Video Transcription
Jong Lee 00:02
Steve, hi everyone here. I have a brief life sciences interlude in the world of med tech. So hopefully it'll be a good break for all of you. Day Zero Diagnostics is a company based out of Boston, Massachusetts. We're a spin-out from a group that was all affiliated with Harvard and MIT. We have a very focused mission, which is we believe we're on a mission to change the world of molecular diagnostics from what is currently a PCR-based paradigm to a next-generation sequencing-based paradigm that can sit in the microbiology lab of hospitals. The company has raised $49 million to date in equity, as well as $23 million in non-dilutive funding. Importantly, one of the sources of our equity funding is Becton Dickinson, which is one of the three largest strategics in this category. And we are currently raising a Series B that we believe will end up being anchored by another strategic that's not Becton Dickinson, giving us a nice setup for a potential exit within the next 24 months. To just put out the premise of what we're describing, if you've had an opportunity to invest in a PCR-based diagnostic, I'd ask you all to just think again. We're currently at what we consider to be the last phase of PCR diagnostics. In the next three years, there will be a clear pathway for PCR diagnostics to come to fruition. But if you invest in one now that's targeted for a five to 10 year life, you'll be entering a time period when next-generation sequencing will actually be a very competitive alternative that's more capable and faster and provides a greater ability to do diagnostics than what PCR diagnostics are capable of doing. And we believe that, because we track the data, next-generation sequencing, particularly Oxford Nanopore sequencing, has gone through a fairly large transformation over the last several years. You can see that the accuracy of the next-generation sequencing platform has radically improved from 2014 when it was introduced, to 2022 we believe it's now a clinical-grade accuracy. You can also see that the cost for next-generation sequencing has dropped dramatically. We track next-generation sequencing as going through a one-log reduction every five years in unit cost. So something that used to cost $100 a gig base of data to generate becomes $10.05 years later, becomes $1.05 years after that, becomes 10 cents five years after that. So we believe the next-generation sequencing will be the platform for molecular diagnostics that dominates the next 40 years, and we're about to hit a once-in-a-generation transformation of the underlying technical profile. In order for all of this to come to fruition, however, you need a bunch of enabling technologies, and that's what we've developed as Day Zero Diagnostics. So we have a system that we've developed that currently sits in our lab and runs clinical samples, where we have an ultra-high enrichment system that allows us to extract high-quality genomes directly from clinical samples without the need for culture, which is a stumbling block today for clinical diagnostics. We have a partnership with Oxford Nanopore Technologies to integrate sequencing directly onto our platform so that the microbiology lab is capable of doing next-generation sequencing without having to send it out to somebody else to run that diagnostic. And then we have a computational platform called Kinome, which has been developed to provide ID and, really importantly, antibiotic susceptibility testing directly from the genome in a way that breaks through the barriers of traditional genomic approaches. We believe we're the only company that's capable of direct-from-blood ID and AST. Our first indication, our lead program, is for bloodstream infections. We have a second program that's currently going with respiratory infections. Those two sources of infections are the two primary causes of sepsis, which is high mortality, super high cost. CMS believes that they end up paying $62 billion in sepsis care costs alone each year, and has a lot of motivation to see this kind of diagnostic come to market. And just to put up kind of an explanation for why sepsis is so intractable, why bloodstream infections are so intractable, it's because the real fight that you have when you're treating a severe infection is a fight against time more than anything else. We live in a world where you have antibiotic resistance that's growing. So a lot of therapeutics that are designed to treat infections don't work. How you select the diagnostic and how you select the therapy matters a great deal to the outcome. And unfortunately, the key piece of information that a clinician needs to make that choice is not available to them until two to five days after they start their diagnostic, and the primary reason for that is that you need culture in order to be able to run phenotypic AST tests. You basically can't get an AST result without culture today, and we've developed a solution where you can go direct from clinical sample, without the need for culture, and get accurate AST results. Because physicians don't have that information at the time of treatment, they use what's called empiric therapy. It's what everyone thinks of as the way to treat an infection: you try to get ever increasingly more powerful drugs to carpet bomb, really, whatever the infection is, because you don't know, and you don't know what the antibiotic resistance pattern is. So in the case of sepsis, patients are typically getting treated with a cocktail of three different hard-hitting antibiotics that have comorbidities associated and so forth. A lot of the time it's not enough, and some of the time it's too much. And both of those have consequences. I'll skip this. What we've developed is a method of doing what we call ultra-high enrichment. So we take a sample of blood, collect it in a Vacutainer, we run it through our process, and we get 10 logs of enrichment. What does 10 logs mean? You know, it means that we start with a sample that has a billion times more human DNA than bacterial DNA in that sample of blood, because for a bloodstream infection, a single cell per milliliter of pathogen is actually a clinically relevant load. So that's why you need culture today. But what we've developed is a method where you can go from a sample that has a billion times more human DNA than pathogen DNA; by the time we're done, and we're sequencing with our process, we typically generate about 1000 times more pathogen DNA than human DNA in that ratio. And what that means is that we work with the entire genome sequence when we do our analysis. If you looked at a PCR diagnostic or if you looked at cell-free DNA, you know, liquid biopsy approaches, they typically retrieve less than 3% of the genome of the pathogen. May be enough to do ID, but you experience a high rate of false positives, and you don't retrieve enough genomic information to actually call antibiotic susceptibility. Once we get through the enrichment process, our Kinome algorithms, Kem ID and Kem GST, which have been built for regulatory-grade diagnostic purposes, FDA clearance targets, provide a comprehensive bacterial and fungal ID with virtually no false positives, which is unheard of in the field of sequencing diagnostics, and then we provide a phenotypically accurate AST result. It's a lot of data shown here, but this is just to kind of demonstrate that when you work with whole genome sequences, you get extraordinarily good signal-to-noise ratio. You can really tell what's in the sample, which is all the gold dots, and how separate they are from things that are false positive calls, which are the blue dots. If you looked at the same data set from a cell-free DNA approach, you would find something that looks more like what's on the right, almost impossible to threshold in order to separate signal from noise. Importantly, we're really committed to the idea that antibiotic susceptibility results are a critical part of a diagnostic result. That ID is not enough. It's not comprehensive enough to drive therapy. And what traditional methods of doing genomic AST do is they look for the presence or absence of a resistance marker, of a known resistance marker that's been annotated, published, captured in a database. We believe that fundamentally, that is incapable of replicating phenotypic AST. And the reason for that is shown in this analysis we've done of 30,000 samples where in 19% of susceptible samples there's actually the presence of a known resistance marker, because not all resistance markers are deterministic. More importantly, in 21% of resistance samples, bugs that basically are resistant to the drugs that they're being treated with, there's actually no known resistance marker in those genomes, and so it's because not all mechanisms of resistance are known or understood. So our approach is a modern machine learning approach, where we've built the world's largest data set of AST results with pathogen genomes, and then we train our algorithm to learn what the associations are in the genome that result in predictions of AST. We find that our accuracy using our machine learning AI-based approach has about a 15 to 20 point advantage for satiating traditional genomic resistance marker detection approaches, and we believe that resistance marker approaches are incapable of becoming better than that. I'll skip through this quickly just to say the field of infectious disease diagnostics, from an investment standpoint, is a bit of a mess. Our thesis about that is that there's an awful lot of solutions that are me-too solutions that are incomplete solutions to the problem they're trying to solve, which have made them difficult to adopt and difficult to get an economic return on. We are trying to create a complete solution that's been recognized by people who are players in the space. Becton Dickinson, Danaher, all of these companies have recognized that the work that we're doing is significantly different than things that they've seen, and we think it sets us up for an exit where a strategic can be a leader in the next generation of molecular diagnostics that moves away from PCR and becomes a world of sequencing-based diagnostics. And a lot of clinicians will tell you the same thing, that this is an unmet need that they're looking for a solution in.
Jong Lee 09:57
So we're targeting the $8 billion category that we believe will be a growth category in this transition. We're starting with bloodstream infections. We have programs for respiratory infections as well as complex UTIs. We think this will become a central pathway through which clinical samples come into a microbiology lab that parallels the traditional culture-based pathway that we see today. Thanks very much. Applause.
