Video Transcription
Karim Aly 00:00
Good morning, everyone. I put this up here because people are always curious what it looks like, so
Karim Aly 00:06
I thought I'd just get ahead of the curve on that one. Thank you for joining me. Let me start by framing our discussion today. So on the surface, most people think about disease detection. They don't see anything wrong with it, right? Because we sometimes get sick, we go, we get diagnosed, we get treated most of the time. We get better, and we forget all about the whole disease detection landscape. But the truth is, if we take an active look at it, and we start to peel away the layers, we start to see some really important fractures, problems that exist in the system. We have tests that are available that are just underutilized. We have tests that are heavily utilized, so outrageously and unsustainably expensive to the system. And worst of all, we have tests that are affordable and available but just inaccessible to those who need them the most. So you start to quickly realize that the system is lacking and something needs to be done. But the solution here is not to tear the system down and rebuild it. That would just be impractical. So rather, we think what we should do is augment it in a very meaningful way, and we believe the path to get there is smell. So smell actually isn't a new form of a diagnostics tool. This man right here, Hippocrates, who we named the Hippocratic Oath after, was using smell as a diagnostics tool from the very early days. In fact, one of the most common diseases we know, diabetes mellitus or mellitus, depending on where you come from, owes its name to the sense of smell. Mellitus is Latin for honey, and it's named that because physicians used to use their sense of smell to smell or even taste sweet urine to diagnose it. The problem with the sense of smell is it wasn't very scalable, because our sense of smell varies amongst individuals, which means if you can't standardize it, you can't scale it. But as it turns out, urine is not the best modality for smell-based health information. It's actually the breath, because in every breath that we exhale, a sliver of it is made up of VOCs that are exchanged between the circulatory system and the respiratory system and exhaled, and the right combination of those VOCs in the right concentrations are reliably indicative of a biomarker fingerprint for a disease. Now the problem is we could never interpret it. So every time we're breathing out, we're breathing out a real-time health snapshot, but we couldn't interpret it. So 10 years ago, we started off with a very clear North Star. Let's digitize the sense of smell. And what we did is we got inspired by the goat of product design, the greatest of all time, Mother Nature. We said, "How does Mother Nature do it?" And what we decided to do is try to emulate it as closely as possible in the digital world. It took us seven years and a boost from NASA, but we did it, and now we use it to decode breath and detect disease, just non-invasively and painlessly and in real time. So the device allows for a real-time breath test that's non-invasive, that's quick to administer, quick to produce results, is very cost-effective. We're talking about $1 per unit, and very portable, enabling it to be taken to the patient. So we can actually, for the first time, take the diagnostics to the patient, whether that's in a primary care center or a clinic, all the way to a pharmacy, to a school, or even a rural community. This is the device. This is what it looks like. It's fully integrated. In this device, you have the ability to automatically capture the alveolar portion of the breath, which is VOC-rich. You have the ability to eliminate humidity without losing any of the VOCs that we're looking for. You protect the ambient environment from contamination, and you process results in real time. If needed, it can also operate offline and still produce real-time results because the models can run on the edge, which makes it really useful for rural environments and, you know, lower or no connectivity environments. That's the hardware piece. The software piece is the AI, and it plays a really critical role here because that AI is what eliminates the noise, eliminates the confounders, and really hones in on the diagnostic biomarkers that we're looking for. The key competitive advantage we have in AI is we have a bunch of proprietary methodologies that allow us to build models very quickly with very small data sets, which means we do it fast and we do it cheap, and we can deploy them over the air to every device in the network in real time. Bringing it all together, we have a full stack mode, and we own that entire mode. It's proprietary and patented, which gives us full control over the supply chain, but it also gives us full control over the commercialization strategy. Right now, the mouthpiece is the only consumable, but we can actually make any component in this device a consumable to drive recurring or repeatable revenue in the future. And in terms of operating cost, like I said, we're talking about $1 per administered test, so we can be competitive regardless of whether you're talking about a gold standard of an MRI or a blood test. We envision a diagnostics platform that covers eight diagnostic areas. These five areas that are being called out are areas where we're currently doing work, either live clinical studies or research work in our labs. Ultimately, our vision is to create a breath panel that is able to detect or screen for multiple diseases in real time from a single breath and deliver results before you walk out of the room. That's the dream, and right now, we're running four clinical studies, and we have one that's been completed in cholesterol. So let's talk about results in cholesterol. The results were impressive. We delivered results that were on par with a blood test, but no blood draw and instant results. What you see here is the visualization of the breath samples. Green represents people with high LDL, people with high cholesterol. Gray represents these same people after they were put on six weeks of statin treatment. So you can even visually see a very distinct separation, a very distinct fingerprint between the two clinical states. But breast cancer is just as impressive, if not more impressive, because we were able to detect breast cancer with the same precision as the gold standard, a mammogram. But despite being able to rival a mammogram's performance, trying to displace a mammogram is probably not the most astute strategy. So we started looking at what are the large population groups that are underserved by the current standard of care, and it wasn't hard to find women with high breast tissue density. For those of you who are not familiar with this medical challenge, women with high breast tissue density are very poor candidates for mammograms because the sensitivity is very low. The tissue and the malignancy both appear as white in the image, so it's very hard to differentiate. So what you end up getting is a mammogram result that's about 35% sensitivity, which is worse than a coin toss. That's abysmal. It's shameful. But what will surprise you more is how big this population is. Approximately one in two women that are breast cancer screening eligible suffer from high breast tissue density, making them poor candidates for mammograms. So that's about 40 million women in North America alone. So how well do we perform? What is our diagnostic performance for this population group? Well, more than twice as good as the gold standard, right? And again, instant results. Non-invasive, no radiation, painless, and in this case, this is massive clinical utility, and we're not competing; we're not trying to displace the mammogram. We're actually adjunctive to it. So in a sense, the relationship between us and the mammogram is symbiotic, and this is the foundation of our go-to-market strategy. We are on track for a Q1 2027 breast cancer screening test authorization from the FDA across three clinical intended uses, the first and the primary one being adjunctive to mammography for women with high breast tissue density. Now it's hard to imagine what the scale of a multi-disease screening breath panel is, but if you zoom in on breast cancer, and then you zoom in even further on a jump to screening for women with high breast tissue density, you're already talking about a billion-dollar market size. So the commercial opportunity is very, very compelling. But while we focus on breast cancer, we're hedging that risk by partnering with the Gates Foundation to build a suite of infectious disease screening tests for low and middle-income countries, starting with tuberculosis. The World Health Organization, who, incidentally, is funded by the MGF, has called for a non-sputum, fully portable, low-cost test to drive their global screening programs. And we were selected by the Gates Foundation to turn that into a reality, and we're targeting a test in Q1 2028. So in closing, we believe our technology, our solution, has the potential to revolutionize health care for every stakeholder: providers, payers, systems, and of course, patients and their loved ones. This is not a zero-sum game, and this is what makes the vision so compelling to us. And in pursuit of that vision, we're currently raising $25 million. We have 50% of that already circled, 5 million of which is coming from the Foundation Strategic Investment Fund. So the foundation handpicks a small group of companies to invest in every year from the thousands of grantees that they have, and we were privileged to be one of those. The round proceeds will be used to take us through regulatory authorization for both breast cancer and tuberculosis, as well as expand our exploratory clinical programs. So that's what I've got today. But if anyone is interested to learn more, please come find me, and if you'd like to discuss how you potentially can get involved in our journey, I'd welcome the chance to talk. Thank you.
