complex samples, the analysis pipeline for making reliable organism ID calls, and how to profile antibiotic susceptibility without culture. These are being worked on by us and others.
However, sequencing is reaching a new level of technical maturity which suggests that we are in an exciting time for infectious disease diagnostics. The next 5 years will be different than the past 5 years and we can’t wait to see the impact it will have on our field. If you would like to hear more about our thinking on the evolution of this space, please reach out and let’s get talking.
Jong Lee
CEO & Cofounder
Day Zero Diagnostics
You can read more about DZD, our HAI surveillance services, case studies, and WGS & machine learning implementation into infectious disease diagnostics here.