Rapid Identification Of Antibiotic-Resistant Infections: Interview with Jong Lee, CEO At Day Zero Di

Antibiotic resistance is a growing concern, with some predictions suggesting that routine surgery could be unacceptably risky in a future where many antibiotics have become obsolete. Part of the problem lies in the time it takes clinicians to diagnose an antibiotic-resistant infection. Current techniques involve lab technicians culturing a bacterial sample until it can be analyzed for drug resistance. The whole process takes days, and by the time the test provides an answer, a patient could be dead from sepsis.

To address this, clinicians typically administer broad-spectrum antibiotics in an effort to bring an infection under control. However, this does not always work if the bacteria present are resistant, and can result in further drug resistance. Administering targeted treatments requires rapid diagnosis.

These issues have inspired Day Zero Diagnostics, a company based in Boston, MA, to develop a suite of technologies to enable rapid identification of drug-resistant bugs. These include sample preparation technology for bacterial DNA-enrichment from blood samples, sequencing technology, and proprietary machine learning algorithms to identify the presence of drug resistant bacteria.

Medgadget had the opportunity to talk to Jong Lee, CEO at Day Zero Diagnostics, about the company’s technology.

​

Conn Hastings, Medgadget: Please give us an overview of antibiotic resistance and the threat that it poses.

Jong Lee, Day Zero Diagnostics: Antibiotic resistance is an escalating public health crisis, where the spread of drug resistance is outpacing the development of new antibiotics. Antibiotic resistance causes at least 700,000 global deaths annually. By 2050, this figure is expected to grow to 10 million—more than deaths due to cancer—if current trends continue. Infections due to antibiotic-resistant bacteria are more dangerous for patients and more difficult and expensive to treat. They often require extended hospital stays, additional follow-up doctor visits, more costly and toxic alternative treatments, and have an increased risk of mortality.

Additionally, the rising prevalence of antibiotic-resistant organisms has dramatically increased the risks of healthcare-associated infections (HAIs), which already affect 4-5% of hospitalized patients in the U.S. and result in 99,000 patient deaths per year. As drug resistant bacteria become more and more common, so will deaths from life-threatening infections like sepsis, which is the body’s unregulated immune response to a severe bloodstream infection. Recent studies show that 20% of deaths globally are already associated with sepsis. In the United States, sepsis is estimated to be responsible for one third of hospital deaths.

Today, preventing septic shock relies on treating patients with broad-spectrum antibiotics because the risk of death from sepsis increases by almost 8% for each hour an infection goes without appropriate treatment. This practice can make subsequent culture-based diagnostics less sensitive and can also contribute to the growth of antibiotic-resistant organisms.

Medgadget: How are antibiotic-resistant infections currently diagnosed and what problems do such approaches cause?

Jong Lee: Early and effective treatment of antibiotic resistant infections has been proven to save lives. However, current technologies cannot provide actionable information quickly enough to enable early, targeted antibiotic treatment.

Today, when a patient goes to the hospital with signs of an infection, samples are collected and sent to the lab, where they are cultured on various media to see if they grow. Only after the bacteria has been given time to grow (usually 1-2 days) can they be identified and tested for drug resistance. This outdated process can take 2-5 days to return a result when it works, but culture can also fail to grow in 30-50% of severe infection cases. In the meantime, every hour that a patient remains inadequately treated increases their risk of complications or death.

With little to no information to go on, physicians are forced to treat with empiric therapy: powerful, broad-spectrum antibiotics that are expensive, can have significant toxicity, and are increasingly less effective due to the spread of multidrug resistant pathogens.

Medgadget: Do you think that rapid diagnosis will be a key factor in fighting antibiotic-resistant infections and reducing the manifestation of further drug resistance?

Jong Lee: The world is in desperate need of a rapid diagnostics that can perform comprehensive antibiotic resistance profiling without the need for time-consuming cultures. Broad spectrum antibiotics are critical tools for fighting life-threatening bacterial infections and the best way to ensure they continue to be effective when needed is to avoid their overuse as a first line treatment. But without rapid diagnostics that can provide a comprehensive antibiotic resistance profile straight from a clinical sample like blood, it is nearly impossible for physicians to choose targeted antibiotic therapy. Instead, they must “carpet bomb” the infection using broad spectrum antibiotics that encourage the development of even more resistance.

Medgadget: Please give us an overview of the technology developed by Day Zero Diagnostics. What does it do and how does it work?

Jong Lee: Day Zero is pioneering a new class of infectious disease diagnostics using whole genome sequencing and machine learning.

We aim to change the way infectious diseases are diagnosed and treated by developing a sequencing-based rapid diagnostic that identifies, within hours, both the species and the antibiotic resistance profile of a bacterial pathogen. Our diagnostic technologies are focused on three major areas that work in tandem with commercially available whole genome sequencing technology: sample preparation, machine learning algorithms, and big data.

Day Zero’s system is designed to extract bacterial DNA directly from a patient sample, such as blood, for whole genome sequencing with unprecedented sensitivity. Day Zero’s proprietary machine-learning algorithm then analyzes the genomic data to rapidly identify the pathogen and determine its antibiotic susceptibility and resistance profile, allowing physicians to confidently and quickly prescribe the most effective antibiotic. Unlike current molecular diagnostic approaches that test just a handful of species and provide little to no information on antibiotic resistance, the new system would allow for the simultaneous testing of a broad range of bacterial species and their antibiotic susceptibility, along with the ability to identify novel resistance determinants as they emerge.

DZD’s system is poised to help patients with severe infections receive the most effective antibiotic treatment on the first day they are admitted to the hospital—day zero—rather than being treated with multiple days of toxic broad-spectrum antibiotics. The system is intended to help physicians quickly and accurately diagnose and treat life-threatening superbug infections.

Medgadget: Please give us an overview of the results of the recent trial of the technology at the Boston Medical Center and the Blood2Bac validation study.

Jong Lee: Interim results from the Boston Medical Center Rapid Bacterial Identification Trial (BRABIT) and results from a separate validation study of Blood2Bac were recently presented at IDWeek 2020. Both studies demonstrate the promise of our new technology to quickly and accurately diagnose superbug infections. They suggest that our approach could improve the way physicians diagnose and treat life-threatening bacterial infections and most importantly, save lives.

The interim analysis of BRABIT included samples from the first cohort of patients enrolled in the study and evaluated our rapid, culture-free sequencing-based bacterial identification assay in patients with suspected bloodstream infections (BSIs). The assay is powered by Blood2Bac™, a proprietary sample preparation technology for ultra-high enrichment of bacterial DNA from clinical blood samples and Keynome®, our machine learning algorithm for species identification and antibiotic resistance profiling.

The results showed that our culture-free assay for determining the presence of a bacterial infection and its species were concordant with clinical blood cultures in 96% of samples. They also suggest the assay is potentially more sensitive than culture-based diagnostics, with the ability to confirm an infection even when same-day blood cultures failed to grow. In addition, there have been no false-positive results to date. False positives occur when a diagnostic detects an infection that is not truly present. This type of error is often made by other sequencing and molecular diagnostics and makes relying on them for clinical decision making particularly challenging.

The results of a separate validation study of Blood2Bac evaluated the technology’s ability to recover the whole genomes of pathogens and tested it across 50 bacterial species spiked into whole blood at concentrations as low as 1 CFU/mL (colony-forming unit per milliliter).

The results demonstrated that Blood2Bac, when paired with Keynome, was able to achieve an average of 95% whole-genome coverage and correctly identify all 50 species with 100% accuracy even at single-digit bacterial concentrations. The ability to test for a broad range of infections and recover almost the entirety of the pathogen’s genome directly from whole blood represents a significant advancement over current molecular diagnostic approaches that are limited to testing just a handful of species and only recover a very small portion of the pathogen’s genetic code.

Medgadget: When do you anticipate that such technology will be available and widely adopted?

Jong Lee: We already offer services to hospitals today that make use of certain elements of our technology. For example, our epiXact service for investigating suspected hospital-acquired infection outbreaks using whole genome sequencing is already available commercially. We are exploring making other elements of our technology available for clinical use in the near future. But designing, building and testing our fully integrated in vitro diagnostic will be the focus of our company for the next few years now that we have completed development of our core technologies. We will ultimately plan to pursue FDA clearance to introduce the system in the United States, as well as a CE Mark.

We will continue to collaborate with leading institutions to test our diagnostic approach in clinical studies over the next several years. Those studies will allow us to demonstrate how our diagnostic could impact clinical decision making and educate clinicians about our potential performance so the use case is well understood by the time our system is introduced.

In addition, we anticipate doing some pilots with leading academic medical centers on the use of sequencing to provide a big data view of the flow of pathogens through the microbiology lab. These pilots are designed to start articulating a vision for the future of hospital microbiology that we think will be transformative.


Source