Good afternoon, everyone. My name is Matt Sykes, Life Science Tools and Diagnostics Analyst at Goldman Sachs. I have the pleasure of welcoming Katherine Stueland and Kevin Feeley, CEO and President and CFO of GeneDx. Katherine, Kevin, thank you.
Thank you so much for having us.
Thank you. Really appreciate you guys being here. Maybe if we just kind of start off, for the benefit of everybody, just give us an overview of the business today and kind of touch on some of the strategic initiatives that you guys are focused on.
Certainly. GeneDx was actually formed 25 years ago at the National Institutes of Health. The team really became renowned for being able to diagnose the hardest to diagnose patients. Over the past four years, we've really been focused on driving utilization of our exome testing and our genome testing. What we've been able to amass over the past decade or so is the largest rare disease data asset that exists in the United States. We've diagnosed more children with rare diseases than any other lab. It really comes down to the underlying data asset that is informing every single patient that we're able to analyze. That data asset is comprised of over 800,000 exomes and genomes, all enriched for rare disease, and is complemented by a phenotypic data asset of more than 6.1 million phenotypic data points.
It is the combination of those two data assets together that sets us apart in terms of our ability to drive a more definitive diagnosis for more patients. We have been focused on solving what today is still a five-year diagnostic odyssey, which we can solve within a matter of weeks, if not hours now. As we think about this year and beyond, we are driving utilization in the pediatric outpatient setting, and we are starting to drive utilization more meaningfully in the NICU setting as well.
Got it. Maybe just as a follow-up to that, and we actually still get this question a lot from investors, what makes GeneDx test so differentiated? Kind of what has created your market-leading position up to this point?
There are really two things that have set us apart, and then I'll add a third for our future. Our data asset, as I mentioned, is comprised of more than 800,000 exomes and genomes, all enriched for rare disease. 60% of the time, we're running mom and dad as well. We started investing in Medicaid populations before there has been Medicaid coverage. We have a database that is highly representative of the population, which is really unique in the genomic space. As I mentioned, the more than 6 million phenotypic data points are all of the clinic notes that we get that give us insights in terms of symptoms, family history, and other relevant medical information. It is the combination of that that informs our expert geneticists on whether or not they can upgrade or downgrade a variant of unknown significance.
They want to get to about a 90% confidence interval before they're upgrading or downgrading, which then ensures that our test is even more conclusive and more accurate for the next patient that we're actually sequencing and interpreting. It's a virtuous cycle of product improvement based on data assets. That has been at the core what sets us apart. It's the reason why an exome is not an exome, a genome is not a genome. There are other exomes and genomes on the market, but without that underlying data asset, it's a less conclusive and therefore less reliable answer that clinicians are getting. That's why 8 out of 10 geneticists rely on us for an exome or a genome. The robustness of our data asset is a really important reason why you can't just buy a sequencer and start running exomes and genomes.
You'd have to actually recreate the breadth and depth of that unique data asset. The second piece that sets us apart is the scale that we've built. In addition to having the most accurate tests out there, we also can run testing faster and more cost-effectively, which means that if it is in the NICU, we were able to make a significant product improvement where now we can run a test and interpret a whole genome for a baby in the NICU in 48 hours. It is that scale that we've been able to build that also contributes to our strength. That scale also contributes to our ability to lower our cost of goods, which then enables higher gross margins, which means we can really start investing back in even more scale. Scale is the second pillar.
I would say the third part, as we think about the future, are commercial might and muscle. We have built what we believe is a best-in-class commercial team with our sales team complemented by medical science liaisons. Just last week, we launched a brand campaign to really start raising awareness of GeneDx and the benefit that we are providing to more and more families to put an end to that diagnostic odyssey.
Got it. I mean, one of the things that I was so impressed with a few years ago is when you have a number of different potential end markets to go into. You have expert geneticists, ped neuro, NICU, general pediatricians. I think you realized very early on that you're only getting paid by some of those cohorts, and therefore you shifted your commercial attention to where you're getting paid, which sounds very simple, was actually contrary to what a lot of your peers were doing, which got them into trouble. Now that you've actually established some pretty nice market share and penetration in these markets, I'd love to kind of get a mark-to-market on sort of what your penetration and market share is in expert geneticists, where I think it's the highest. Ped neuro, NICU is new, but then general pediatricians.
Maybe just help everyone understand where you are in each of those markets and I guess how those S- curves kind of stack up so that you can keep driving growth momentum as you move through those markets.
You're absolutely right. The starting point for us and where we historically have enjoyed the support of clinicians has been in expert geneticists. If you're a geneticist, you typically know who GeneDx is. Most other clinicians have not known who GeneDx was. We have about 80% market share in those expert geneticists, and we want to continue to make sure that we have a predominant market share because they're an important ambassador for other clinician types moving forward. Part of the problem is there's about a 9-18 month wait time to get to see a geneticist. If you think about that five-year diagnostic odyssey, what's typically happening is a family is going in to see a general pediatrician. They get referred out to a specialist. The specialist may run some tests, and then they would historically be referred out to a geneticist.
That 9-18 month wait time contributes to that five-year diagnostic odyssey by driving utilization earlier on into that specialist setting. The pediatric neurologists were only about 14% penetrated into the patient population that a ped neurologist is seeing. We still have a lot of room to continue to grow in that segment, diagnosing epilepsy, developmental delay, autism, and now we've started moving into cerebral palsy as well. We'll continue to find additional symptoms and use cases and indications in that pediatric neurology setting because we're only about 14% penetrated. As we think about the general pediatrician setting, our team's not spending time there. You're absolutely right.
That's an area where there is not sufficient reimbursement, and we want to continue to ensure that as we drive volumes, we can get paid for those services, not an unreasonable ask, and part of what has enabled us to reach profitability last year. We are committed to profitable growth, which is part of the reason why the general pediatrician has not been a focus for us, but strategically, we do want to get there. Once AAP guidelines, which are more than a decade out of date, and once we can feel confident that there is a clear path to reimbursement, we will move there in the future. As we think about the NICU setting, it's really interesting dynamic of babies in a NICU actually get a genetic test.
Yet we've been able to publish data that shows that 60% of babies in a level 4 NICU would benefit from that test. Of the 800 or so level 3 or level 4 NICUs, about 20% of them are customers today, yet they're only using, I would say, a smattering of testing. We want to drive broader utilization in those settings.
Got it. Could you help us sort of you size sort of the penetration? Can you size the NICU opportunity in general, and how quickly can you drive penetration in that setting? How should we think about pacing in this segment?
Yeah, so there's 800 level three and level four NICUs in the U.S. They're 800 approximately. We would only target level three and level four NICUs. The level one and level two is where you would typically end up with premature birth, low birth weight, or some physical abnormality. There's about 450,000 babies that go through those level three and level four NICUs, and we've always assumed about 50% of those babies would have a meaningful change of care with intervention of a rapid genome. We published a definitive study, peer-reviewed study in February called Seek First that was in partnership with Seattle Children's and the University of Washington, which laid out inclusion and exclusion criteria and demonstrated that 60% of level four babies had a meaningful change of care.
That would say there's roughly a quarter million, 225,000 tests a year that should be run in the NICU. Today, less than 5% of all babies in the NICU are getting any genetic test. If you look at GeneDx over- 20- years, we've developed entrenched relationships with every expert geneticist around the globe and many children's hospitals. About 20% of those NICUs are existing GeneDx clients, but have been reserving exome and genome just for their most complex cases. It's building on that Seek First data. It's implementing a more seamless ordering process through Epic Aura. It's recently launching a two-day ultra-rapid turnaround time for that whole genome. It's a whole new subset of health economic data that we've brought in the commercial team that tells us we've now built the foundational pieces to begin to open up that NICU market.
It will be a core part of our growth strategy, not just the second half of this year, but into 2026 and into 2027, as ultimately we get to what we think is a quarter million children a year that would benefit from a rapid genome in the NICU.
Just given the I know you just recently launched it in April, but any sort of feedback on initial uptake for the ultra-rapid, the 48-hour turnaround time test?
Yeah, it's early. I'd say the ultra-rapid became orderable in April, but really is part of a suite of enhancements around the experience. That two-day turnaround time, but then also coupled with Epic Aura and other features that we launched. It's very early days, but through two months of experience, it's tracking to where we expect it to be. Ultimately, in the near term, see a place in the market for both the two-day and the five-day turnaround time products that we have. They're very different price points. We actually believe the clinical and economic argument is even more compelling for the two days, but it's going to take time for us to get in front of hospital administrators in the C-suite to bring them through that health economic data using their hospital's data. We've now equipped the commercial team to have that tool available.
More to come on where we think ultimately the mix will land, but do see a place in the market for both that five-day and two-day price point.
Got it. You're adding some additional indications this year in the specialist space. Cerebral palsy, you mentioned hearing loss and a couple of other indications. How would you characterize those opportunities going forward?
The first thing that we always look at when analyzing what additional patient populations to focus on, what is the best test from a clinical outcomes perspective? Part of the limitation with multi-gene panels, by default, they're only looking at a subset of genes. If you think about any number of these indications, cerebral palsy, there's 300 genes associated with it. Autism, there's 800 genes. Epilepsy, there's 768 genes. By utilizing a multi-gene panel, you're just leaving so many genes and therefore patients behind with inconclusive results. We first look at, is an exome or a genome the best test for clinical outcomes? Then we take a look at whether or not there is a path for reimbursement. That is both from a commercial and Medicaid perspective. I would say the additional factor that we contemplate is, are there clinical trials?
Are there new FDA-approved therapies? The actionability of the information is something else that we're taking into account. Those are the main factors that we think about when we think about additional indications. We have a robust pipeline of additional indications that we will be rolling out in time. The goal with more than 10,000 rare diseases that we know of today and with new gene disease discoveries, as we're opening up access, we're actually finding that the prevalence of these diseases is actually quite higher than what was anticipated. The real industrial wisdom of looking at a genome's worth of information and opening up access is helping us contribute to a broader understanding of the true prevalence of disease. We were never intending to be an epilepsy company.
We were intending to be able to and are intending to diagnose any genetic disease as early as possible. Ensuring, to your earlier point about opening up access to more and more patients and ensuring that we can get paid for those services, you're always going to hear us talk hand in hand about volume and about our ability to improve our average reimbursement rate. Investors have our commitment to that moving forward as we think about additional indications in the future.
It was just the second quarter of 2023 that for the first time GeneDx stepped out of calling on expert geneticists. There are about 2,000 of those in the country and into the first of what will be many additional call points, that being pediatric neurologists. There are about 2,000 of those. We touched on cerebral palsy. That is a new indication that will be focused on commercially at that same call point of ped neuro. That is about two years old now. A lot of excitement about bringing something new to that relationship with the ped neuro. At the same time, April for the first time selected the next of what will be additional call points. The pediatric immunologist, a new call point altogether in April. There are about 600 pediatric immunologists in the U.S.
I think it should be an indication that over time we will be stepping into additional call points, but this is really the first one in two years that we expect to be able to report back on later on this year and into 2026 on what the uptake is. A lot of excitement across our commercial team about having a new cohort of doctor altogether to speak to.
One of the things I've been thinking, one, the more and more people hear from you when I talk to you, you realize that rare disease is actually not that rare. Two, one of the positive things coming out of the new administration as it relates to Vinay Prasad at CBER and the FDA is that they seem to really want to accelerate rare disease drug development. You and I have talked in the past about many of these diseases you may find are not druggable. Therefore, is that actually a hindrance to doctors prescribing the test if there's nothing they can do about it? Maybe so, maybe not, but do you see that as a tailwind? If we get more accelerated approval of rare disease drugs, should that spur more testing, or is the connectivity between those two things fairly loose?
Absolutely. I think the frameworks that we're hearing, and I've worked on several FDA approvals of rare disease therapeutics, they echo what clinicians would say. They echo what parents would say, which is ensuring that you keep safety standards high and that there's flexibility on efficacy. There are a lot of really important and I think well-rounded conversations on this topic. Without a doubt, a focus on rare diseases can only be helpful to us in our quest to be able to diagnose more children as early as possible. Yes, it is helpful. I would say it's also really important, and I talk to parents of children with rare diseases often. Last week had two conversations. I do think we need to change the conversation about actionability because yes, there's druggability. FDA is helpful.
Clinical trials are helpful, but there's a lot of really important action that can be taken with a diagnosis. It can be physical therapy. It can be occupational therapy. It can be speech therapy. It can be sharing best practices as a child is transitioning from preschool into elementary school, parent-to-parent interactions that help support one another. As we continue to uncover exactly what the burden is on these families economically from a clinical perspective, obviously, but then also in terms of the psychosocial impact for these families, there's so much more that can be done. I have yet to hear of a parent say they wish they hadn't had the diagnosis. While devastating, it provides so much clarity, and it provides actionability in more ways than what most providers, I think, really fully appreciate.
Having the diagnosis also can lead to actionability in that it can stop the action of the prolonged odyssey and search for an answer, which is putting families in a mountain of medical debt, and it is exacerbating inefficient healthcare spends across the country. To Katherine's point, we need to change the definition of actionability and also realize that finding a diagnosis can put an end to using less efficient, effective technologies to search for an answer. That is exacerbating healthcare spend that is just unnecessary and that a genome-first approach really can aid towards healthcare cost efficiencies in this country.
Kevin, for you, you've driven exome and genome testing to 40% of your overall test mix. Are you still seeing the same rate of success in converting panels over to exome? Should we still be looking for that sort of 200-300 basis point improvement in mix every quarter? Is that the right way to think about it still?
Yeah, I think so. With 40% of all tests produced in the first quarter, nowhere near what we think is the ultimate end state, which is all hereditable disease diagnosis being on a whole exome, whole genome backbone. We will get there. Obviously, in that mixed share, we've taken some unilateral actions to turn off and reduce SKUs where we saw individual gene tests or multi-gene panels that are not serving physicians and patients well and not good for our business. At the same time, we'll expect to see a continued evolution of mixed share pickup by exome and genome, replacing what is a generation of legacy multi-gene panels that, frankly, should not be run anymore.
The pace of that change, I think, is less important than it had been in the past for our business because we've now retired about 70% of our test menu, going through a filtering process to say, what's the best test for patients and physicians and what are the unit economics for our business, such that every test left on our menu has passed through a filtering criteria. It is there for a reason. The reason is it's either favorable unit economics, so it's good business, or we see them as near-term candidates to one day convert to exome and genome, such that what's left on the menu, we're comfortable being on the menu. Therefore, that pace of pickup, it may ebb and flow from one quarter to the next, but ultimately see the entirety of our menu moving towards a whole genome.
Got it. And then looking to ASPs, last quarter, you mentioned you currently have a denial rate in the mid-40%. How much room left for improvement is there, and what strategies are you still using to work down no-pays?
I mean, ultimately, we think a theoretical max is something like a payment rate of 75-80% of the time. We're still facing a dynamic on all insurance-based claims being paid roughly half of the time. The gap there is still significant go-get for us to reduce denials, have that be accretive to average reimbursement rates. All of that improvement would effectively fall to the bottom line. It's easier said than done, but over the past two years, we've made radical improvements in our ability to reduce denials. Go back to when we first introduced pediatric neurologists as a call point, we were being denied about 65% of the time, and we've now been able to reduce that to less than 50% of the time. That has led to material improvements in reimbursement rates, gross margin, and operating profit for the company.
While we're very proud of that work to improve that payment and denial rate, it's nowhere near optimized if you think the theoretical max is something like 75-80% of the time. We'll get there really in two ways. If you look at the Medicaid portion of our business, that's 45% of all volume running through Medicaid plans. We service all 50 states. Up until this morning, there were only 33 states that covered exome and genome, but we're servicing all 50 states. Knowing we're going to take zeros, see those denials. Ultimately, do we think all 50 states cover the test? Absolutely. I started with the company eight years ago, and it was zero. A lot of progress has been made, but still a lot more to come. Excited to announce this morning, two new states picked up policy coverage for exome and genome.
The 33 outpatient will now become 35. Those were volumes that we were already servicing, and the only difference now being we'll get paid for those volumes. We'll continually see, I think, improvements in denials on the Medicaid front. What's super encouraging is in those 33 states with policy coverage, we are seeing something close to an 80% payment rate. Pretty clear rules to follow with respect to medical necessity and documentation requirements, and none of what you see in terms of ancillary procedural denials that you see in the commercial front. On the commercial payers, about 45% of all volume been able to reduce denials by taking a payer-by-payer approach. The technology not yet mature where there's a one-size-fits-all policy with respect to medical necessity, documentation, and procedural processes to follow.
Over the past two years, have been working to customize payer-by-payer to adhere to their rules. That work will continue. All to say, we think there's still massive room to reduce denials and improve average reimbursement.
On the cost per test, you've continued to bring those down. How much room is left on that front? If you combine that with ASP improvements that we were just discussing, kind of where does that lead you to a longer-term gross margin target?
Yeah, exome and genome today operating at around 80% gross margin. We talked, we believe there's room to improve average reimbursement rate. At the same time, over the past two years, we've made meaningful progress in reducing cost per test. The majority of the cost reductions over the trailing 24 months have been on what I'll call the wet side of the lab. Full-scale robotics, automation, process design, input cost reductions, the large manufacturers are now seeing a strong amount of competition. We're benefiting from that as well as the rest of the industry. Anything that makes sequencing costs and sequencing speed come down is a net favorable to our business. I think where we're more excited to further reduce cost per test, and it's very much in our ability to do so, is in what I'll call the dry side of the laboratory.
Still represents about a third of all cost. Production cost is in steps around variance analysis, interpretation, curating, and documenting medical records, and then writing the report itself. Today are likely done more efficiently than any other lab given our size and scale and the interpretation platform, but 100% of the samples we report out today are touched by humans, and they're fairly expensive, scarce humans at that. We'd like to get to a point where we can automate a large portion of those dry side steps. Certainly, is in our ability to do that. We believe firmly that there should always be a human portion of our diagnostic processes. We're very much committed to ensuring that. At the same time, there are many steps performed by that group of geneticists and genetic counselors on staff today that could be more fully automated.
I absolutely believe in the next couple of years, we'll be able to deliver on meaningful cost reductions on a per-test basis by driving language models, AI. The recent acquisition of Fabric will help us accelerate what was always a long-term roadmap to automate many of those steps. We were very impressed with their technology and ability to automate many of those dry side steps. Now we're going through the process of comparing our homegrown platform with Fabric's to come out with the best of both worlds in that regard.
Just on last question on OpEx for SG&A, what investments do you still need to make in the commercial teams as you kind of continue to grow volumes and expand your call points? I mean, you talked about additional call points. Is that additional salespeople, or can you flex people back and forth or have them do more than one call point?
The team's optimized. The new call point of immunologists by itself does not necessitate the need to expand the commercial team. I think the commercial team and spend really at a healthy spot right now, at a point to be highly leverageable. We're continually deploying technology that builds in embedded capacity for the sales team to spend more time hunting rather than servicing accounts. That will unlock capacity without having to add headcount. The one caveat there would be if and when the American Academy of Pediatrics, AAP, changes their guidelines. They last wrote their genetic testing guidelines in 2014, so they're 11 years old. When those guidelines change, it will have us target what is close to 60,000 pediatricians in the U.S. That would take an expansion of the sales team that you would notice.
I think we've got all the plans in place for if and when AAP ever changes their guidelines, but until then, feel like the SG&A spend, by and large, plus or minus a couple million dollars each quarter, is fairly well optimized and ready to be leveraged. Excited to announce we did invest in an overall brand campaign that just launched last week, really to drive brand awareness towards the problem we're trying to solve, which is the unmet need of kids around the country who are not getting fast enough access to the testing that we can provide in order to help them alleviate their symptoms and get the care they need. Not overly expensive, but an example of some incremental investments we're making into sales and marketing.
You should continue to see us drive revenue and see the percentage of SG&A decline over time as the revenue growth far outpaces the need to invest incremental dollars into SG&A.
I mean, Katherine, we've talked a lot about the pediatrics segment, where you guys are focused now, but just talk a little bit about the adult opportunity in terms of size, in terms of what's your timeline for really kind of focusing on that market?
Certainly. I mean, I think what is encouraging, we're seeing investments on the therapeutic side of things and companies that are focused on cardiogenetics, on therapeutics for adult neurodegenerative conditions like Parkinson's and Alzheimer's, which is there have been just tremendous failures in that space. An earlier genetic diagnosis is going to help in those settings as well. As we think about moving into that space, we do need to wait for reimbursement pathways to be there, and/or we need to be partnering with some of these companies that helps us. It's a bit of the chicken and the egg dilemma. If we can generate the data with pharma companies to be able to go to payers to make the case for why they should be covering testing, that would be super interesting to us.
Ultimately, GeneDx is not intended to be an epilepsy company, a pediatric company. We want to diagnose anyone as early as possible, including a future of being able to do whole genome screening at birth to be able to inform a diagnosis at that very earliest point.
In the time we have left, let's just talk a little bit more about the data business. You have done the right thing, in my opinion, by focusing on making sure you make money in diagnostics before you become a data company, but you do have this data. You mentioned the 800,000, and you have got a very robust data set. What is your strategy to try to kind of monetize or at least use that data for biopharma and make sure that that is actually a viable business for you?
You're right. The strategy today is to be a really good diagnostics business and to drive earlier diagnosis. The data that we amassed, and we were at 270,000 exomes and genomes just four years ago. The fact that we've accelerated that data moat, expanded that data moat is tremendous as a competitive advantage. It helps us with every interpretation. Being able to take a data asset, and again, all of our testing is clinically actionable, to be able to put it to work for ultimately patients to deliver on precision medicine and not just a diagnosis, but a gene therapy, to be able to inform gene editing, we'd love to be able to be an important contributor in that space as well.
You look at other models that exist, whether it's for Regeneron utilizing their genome center and all of the data there to inform their own drug discovery and development. No other company should have to develop their own genome center. That's what GeneDx ultimately is. We are going to continue to be an engine for earlier diagnosis, but also for the most robust data set that is clinically actionable in order to really inform an earlier target discovery and faster clinical trial development, ultimately to get to more options for patients. That is where we want to really put that data asset to work to be able to change the trajectory of those FDA approvals.
Great. Was that right on time? We are out of time. Thank you very much, Katherine and Kevin. I really appreciate it.
Thanks so much. Appreciate it.