Good morning, and welcome to the GRAIL Capital Markets Day webcast. At this time, all participants are in listen-only mode. At the end of the call, there will be a question and answer session. If you have connected via Zoom and would like to ask a question, please use the Raise Hand feature at the bottom of your Zoom window. If you have dialed in by telephone, press star nine to raise your hand. Once called upon, please feel free to unmute and ask your question by unmuting the device microphone or dialing star six. I will repeat these instructions before we begin Q&A. I will now turn the call over to Bob Ragusa, Chief Executive Officer of GRAIL. Thank you.
Thank you, operator. Good morning, and welcome, everyone. It's a pleasure to be with you today for our Capital Markets Day. Next slide, please. Before we begin, I will remind you today's presentation will contain forward-looking statements and more information can be found in the SEC filings. Next slide, please. Joining me today are Dr. Joshua Ofman, President, Sir Harpal Kumar, President, Biopharma and Europe, and Aaron Freidin, Chief Financial Officer. GRAIL has built an exceptionally strong team across all levels of the organization. We are incredibly proud of the achievements that our employees have driven to date in data science, computational biology, medical affairs, laboratory operations, IT, commercial, as well as other areas. We are pleased to represent the company here today. Next slide, please. Today's agenda will begin with a short introduction to GRAIL and our investment highlights.
Josh will describe the promise of multi-cancer early detection, GRAIL's progress to date, and Galleri, our first-of-its-kind multi-cancer early detection screening test. We'll then move to our commercial strategy for Galleri. Harpal and Josh will discuss the scientific evidence that GRAIL has generated, plus ongoing clinical and real-world data collection to support adoption and upcoming regulatory submissions. Harpal will then share GRAIL's opportunities beyond asymptomatic screening, including symptomatic detection and a number of precision oncology opportunities, such as pretreatment, prognosis, and MRD. Finally, Aaron will present GRAIL's financial profile before we open the call for Q&A. Next slide, please. Moving now to the company's background and investment highlights. Next slide. Since its inception in 2016, GRAIL's mission has been to detect cancer early when it can be cured. Next slide.
We know that current recommended screenings, which test for single cancers, is limited and most deadly cancers are found too late. We believe that multi-cancer early detection is a solution for effective population screening. GRAIL is uniquely situated to address one of the most meaningful opportunities in healthcare. Galleri, our first of its kind multi-cancer early detection test, was designed for population scale screening. Our expansive clinical evidence program is setting the standard for the multi-cancer early detection field. GRAIL has a first-mover advantage in an expansive global market. We have completed more than 180,000 Galleri commercial tests and are expanding the commercial adoption of Galleri in the U.S. We have a large global opportunity. We're investing to enable commercial scale. And finally, we have a proprietary methylation platform that yields a product portfolio across the cancer care continuum. Next slide.
Since GRAIL was formed, we have developed an exceptionally innovative technology, drove market development of a new cancer screening category, and built out scale to support the transition to population scale. We are looking today at a number of critical near-term catalysts that drive the next stage of growth. We are looking forward to these upcoming catalysts, including a potential NHS pilot, final data from our pivotal studies, FDA approval, and broad reimbursement for Galleri, as well as additional product launches. Next slide, please. The ultimate opportunity for multi-cancer early detection is very significant. The total addressable market in the U.S. is over 100 million individuals. That TAM expands quickly, with 19 million in the U.K., where we are running the NHS Galleri trial.
Additionally, across the EU, there's a TAM of approximately 160 million more individuals, and in Japan, we see a TAM of 50 million individuals. Clearly, we're at the very beginning of an exciting journey. Next slide. In delivering our mission to detect cancer early, when it can be cured, GRAIL envisions population scale, multi-cancer early detection. Achieving such an ambitious goal requires significant investment and resources directed towards the broad reimbursement we anticipate in the future. To that end, we are laying the groundwork today to intercept that eventual opportunity. We have already made large investments to support broad access, such as establishing a laboratory footprint built for scale. We are making significant strides in market development. MCED represents a new, potentially transformational category.
Our commercial and medical affairs efforts today are designed to build the education and awareness of MCED, drive adoption of Galleri, and prepare the market in order to capitalize on the larger opportunities as they materialize. Next one. GRAIL was formed in 2016 as an idea, and since then has made tremendous progress. We've invested heavily in transformational science, enrolling what we believe is the largest clinical program in genomic medicine. We have presented analytical and clinical validation data in world-renowned forums, and published our findings in leading academic journals. We have successfully developed and launched the first clinically validated multi-cancer early detection test called Galleri. We've also begun to expand the application of our methylation platform capabilities to precision oncology. Next slide. We are proud of the rigor of our data.
We have generated more than 260 scientific and medical publications to date, presenting at major conferences and publishing in leading academic journals. Galleri has found strong commercial reception in a largely pre-reimbursement market. There's robust interest from providers and commercial partners. We have already completed over 180,000 Galleri commercial tests. We have signed over 100 commercial partnerships, and our progress here spans a number of channels, including health systems, employers, and life insurance. Importantly, we have over 10,000 ordering providers. We have also had high-profile recognition by Fast Company, The Atlantic, Time, and Fortune magazines for our high level of innovation and impact on the world. Next slide. In addition to the size of the investment required to develop a population-scale MCED test, the timescale is analogous to that of drug development.
Our clinical and real-world experience provide a time advantage in the order of years. Over the past several years, we have significantly de-risked our business. We have generated critical data in our Pathfinder study that demonstrate Galleri can detect twice as many cancers as standard of care screening alone. We have launched our commercial product, Galleri. We have delivered over 180,000 Galleri commercial tests. We have unlocked new product opportunities utilizing our methylation platform. We have scaled our laboratory infrastructure, and importantly, we have progressed discussions with the FDA on the evidence requirements for our PMA, primarily supported by Pathfinder 2, where we have enrolled over 30,000 participants, and the NHS-Galleri study that is fully enrolled and on track to finish final study visits in Q3 of this year.
Additionally, we have gained approval of the REACH study for 50,000 participants in the Medicare population across 3 years. We believe we are well positioned in the field. We have made the investments required to drive a new paradigm in early cancer detection and are looking forward to years of growth ahead. We recognized early on the complexity and scale of the opportunity in front of us and have brought together an executive team based on that knowledge. We've built an exceptionally strong team to take this company forward. Our leadership has extensive experience and a proven track record of execution at industry-leading organizations. We're also very pleased to announce our new board members, Greg Summe, Steve Mizell, and Bill Chase. We are excited to work with these three experienced leaders as we take GRAIL forward.
Now I'll turn it over to Josh to describe the promise of multi-cancer early detection.
Thank you, Bob. Next slide. So now we're going to talk about multi-cancer early detection. Now, as many of you know, the human and economic toll of cancer is enormous and is simply devastating. Nearly every one of us has a story about someone important to us, who's been impacted by cancer. In fact, there are 19 million new cases of cancer globally every year, along with 10 million deaths. Today, cancer is considered the number 2 cause of death globally, but is projected to become the number 1 killer very soon. In a recent WHO analysis, forecasted the number of new cancer cases will rise 77% to a total of 35 million cases by 2050. Next slide. There is simply nothing acceptable about the status quo today in cancer screening. Most deadly cancers are simply found too late.
This is both a problem statement as well as what we view as one of the most enormous public health opportunities at population scale. On the problem side, current cancer screening is extremely limited. The reality is that 80% of the cancer deaths are coming from cancers we are not even looking for and do not have standard of care screening in the United States. We also know that 86% of cancers are not found through screening at all, and that proportion is even higher outside of the U.S. On the opportunity side, is that we know that we have a 4 times improvement in survival rates when cancer is detected early in its localized stage. So how do we capitalize on this opportunity? Next slide.
While multi-cancer early detection is the way to enable early cancer detection at population scale, we simply must find more cancers through screening before they present symptomatically. This may be our only hope of really bending the cancer mortality curve. Today, most people receive 2 or 3 single cancer screening tests, and current cancer screening is not finding enough cancer in the population. In fact, these 5 screening tests, one cancer at a time, are only finding about 14% of incident cancers in adults at elevated risk. That is just not going to do the trick. It is clear that we need to find many more cancers early, and we need to screen for many of the less common cancer types.
As you may know, there are over 200 cancer types, so it's simply impractical to add 10, 15, or 20 more single cancer screening tests to our current single cancer screening paradigm. This would result in very high costs and high cumulative false positive rates that would be prohibitive, prohibitive because they would really overwhelm the healthcare system with all these false positives, and I'll show you some data about that in a moment. Single cancer screening tests are simply impractical for most of the less common cancers, and they will never be cost-effective. And remember, these less common cancers cumulatively account for the majority of cancer deaths today. So we believe that a single MCED test that finds a shared cancer signal across many cancers and can identify where in the body that signal is coming from is the best solution to this vexing healthcare problem.
Next slide. This is the illustration. Simply adding single cancer tests, more and more of them, will not get us to where we need to be. If you look at the results of the PLCO study, a large trial sponsored by the NCI, sequential screening with a number of single cancer tests for only four cancer types led to an individual's false positive rate of over 50%. And if you look at this illustrative example, you can see why. For each single cancer test, they each carry their own false positive risk, which you have to add up at the individual level. And once you add a few of them together, you see that this false positive rate quickly becomes unacceptably high. Next slide.
So when evaluating a screening test, one of the most important criterion is to look at the number of cancers detected, the yield, relative to the number of false positives generated. So let's look at what the world looks like with and without Galleri. Without Galleri, today, in the top row, standard of care screening identifies less than 200,000 cancers in elevated-risk adults, and in the process, will generate over 8 million false positives every year. According to our modeling, when you add Galleri, we could identify approximately 460,000 more cancers, and because Galleri's been designed with such high specificity, it will minimally increase the number of false positives, making multi-cancer early detection testing one of the most impactful opportunities in healthcare.
Adding Galleri to standard of care screening completely improves the signal-to-noise ratio from approximately 1 to 45 to a combined ratio of approximately 1 to 15. So here you can see that by adding Galleri to standard of care screening, the entire screening system gets much more effective and much more efficient. We can identify many more cancers much more efficiently and reduce the cost to diagnose a single cancer by approximately 65%. Next slide. So through this comprehensive research, GRAIL has confirmed that our methylation platform can identify a cancer signal that is shared by many, many types of cancer. This is the fundamental breakthrough that marks GRAIL's innovation and leadership in this whole field. The test is designed to be used in asymptomatic adults at elevated risk for cancer as a complement, not a replacement, to single cancer screening tests that are highly recommended today.
This is to drive earlier detection of cancers, to enable more treatment options and better outcomes. Abnormally methylated DNA is a hallmark of cancer, and additionally, that abnormally methylated DNA can also tell you where in the body that cancer signal is coming from. In other words, it can predict the cancer signal's origin. Galleri was designed to have a very high specificity to limit false positives and to drive a very high positive predictive value, one of the key clinical measures that I'll talk about more in a minute. Our technology also has a number of other benefits, including a very low clinical limit of detection, a cell-free DNA signal that preferentially detects highly shedding cancers that are typically invasive and very aggressive. Next slide.
Positive predictive value is perhaps the most important clinical measure, and just to recall what that means, positive predictive value is, among those who have a positive test result, what is the likelihood that they will have invasive cancer at that time? You can see here that the PPV for single cancer screening tests is often in single digits, and that the Galleri test stands out among all of these single cancer screening tests with its very high positive predictive value because of its low false positive rate. These features support the impact for population screening. Next slide. Critically, Galleri is actually working in the real world. As you can see from the quote here, our commercial use of Galleri is finding lethal cancers early in the real-world setting. In this case, Roger talked about his early pancreatic cancer that was found.
The majority of these early-stage cancers that we're finding do not have any screening tests. And remember, that when you find localized solid tumors, not only are there effective treatments, but there are often curable treatments, typically with surgery and radiation. We're immensely proud to be able to help patients and providers find these cancers early and enable positive outcomes for patients. Next slide. So what does the test report actually look like? It's not a 20-page report of hundreds of mutations. The test is meant to be easily implemented into clinical practice. It is prescribed alongside recommended screening, and the test results are communicated back to the provider. We designed the test report to provide valuable information for a positive or negative signal, and importantly, it shows the signal origin, the prediction, to allow the doctor to get a very directed and efficient diagnostic evaluation. Next slide.
Because this is so new, GRAIL provides a variety of support services to providers to optimize the path to diagnosis. After a positive signal is reported, GRAIL's medical science liaisons are in contact with physicians within hours to provide support, material, and other resources. We have resource and navigation services available to the physician and the patient. Physicians can elect for peer-to-peer consultations, and for those unusual but very difficult cases, we have a tumor board and a retesting program that is also available. Now I'll hand it back to Bob.
Thanks, Josh. Shifting now to our commercial strategy. Next slide. In the short term, our commercial strategy is focused on multiple stakeholders as we work to build awareness and capabilities to enable scale once broad access is available. Our focus today is on wellness-driven clinician engagement and innovative, value-oriented partnerships, and we are encouraged by the engagements we are seeing in the market. We are seeing enthusiasm from a broad set of health systems, employers, and life insurance partners. To date, we have more than 10,000 ordering providers and more than 100 commercial agreements in place with leading health systems, forward-looking employers, and life insurance partners who recognize the value of detecting cancer early. We are gaining insights about the MCED market today, as well as meeting the early market demand.
Our targeted commercial sales force promotes the physicians, as we also have business-to-business teams in place, driving partnerships and supporting pull-through. In addition to our other channels, we have established partnerships with early adopter payers, as well as groups with potentially higher risk for cancer, such as unions and first responders. Next slide, please. Our focused commercial team drives adoption through direct promotion to clinicians in partnership with the enterprise team, customer service, health system sales, and marketing. We have approximately 200 sales personnel across our channels, and we're taking a targeted approach in the short term to maximize the opportunity before broad reimbursement is available. We're also focused on market education, as we are the leading player in MCED with the largest share of voice. Next slide. Beyond the U.S., the NHS England rollout would represent a substantial commercial expansion.
The ongoing NHS-Galleri trial is a randomized control study of 140,000 participants, aged 50-77 in England. We have an unprecedented commercial agreement in place with the NHS, which could enable national-scale implementation after the full study results in 2026. National rollout will be deployed by an NHS invitation system, enabling significant scale with limited commercial resource requirements. The test is already UKCA marked, and no additional regulatory processes will be necessary to initiate market deployment if the NHS decides to initiate a pilot. We anticipate that the NHS implementation could carry major influence on markets globally. Next slide. We have already built out laboratory infrastructure to scale to enable future growth. We have 65,000 sq ft of CAP-accredited and CLIA-certified facilities between our Research Triangle Park, North Carolina, sites and our Menlo Park, California, site.
In addition, our RTP site alone has approximately 200,000 sq ft, enabling us to continue to scale laboratory capacity substantially with sufficient capacity to support multiple years of growth. In our next version of the Galleri test, to be launched at the end of this year, we've integrated significant automation, which supports both large-scale testing and scale, as well as driving down cost significantly. We have already demonstrated significant scale, as we've run approximately 450,000 Galleri commercial tests and research tests to date. We also have office sites in Washington, D.C., and London, and over 1,300 employees. Next slide. We are looking forward to a major inflection point with FDA approval of Galleri. We anticipate FDA approval will enable broad access with commercial health plans.
In addition, we anticipate passage of the Medicare legislation that would authorize CMS to cover multi-cancer early detection tests that are FDA approved. Following passage of that legislation and FDA approval, CMS would be able to cover Galleri. Broad access enabled by commercial and Medicare coverage is expected to transform the MCED opportunity, and the work we are doing today will help us prepare for the scale to meet that opportunity. Now I'll turn it over to Harpal to walk through GRAIL's scientific background and clinical evidence. Harpal?
Thank you, Bob. So let's now talk a little bit about how our technology works. Next slide, please. So many of our cells shed fragments of DNA into the bloodstream as they turn over or die, and this is also true of cancer cells. And DNA from those cancer cells contains information that can tell us something about the cancer. So the first step is that we look for these DNA fragments and isolate them from the blood plasma. We then sequence these DNA fragments using next-generation sequencing, looking particularly at methylation sites across the genome. Next slide, please. Now, the role of methylation in normal biology is to tell genes when to switch on or when to switch off, and methylation is a process that silences those genes.
And although methylation is established during normal human development, we know that methylation patterns are modified during aging, by disease, or by our environments, and this enables us to distinguish cancer from non-cancer. What we have found is that there are patterns of methylation associated with cancer that are different from the patterns that we see in normal cells. Increased methylation, which turns genes off, we see in regions which usually suppress tumor development. By corollary, we see reduced methylation, in other words, turning genes on, in regions which usually promote tumor development. And what we've seen is that these patterns are shared across many different types of cancer. Now, abnormal methylation and epigenetic reprogramming, in general, is a hallmark of cancer. And moreover, epigenetic programming is also a fundamental process which determines how individual cell types are formed or differentiated from a starting embryo.
And so these methylation patterns can also give us a fingerprint of the cell type that is giving rise to the cancer signal, and this enables us to predict the cancer signal origin, as I'll come onto in a couple of slides' time. Compared to other genomic alterations, methylation is a much more ubiquitous signal. There are around 30 million methylation sites across the genome, which provides a rich basis for interrogation, whereas mutation panels typically only examine a relatively smaller number of sites. And for all these reasons, methylation seemed like an excellent opportunity to look for a cancer signal. Nevertheless, GRAIL wanted to be absolutely sure that this was the best approach. Next slide, please. So we started with a completely unbiased discovery program.
We explored multiple different approaches to see which approach would give us the best signal-to-noise ratio when looking for a cancer signal across a diverse population of people with different backgrounds and different potentially confounding conditions. We explored mutations, fragment lengths, chromosomal changes, and methylation. In exploring these different approaches, we were inspired by the concept of limit of detection from analytical studies, and we looked for a clinical limit of detection to understand our clinical signal detection performance. We defined this as the value where the probability of detecting cancer, a cancer case, was 50% at 98% specificity. You can see in this chart how the different approaches performed in that context. The error bars you can see are the 95% confidence intervals for each approach.
If you look down towards the bottom of this chart in the beige, you'll see that whole genome methylation gave us one of the lowest clinical limits of detection. What we want to see in terms of a limit of detection is the lowest value possible. Then, with further development and deeper sequencing, looking at a smaller number of methylation sites, we developed what we call our targeted methylation panel. As you can see in the blue here, at the bottom of the slide, this gave us an order of magnitude further improvement. Really importantly, adding other approaches to this targeted methylation approach did not improve performance any further. Next slide, please.
So the way our test works is that having sequenced the DNA, looking for these methylation patterns, we then run this information through a two-stage classification process using advanced machine learning. The first stage is that we look for patterns that enable, enable us to discriminate between DNA that's coming from a tumor, from DNA that might be coming from normal cells or other signals in the body, and this enables us to call a cancer signal. The second step then is if we, if we have detected a cancer signal, we use that same methylation information to predict where in the body the signal has come from, and we call this the cancer signal origin. Because we can predict the cell type that's given rise to that signal, as I touched on earlier. Next slide, please.
Now, cancer signal origin prediction is increasingly recognized as a critical component of a multi-cancer early detection test. What we want to be able to do is to efficiently guide the next stage in the assessment of whether and what type of cancer a person has, once we've detected a cancer signal in their peripheral blood. Because until this point, we have no knowledge of the location of the disease in the body. What you can see on this slide, on the right-hand side, is statements from a number of opinion leaders on the importance of having this CSO capability and accuracy. And what you can see on the left-hand slide, the left-hand side, is a quote from the FDA summary report, following an advisory committee meeting held in November regarding multi-cancer detection.
The panel believes that multi-cancer detection tests should have a tissue of origin component to the device, as it would guide targeted diagnostic work-up and minimize the risks." So that's a bit about the science and the technology. Next, we have to prove that this approach works in the real world. And so in developing Galleri, we've undertaken one of the largest clinical evidence programs in the space, designed to validate our approach and generate the evidence we need to secure regulatory approvals and reimbursement. Next slide, please. This is a program that's involved multiple studies totaling over 385,000 participants overall. More than 21,000 of these participants were included in studies that supported the development and launch of Galleri, summarized on the left-hand side here.
A further more than 170,000 individuals have been enrolled to date in interventional studies, which will support our PMA, summarized in the second column here. Let me start with our discovery and clinical validation work. The first study we undertook was our foundational study, called the Circulating Cancer Genome Atlas study or CCGA. This was an observational study, a case-control study with around 15,000 participants. The first two parts of CCGA were focused on deciding which technology to use, what I've already summarized in terms of our unbiased discovery effort that led to us choosing methylation as the technology to move forward with. CCGA 2 enabled us to further develop that targeted methylation assay to improve sensitivity and specificity.
We then used the third part of CCGA to validate this clinically in the case-control population, and it was this that established our initial performance parameters for our multi-cancer early detection test. Pathfinder was a study of more than 6,000 individuals in the U.S., and this was our first implementation in the intended use, asymptomatic population. What we were really pleased to see was that the performance replicated what we had seen in the CCGA case-control study, and what we had predicted for our performance. Importantly, we did not see the degradation in performance that's typically associated with the transition from case-control studies to asymptomatic populations. This speaks both to the quality of our test design, but also the rigor of our science.
CCGA and Pathfinder have both been published in peer-reviewed journals, and the results have been presented at major conferences, including the American Society of Clinical Oncology and the European Society of Medical Oncology, or ASCO and ESMO. Moving to the second column here, we are currently conducting two registration-enabling studies. NHS-Galleri is a study of over 140,000 people being carried out in England. It's a randomized controlled trial designed to demonstrate the clinical utility of our MCED test at population scale. The primary objective will be to demonstrate a reduction in the incidence of late-stage cancers, what is sometimes referred to as stage shift. Because this is also a very large study, we'll also be able to use it to confirm our performance. We'll be able to generate large-scale safety data, including understanding what happens with post-positive diagnostic investigations.
We'll be able to to model the mortality benefits of implementing a screening program using our MCED test. And we'll be able, for the first time, to look at the value of screening annually, because NHS-Galleri involves screening people over three rounds, spaced one year apart. NHS-Galleri is fully enrolled, and we're very close now to completing that third round of screening, and we will complete the blood draws in July of this year. Pathfinder 2 is a 35,000-person study being carried out in the U.S., of which over 30,000 have already been enrolled. Pathfinder 2 will generate additional U.S.-based safety data in a much larger study, and we'll also be able to demonstrate data across a number of different diverse groups, including a number of ethnic groups.
So these two studies collectively will support our PMA, which is currently in process with a rolling submission under a Breakthrough Designation. Over on the right, we'll soon launch our Galleri Medicare study. This is a real-world evidence study, which will enroll 50,000 individuals, again, allowing for 3 annual tests, in this case, in a Medicare population, together with a synthetic control. Galleri Medicare will provide additional evidence on the clinical validation and utility and support a Medicare reimbursement decision after FDA approval. And I will stress here that Galleri Medicare is intended to be a post-PMA study, and you'll hear more about this study from Josh in a little bit. So these are some of the critical studies that are involved in validating the performance, but also then generating the evidence that will support approvals and reimbursement.
So let's now look at some of the data, starting with our CCGA study. Next slide, please. So CCGA demonstrated that we can detect a cancer signal across more than 50 different types of cancer. Importantly, we've been able to do this with an extremely low false positive rate of 0.5%, which compares with some of the much higher false positive rates that Josh showed you earlier. And together, this means that we have modeled with CCGA that we would have a positive predictive value of 44%. In terms of our ability to predict the cancer signal origin, we've been able to do that with an accuracy of just under 90%. And really critically, are we finding cancers early? Well, we pre-specified a group of 12 types of cancer that represent just under two-thirds of all cancer mortality in the United States.
What we found was an almost 70% sensitivity for finding these cancers at stage two, which is regarded as an early-stage cancer, for which almost always there is a curative treatment option, available. CCGA enabled us to establish our critical performance parameters for MCED testing, and I'll now hand over to Josh, who can tell you what we saw in our Pathfinder study.
Thank you, Harpal. Next slide, please. What I really want to emphasize here and reiterate what Harpal mentioned is that we did an important interventional trial called Pathfinder, and we show here that the data in Pathfinder actually replicated what was seen in our case-control study. Just to emphasize what Harpal said, this is very unusual to see in this field. In fact, it's not been seen in this field until now. Pathfinder showed that our positive predictive value and CSO prediction were consistent with what we saw from our case-control CCGA study. We did not observe the degradation that so many in the field had seen, and that was, quite honestly, expected. Specifically, our PPV at 43% is in line with what we published and modeled from CCGA.
So this evidence together supports the generalizability and the actual robustness of Galleri in our interventional study, involving Galleri return of results on the clinical, diagnostic, and care pathways in the intended use population outside of a case-control study. Really, really important. Next slide. In Pathfinder, which we presented at ESMO in 2022, we showed that Galleri more than doubled the number of cancers identified when added to standard of care screening. And we're really pleased that these studies continue to demonstrate consistent results across all of our observational and interventional studies. Almost half of the cancers detected by Galleri were in their early stages, 1 and 2. The PPV, as I already mentioned, was 43%. Again, an order of magnitude higher than the PPV seen from leading single cancer screening tests shown here on the slide in green.
The localization accuracy, critically important, and the CSO was also very high, leading to very efficient diagnostic workups. Critically, there were no serious adverse events related to the diagnostic workups in the study, and patient satisfaction scores were uniformly high, even among those who had false positive results. Next slide. So we believe that the reason our test performance is so consistent has been enabled by the fundamental biologic features of circulating cell-free DNA. Biologically, we know that higher levels of tumor cfDNA are associated with more aggressive cancers across all stages. So there are early-stage cancers that are shedding lots of DNA, and there are late-stage cancers that may not be, so it's very stage dependent. Because of this, our MCED test preferentially detects the more lethal cancers, the invasive and aggressive cancers, which may help avoid overdiagnosis. And of course, this is what we would want.
We would want an MCED test to preferentially detect the cancers most likely to result in death, aggressive and clinically significant cancers warranting treatment. This is what Galleri detects. Next slide. This slide is a little complex, but it illustrates the point, I think, very well and has now been published. Data across our clinical studies suggests that although Galleri detects signals for some of the most aggressive cancers, the detection of cancer signals for slow-growing or indolent cancers, which people are less likely to die from, is actually quite low. These are Kaplan-Meier survival curves, which shows survival over time. Cancers detected by Galleri, seen here in blue, have a very similar course and prognosis to that expected, based on our analysis of the SEER database, which are shown in hashed lines.
You see the blue line of Galleri-detected cancers really sitting on top of these hashed lines, which is what you would expect based on SEER. At any given stage, survival was worse with cancers detected by Galleri compared to cancers that were not detected by Galleri. More specifically, the blue curves, which represent the Galleri-detected cancers, are steeper than the purple curves, which represent undetected cancers, meaning the cancers detected by Galleri had a prognosis similar to SEER, whereas those not detected by Galleri had a prognosis much better than what would be expected. These findings were consistent across stages and across age. What these data suggest is that indolent cancers, the ones that people are worried about, that we are over-diagnosing and over-treating, the ones that people may die with and not die from, are unlikely to be detected by Galleri.
Thus, Galleri is unlikely to contribute to this problem of over-diagnosis and the harms associated with overtreatment of these over-diagnosed cancers. Next slide. Now we're going to go into a little more detail on our registrational program. Next slide. These studies are progressing well and are on track. We have enrolled more than 30,000 of the planned 35,000 in Pathfinder 2, and in the NHS Galleri study, it's anticipated to complete the third and final round of blood draws in July. Study results for the NHS Galleri trial are anticipated in 2026. Interim results from Pathfinder 2 are expected in the second half of 2025, after we have a full 1-year follow-up on the first 25,000 participants. Together, as Harpal mentioned, these studies include approximately 175,000 participants and will be the pivotal package supporting our PMA submission.
I think you can all appreciate the depth and breadth of our clinical genomics program. Next slide. The NHS Galleri trial is, you know, really a unbelievable study trying to address the long-term ambition of the NHS, which is to reduce the proportion of cancers detected in their late stage and increase early-stage detection. This is the first randomized controlled study of an MCED test of 140,000 participants between the ages of 50 and 77 to evaluate the clinical utility for population screening. We're on track to complete final study visits in the third quarter of this year, and final results anticipated in 2026. Next slide. Pathfinder 2 is a multicenter study in the United States with return of results in our US healthcare system. It's enrolled over 30,000 participants to date, and this is really our pivotal US performance and safety study.
While the UK NHS trial is three consecutive years of annual testing, this is one year of testing with three years of long-term safety follow-up. Next slide. In November, we announced the REACH Medicare study. Here, as you know, Medicare beneficiaries are among those most at risk for cancer due to age and other risk factors. So there's an enormous unmet need for earlier cancer detection in the Medicare population. We announced the Galleri Medicare study, called REACH, which is a prospective, multicenter study to assess the clinical impact of Galleri when added to recommended standard of care screening in the Medicare population. This study will recruit 50,000 individuals who will receive usual care, plus Galleri for three consecutive years. Because this study was approved by the FDA as an IDE, the cost of the test and the related services for study participants will be covered by Medicare. The individuals receiving Galleri will be compared to a similarly sized cohort from EMRs to understand the clinical impact of Galleri in this real-world cohort. Next slide. Now, real-world evidence is a critical part of our entire evidence program. We launched Galleri in 2021 and have been committed to studying the impact in real-world use. We just mentioned the Galleri Medicare study. That will be a real-world study to help us understand the actual clinical impact of Galleri under usual care conditions. In addition, we designed the REACH Galleri Medicare study
Beyond that study, we have an ongoing clinical surveillance program, which enables us to assess performance and safety from cases that we follow, to generate data and to systematically evaluate evidence from our commercial use. These learnings are critical to enable our own test performance improvements over time, but also to enable the paradigm shift in early cancer detection. Next slide. Importantly, we are making great progress in our rolling modular PMA. As you know, we're in Breakthrough Designation with the FDA. We have an ongoing modular submission that is progressing well. Importantly, all of our product development efforts that support Galleri updates and our PMA submission have now been aligned. They're all occurring under full design control to ensure we meet the regulatory standard. And now to talk about the opportunity beyond asymptomatic screening, I'll hand it back to Harpal.
Thank you, Josh. So everything we've spoken about so far has been asymptomatic screening. But as we think about what we've developed in GRAIL, we've developed not just a product, but a platform. Our targeted methylation platform can enable us to provide solutions across the continuum of cancer care. And we're really excited about two areas, in particular, symptomatic detection and precision oncology. Next slide, please. So let me talk about symptomatic detection first. As we have already alluded to through this presentation, the vast majority of patients who are diagnosed with cancer, over 80%, first are detected because they present with symptoms. Now, sometimes those symptoms can be very obvious. For example, a breast lump or a skin lesion, and in those cases, the clinician knows what to do next and knows which investigations to undertake.
But in a very large number of cases, the symptoms that a patient presents with can be very nonspecific. So these can be things like unexplained weight loss or anemia, and in these cases, cancer might be the reason that the patient has symptoms, but it could be something else altogether. And if it is cancer, it could be many different types of cancer. Now, because cancer is one of the possibilities, these patients typically undergo extensive investigation. We know that more than 70% of these patients will undergo some form of imaging. Over a third will have a biopsy, and in many cases, patients will have several different investigations. But in practice, only around 4% or fewer of these patients will actually be diagnosed with a cancer.
We also know that many of these patients, over a quarter, can take many months to get to a diagnosis, because they go through a sort of diagnostic odyssey as they go through these multiple different investigations. So we believe that multi-cancer early detection offers the opportunity to realize this unmet need to accelerate diagnosis where a patient is presenting with these nonspecific symptoms, but also to avoid harmful procedures where cancer is unlikely to be the reason that the patient has those symptoms. This is a very large unmet need. We estimate that between 15 and 20 million patients every year in the United States are presenting with these nonspecific signs and symptoms.
The other really interesting and exciting opportunity with this, with this unmet need, is that there is the possibility to pursue a reimbursement approach using existing coverage pathways, because this would be a molecular diagnostic test. So because of this opportunity, and because we believe our technology is ideally suited to this opportunity, we've already started generating evidence for this group. Next slide, please. So we've carried out a study that we call SYMPLIFY. This was carried out in the UK in partnership with the University of Oxford and enrolled more than 6,000 people. And we were very excited by the results that we saw. First of all, our PPV with this group, as you can see here, was over 75%.
We also had an extremely high negative predictive value, and this is really important in this population because we don't want to miss a cancer if a patient does indeed have it. As you can see towards the bottom there, we also had a very high accuracy in our CSO prediction, and this is really important, again, in this population, remembering that these are patients that are presenting with nonspecific symptoms. So if I go back to the example of unexplained weight loss, unexplained weight loss can manifest in up to 15 different types of cancer. So being able to direct the next stage of clinical investigation, if someone does have a cancer signal, becomes incredibly important in this population.
Indeed, we know that often the cancer type that the patient was ultimately diagnosed with was different to the one that the physician initially expected or suspected from the way that the patient presented. Interestingly, we saw the highest performance in patients referred for a suspected upper GI cancer, which, as it turns out, is where the largest unmet need is in this population. Next slide, please. The second area that we're really excited about is precision oncology. This is the notion that once a patient has been diagnosed with cancer, how do we optimize their treatment to drive the best possible outcome for them? Now, precision oncology is a rapidly growing field with a significant unmet need and therefore a significant market potential. And we see applications for our targeted methylation platform across this continuum.
Most of the market focus so far in precision oncology has been in what we call MRD, which is the advent of tests to detect the presence of microscopic disease after curative intent treatment, because often this can presage a higher risk of recurrence or relapse. But we also believe that there are really important use cases, first of all, in pre-treatment prognosis, because if a patient has a much worse prognosis, we might want to intensify treatment at that earlier stage. In being able to detect recurrence before symptoms reappear, or before there's clinical evidence of recurrence, because again, we might want to start treatment earlier in this population. And last but not least, in monitoring patients when they are on treatment, to see if they're still responding to that treatment or not, or should we change the treatment regimen that they're on?
Next slide, please. Most of our work so far in precision oncology has been in partnership with biopharma companies, and we've been working with a number of the leading oncology companies to explore the potential of our technology in their clinical programs. This area of work is also starting to generate meaningful revenue for GRAIL. One example is a strategic collaboration with AstraZeneca, which we announced in 2022, to develop companion diagnostics for patients with earlier stage disease. Indeed, we recently announced the start of a prospective clinical trial with AstraZeneca in Japan, in stage 1 non-small cell lung cancer, using a customized GRAIL test. Importantly, we also see significant potential to expand beyond biopharma with our own clinical products, for example, in MRD, and we're in development for the first of those. Next slide, please.
Now, we believe our technology has a number of significant advantages in this space. First of all, our platform doesn't require us to develop a bespoke panel for each patient, and neither do we require a tissue sample. We just need a blood sample with limited plasma input. We're able to deliver a very quick, between 7 and 10 days, clinical turnaround, which is critically important when we're dealing with cancer patients who may have aggressive disease. And we can also provide a quantitative measure of the tumor burden, so that we can track how patients are doing over time, particularly in that clinical monitoring phase.
Of course, because of everything else we've already described, this has application across multiple different types of cancer, which is attractive both to pharma companies and clinicians, because it means that they can work with, and get familiar with, a platform that they can use across multiple different types of cancer. Now, we validated our performance, and we have demonstrated. Indeed, our pharma partners tell us that we have sensitivity on a par with the tumor-informed methods that they often use. We have robust and reproducible test performance, and we have an extremely low assay failure rate, which again is critical in this population, and again is something that our biopharma partners tell us is very important, vis-à-vis some of the other platforms that they use.
What we do in the precision oncology space is that we're able to optimize class classifiers for specific use cases and indications. We're very excited with some of the data that we're seeing. Next slide, please. We've now started to present and publish some of this data at key conferences, such as the American Association for Cancer Research. What you can see summarized on this slide is some of the different use cases and cancer types on which we have already presented data. But let me talk about one example, specifically, on the next slide, which is data that we presented at the North American Lung Conference last December. Next slide, please. So this is a lung prognosis test that is carried out prior to surgery with Stage One lung adenocarcinoma patients.
And what we would demonstrate, what we were able to demonstrate with this, with this study, is clinically meaningful risk classification. So on the left-hand side here, you can see a Kaplan-Meier curve, similar to the ones that Josh showed you earlier. And what this shows is the relapse-free survival of patients who either did have a positive test, those in purple, versus those who had a negative test in the yellow there. And what you can see here is that this clearly discriminates patients at high risk of early relapse from those at lower risk of early relapse. And the conclusion here was that test positive stage one patients experienced recurrence rates similar to unselected stage two patients, suggesting that these patients would benefit from intensified therapy at the earlier stage.
This study had a hazard ratio of 3.8, as you can see here on the right-hand side. It's this assay that we're using in the Japan study that we're carrying out with AstraZeneca, that I alluded to earlier. So we're really very excited by the possibilities that we see beyond asymptomatic screening to further build the GRAIL business using our core platform. This is a key part of our story going forward. With that, I'll now hand over to Aaron to talk about some of the financial aspects of our presentation. Aaron?
Thank you, Harpal. As Harpal said, we'll now spend some time looking at the financial profile. Next slide, please. GRAIL's revenues for Q1 2024 were $27 million, which was approximately 36% growth over the same period of the prior year. Full-year revenues in 2023 were $93 million, representing approximately 68% growth over the full year of 2022. As Bob's explained, we are seeing early adoption of Galleri prior to broad reimbursement. We are making an investment in building the MCED market category, which will allow us to drive uptake more effectively once we have broad regulatory approval and reimbursement. Since launch, we have been learning which investments drive commercial pull-through most efficiently, and we know that sales will titrate based on our level of investment.
Going forward, we expect our revenue growth to fluctuate based on the level of investment we make in building the MCED market. In the near term, we expect continued investment focused on obtaining broad regulatory approval, reimbursement, and access. We are focused on making significant reductions in our COGS per test and bringing down our operating loss over time. As we progress through major milestones, we expect to have line of sight to profitability. Next slide, please. At separation, we'll have approximately $1 billion of cash on our balance sheet. With 2.5 years of cash, we are fully funded through the end of 2026 and are well positioned heading into major milestones, including NHS data readout and the filing of our PMA. We expect our core US Galleri business to grow at 30%-50% rate in 2024.
This rate does not include any potential contribution from an early pilot with the NHS or other international opportunities. We are also not including any guidance on our pharma business, as it can be lumpy, as timing of projects are dependent on specific pharma program timelines. In 2023, our cash burn, adjusted for the cash-based LTI program that will convert into an equity-based program upon spin-out, was $532 million. Our cash burn for the second half of 2024 is expected to be approximately $250 million. We have levers to pull on across our P&L to manage our burn and ensure we can invest in our core value drivers, including UK implementation, MCED approval, and a commercial rollout, and growth of our precision oncology business. Next slide, please.
Galleri has established commercial and set leadership and achieved a greater than 90% year-over-year revenue growth in 2023. Through March of 2024, GRAIL has sold more than 180,000 commercial Galleri tests, and significant runway remains, with more than 100 million individuals in the intended use population in the United States alone. We continue to make significant investments to increase market awareness, to drive expansion of the Galleri provider base, and implementation of early detection clinics. Key drivers for the future include pricing and reimbursement, international deployment, establishing consistent annual testing intervals, and driving screening compliance. The market opportunity will transform with FDA approval in the US and obtaining national or commercial payer coverage broadly. GRAIL has invested in technological and automation improvements that will support price reductions to serve a broad population and enable acceleration, accelerated adoption. Next slide, please.
There's a significant need in the oncology market to identify residual disease or recurrence early to inform treatment decisions. To date, we have ran more than 6,000 samples across multiple pharma partnerships. These partnerships are generating a pipeline of companion diagnostic opportunities. We are additionally exploring opportunities to enter the market with a clinical LDT for precision oncology use cases. Next slide, please. We've talked about our infrastructure investments and the anticipated transition to an automated platform this year, which will enable a significant reduction in costs. Over time, we anticipate further volume-based efficiencies to get us to target gross margin of 55%-60%, which includes a royalty we will owe to Illumina after separation. Further development and technological advantages are additionally possible. Next slide, please. Today, our revenue is primarily from Galleri in the U.S., with contribution from our precision oncology business.
In 2023, U.S. Galleri revenue was $75 million, or 80% of total revenue. Over the next few years, we anticipate additional diversification of revenues as we launch internationally and introduce additional products. Importantly, in markets like the U.K., where launch in a national system is possible, the operating requirements are much more limited as compared to the United States. Because of this, we expect to drive a strong operating margin with lower pricing. Next slide, please. We have heavily invested in R&D since our founding, which has created a competitive advantage for us. As we complete the major clinical studies and roll out of our lower cost automation platform, we anticipate the R&D cost as a percentage of revenue to fall significantly, even as we continue to invest in innovation and new product development. Next slide, please.
You can also expect SG&A costs to become more efficient over time as we build Galleri channels and progress pre-reimbursement sales through focused initiatives. We expect our current sales force to become more efficient over time. Next slide, please. In 2016, when GRAIL was formed, we have developed an exceptionally innovative technology, driven market development of new cancer screening category, and built out scale to support the transition to population scale. We are looking today at a number of critical near-term catalysts that drive the next stage of growth. We are looking forward to these upcoming catalysts, including potential rollout of Galleri and an implementation pilot by the NHS, final data from our pivotal NHS-Galleri and Pathfinder 2 studies, FDA approval of broad reimbursement of Galleri, and additional product launches. Next slide, please. That concludes our prepared remarks.
Operator, we'd like to turn it over to questions now.
Thank you. We will now move on to Q&A. If you have connected via Zoom and would like to ask a question, please use the Raise Hand feature at the bottom of your Zoom window. If you've dialed in by telephone, press star nine to raise your hand. Once called upon, please feel free to unmute and ask your question by unmuting your device microphone or dialing star six. We will ask that you limit to one question and one follow-up at this time. I will now pause a moment while we wait for the queue to assemble. Our first question comes from Vijay Kumar from Evercore.
Hey, guys. Thanks for hosting this and taking my question. I guess my first question is, Josh, perhaps, you spoke about PPV, positive predictive value. And when I look at the screening space, there's a lot of factors and metrics which are being thrown around, right? For different cancers, different metrics being thrown around, whether it's early stage, asymptomatic, you know, sensitivity. Like, I guess when you look at MCED, how do... What do the guidelines say on what metrics are, you know, bodies gonna look at? Is it PPV? Why does PPV matter for MCED tests? Why shouldn't we look at traditional metrics like a sensitivity metric? And how well understood is this metric by physicians, right?
When you think about test adoption, is this a well-understood metric? Maybe some thoughts on that.
Yeah. No, it's a great question. So again, the traditional metrics will obviously be looked at. Remember, single cancer screening tests really focus on being highly, highly sensitive, 'cause you don't wanna miss anything, and you wanna find everything, and you accept a very high false positive rate. You can look at mammography, low-dose CT, stool-based colon cancer screening, go down the list. That's quite different than what you want in multi-cancer early detection tests, where you can detect somewhere between 50 and 70 different types of cancer. You can't accept that high false positive rate. So you really focus on specificity, and our specificity for our assay is at the cancer signal level. So we have one specificity, and then the sensitivity will vary by cancer type, depending on how much tumor DNA is being shed.
You have high shedding tumors, which are aggressive and invasive, which will have very high sensitivity, and lower shedding tumors, kind of the slow-growing indolent ones, where there will be lower sensitivity. The reason PPV is so important, because as a clinician, you do not know what the cancer status is of the patient, and so the positive predictive value becomes the most important thing you deal with as a doctor. Yes, doctors are very familiar with PPV, and that is what matters most to the doctor, because that tells you if I have a patient in front of me with a positive result, how likely is it that they're actually sitting before me with cancer? In our assay, it's somewhere between 40% and 50%. Again, an order of magnitude higher than that seen with single cancer screening tests.
The short answer to your question is we will look at all the traditional metrics. There will be a single specificity, and PPV becomes really the critical metric for multi-cancer early detection testing to avoid all the false positives. The other critical metric, of course, is the signal origin prediction, which enables a very efficient and effective diagnostic evaluation.
... That, that's helpful, Josh, and my follow-up perhaps is on when you look at the catalyst here, it looks like the NHS study in the back half of this year, that's the most near-term catalyst. So what are we looking for when the study is published? Is it PPV we're looking for, or are we looking at this cancer signal of origin? What kind of metrics are we looking at? And you did mention a potential NHS expansion of the commercial roll-out. Do we have some hard metrics on, you know, what's the minimum bar that we need to clear before this is rolled out in the UK?
Harpal, you wanna take that one?
Yeah, sure. Thanks, Vijay. Well, look, let me say a couple of things about this. First of all, the most important thing to say about NHS Galleri is that it's intended to provide meaningful clinical utility evidence when we get the final study results, which will be in 2026. And that will be the basis on which the NHS decides whether or not to implement a population scale program, you know, really rolling out a full-scale national screening program. But also, as we've already touched on, provides data for the FDA PMA submission. So though, that's really the most important thing to focus on with NHS Galleri, and as I say, we will get those results in 2026.
But what we also discussed with the NHS some time ago, was that if we see, just looking at the first year only of the study, if we see some really compelling information there, then the NHS might be willing to start an implementation pilot earlier than 2026. Now, I should clarify that that would only look at the first year of the study, and they would only look at certain metrics, which is not the primary objective of the study. So that's a process that's currently underway, and we'll know this summer. I should clarify, though, that there's no intention at this stage to publish those results from the first year of the study.
The only purpose of it is to enable the NHS to make a decision as to whether it's so compelling that they want to start a pilot implementation earlier than they otherwise would. In answer to your question about what they'll look at, they'll look at PPV, yes, and they'll also look at the number of stage four cancers and how many cancers we find overall. So that's the sort of thing that they're gonna look at, but... We'll know that, as I say, later this summer.
Thanks. Thanks, guys.
Thank you. Our next question comes from Sung Ji Nam, from Scotiabank.
Got that, I was on mute. Thanks for taking the questions, and thank you for the presentation. I have a couple of questions that are related, so I'll ask them at the same time. Just wanted to clarify that the FDA submission will be for asymptomatic population 50 years and older, and not targeting high-risk populations or, you know, symptomatic patient population, et cetera. Then related to that, kind of, you know, the question was asked in terms of endpoints that the FDA might be looking for, and I think you touched on, with the NHS, you know, looking for stage shifts. So do you think stage shift would be an acceptable kind of near-term endpoint for FDA and other regulators?
And then kind of when, you know, how are you thinking about, or how should we think about just kind of, you know, being able to determine mortality, right, longer term, for tests like an MCED test? Thank you.
Yeah, that's a really good question. Thank you. You know, the first piece, you know, so we are submitting for an asymptomatic, and it is for 50 and above, which is, you know, our view of what elevated risk is. You know, from a criteria standpoint, you know, maybe a few things from the FDA hearing. They were very clear that mortality data was not gonna be required, you know, for approval of an MCED. They also did say that CSO capability, which the Galleri test has, is a very important and critical criteria for an MCED test. And they also indicated that we would be able to use aggregated, you know, aggregated data as well.
So I think, you know, those elements speak, you know, work very much in the favor of Galleri and the way we've set up the technology. Josh, you wanna add anything to that?
Yeah, just that, you know, the FDA was pretty clear, they're gonna be looking at clinical validation, really not focused on clinical utility like mortality or stage shift. So we believe, you know, we will provide them with all the data that we have, and as Harpal explained, we will be looking at an absolute reduction in late-stage detection in the NHS-Galleri trial, which we will submit to the FDA over time, but that's not gonna be what they're focused on. They're clear that they'll be looking more for clinical validation and performance and safety. And so that's where we are with the FDA. On the NHS-Galleri trial, I think, you know, we've talked about the clinical utility endpoint that's gonna be really important for the whole field, to understand.
And there's, you know, there's a lot of discussion right now about, you know, what are the right types of intermediate endpoints to measure. Mortality, obviously, is something that's been the standard for single cancer screening tests.
... But, you know, we don't believe that doctors and patients, you know, should be waiting a decade or more, for mortality studies before, having access to this well-validated technology. So we don't, we don't think that those studies, are currently, you know, designed or powered, and the NHS study was not designed or powered for a mortality benefit.
Great. Thank you. Thank you. Our next question comes from Subbu Nambi from Guggenheim. Subbu, if you'd like to unmute yourself, please. Thank you.
Hey, guys. Can you hear me okay?
Yes.
Perfect. On MRD, besides the partnerships with biopharma, what is the plan for clinical studies? And will you be doing single cancer studies or multi-cancer studies, knowing that your test can be universal? Any thoughts on how it would compare to commercial tests, bespoke and non-bespoke MRD tests?
Yeah. Harpal, do you wanna maybe take that one?
Yeah, thanks. So, the first part of your question, yes, we would anticipate that MRD, the MRD use cases would be cancer specific. But to sort of the other part of your question, but the same platform should be usable across multiple different types of cancer. But evidence needs to be generated probably on a cancer type by cancer type basis. We have not yet started any clinical studies. But certainly, that would be the plan in the future. And so that, you know, that would be the anticipation. In terms of the performance, I touched on that. You know, the feedback that we're getting from our pharma partners is that they're seeing sensitivity on a par with some of the tumor-informed methods that they're using already.
So we feel, we feel very confident with our ability to have performance that's going to be acceptable in this marketplace. So I hope that answers your question.
Yeah, super helpful. For the pre-treatment prognosis product, how should we think about it commercially, mainly from a reimbursement standpoint and patients who are deemed eligible?
Yeah, it's a very important and good question. You know, at the moment, the work that we're doing with AstraZeneca would be in the context of a companion diagnostic. And so that's certainly the sort of development path that we're on there. As I said, it's already gone into one clinical study with AstraZeneca. And the idea there would be to think about the neoadjuvant or adjuvant treatment of Stage one non-small cell lung cancer patients. For those who are not completely familiar, the current standard of care with Stage one non-small cell lung cancer is surgery and then no further treatment until there's a clinical reason to do so. So the idea here would be, should we instead be invoking earlier adjuvant or neoadjuvant therapy?
At the moment, as I say, our focus is on its development as a companion diagnostic.
Makes sense. Thank you, guys.
Our next question comes from Doug Schenkel from Wolfe.
All right, thank you for taking my questions. So, I do wanna ask a clarifying question on data performance in a second, but I wanna first talk about, I think, some business questions before we get into the data. So Bob, how would you characterize the political risk in the U.S. for GRAIL, as it relates to CMS reimbursement and FDA approval hurdles, you know, heading into an election year, and how are you preparing for the different scenarios? So that's the first one. Second, and I think this is for Aaron, the company's situation with Illumina allowed it to really continue investment at a level that most others in the space haven't been able to do, just because of, you know, how the world has changed over the last few years from a capital market standpoint.
Keeping in mind where we are today and, you know, keeping in mind your move towards likely becoming a standalone company, how should we think about the pace of studies and your ability to invest at the same levels from an R&D standpoint? So let me pause there, and then I'll come back on the data question.
Yeah, thanks, Doug. You know, a couple of things. So you saw, you know, clearly the FDA came out with their, you know, new guidance and rulings. We believe that Galleri, you know, Galleri's, you know, well positioned in that. You know, Galleri's already has breakthrough designation. It's on the market as an LDT, and we're pretty deep into the PMA pathway. So we don't think any of the, you know, recent FDA rulings is gonna have any significant impact on our Galleri program. On the other programs that Harpal's talked about, we, you know, we are evaluating those, but, you know, the kinda core Galleri, we don't think there's an impact given our progress on PMA.
And so, you know, even, even in an election year, you know, things could, you know, probably only, you know, improve from there. But again, we don't think there's risk on the Galleri side. So the political, you know, thing of an election at the end of the year shouldn't have, shouldn't have impact on our ability to invest, and I'll let Aaron jump in a second, but, you know, clearly, if we go as a standalone company, we're gonna have to, you know, look, look very carefully at the, you know, certainly any additional studies that we're, we are doing. Many of the studies that we're engaged on are, you know, already well, you know, well in. You saw the, you know, level of enrollment, NHS is, you know, fully enrolled in, in its third year of study visits at London in July.
You know, Pathfinder 2 is 30 of 35,000 participants already enrolled in. So we're, we're already through, you know, kind of the bulk of it, and we're, you know, hopefully gonna be on the, you know, getting some advantage of coming out of, coming out of some of these studies. It will clearly, you know, we'll have to look very carefully in terms of, you know, future studies and what we do there, but, you know, we're feeling good that, you know, a lot of the work has been done, and we're, we're starting to see some of the benefit of that. Aaron, do you wanna add anything to that?
Yeah, just add a couple things. Yeah, Doug, so as Bob laid out, the primary studies that are underlying, you know, Galleri broad access, those are wrapping up. Additionally, the study that we'll begin in is the CMS Galleri study, which, as Josh pointed out, you know, CMS is covering the price of the test, reimbursing for it, and then also the workup. So it'll be a pretty efficient study to run from a capital perspective. Outside of the studies in R&D, we mentioned we're launching the automated lower cost, you know, version of the test this year. So that multi-year development effort will also be behind us.
So, you know, between, you know, the R&D components supporting Galleri, you know, winding down or coming to fruition, you know, we'll, we believe we've got levers through the rest of R&D, and then in sales and marketing to change levels of investment as we need to, as we cross these inflection points.
Okay, that makes a ton of sense. Thank you. Thank you for that. And then if I can just sneak in 2 quick follow-ups. First, have you shared any data on repeat testing? And I guess in this context, define repeat testing as folks that, you know, maybe got Galleri done one year and then, you know, came back a year later and did it again. You know, essentially, what proportion of folks have ordered more than one Galleri test? And then just a clarification, again, I could be wrong, and I may have misheard you, but I thought sensitivity on CCGA was around 50%, maybe a little bit higher, and then on Pathfinder, I thought it was in the 20s. Is that wrong? I just wanted to clarify there. I think that's a Josh question.
Sorry to, if that's off base, but I just wanna make sure I'm clear on that.
No problem. Just to clarify, sensitivity was not... The Pathfinder was not a study that measured sensitivity. It was a small study with a 1-year follow-up, so that was not reported from Pathfinder. In CCGA, there were a couple different sensitivity measures that were reported. One is, as Harpal revealed, the pre-specified subgroup of deadly cancers that account for almost two-thirds of deaths in the United States, the stage 2 sensitivity there was almost 70%. The 50% you're referring to was across all cancer types that were assessed in CCGA, including some cancers that don't shed DNA into the blood or very much DNA into the blood, indolent cancers. So that's the 50% you're referring to.
Okay, thanks for that, Josh.
And then just to get you, you'd asked a prior question about CMS reimbursement, just to share that, you know, there is bipartisan legislation that's been introduced to give Medicare, Bob mentioned this in his introduction, the authority to provide coverage for Galleri. It's not a mandate to cover it upon FDA approval. That legislation has a tremendous amount of support. Over 320 congressional sponsors, over 700 stakeholder groups supporting it. I think the stakeholders have really driven this whole process and made a great piece of legislation that is broadly supported across both aisles and in both houses of Congress. So we're very optimistic that even in an election year, that could get passed this year.
Okay, great. Our next question is from Puneet Souda, from Leerink.
Yeah. Hi, guys. Hopefully you can hear me.
Yes.
Thanks, and thanks for doing this. My first question, a bit of a clarification. Can you clarify what is your expectation for pricing in U.K.? And a related question on that is, you know, you know, how sustainable is the patient pay if the MCED reimbursement, as Josh talked about, there's support for that, but if that was not to materialize, what I'm trying to understand is, you know, reimbursement is a very big consideration for any diagnostic, and how you get paid for these assays is what is the sort of protection to the downside if things were not materializing in for MCED? And if you could clarify on U.K. Thank you, and I'll follow up.
Yeah, so maybe start off a little bit on the reimbursement. So right now we're really looking at a, think about it as a phased reimbursement. We already have, you know, some large self-insured employers that, you know, effectively reimburse the test for their employees. So it's both a benefit for employees, but it's also reimbursed for them. You know, we see similar things in the life insurance channel, where many of the life insurance companies are paying for the test for insured policyholders. You know, clearly in the U.K., we're seeing the potential for a pilot, but then in 2026, with broad reimbursement there, the potential for that. You know, so we're seeing across the board, there's another number of opportunities for reimbursement outside of, you know, kind of the traditional channels.
The CMS, even the CMS REACH study, you know, is effectively reimbursing those tests for us. You know, in the pathway where, you know, where the legislation is, you know, if assuming FDA approval, we'll have a pathway for, you know, the large commercial payers. But for CMS, the, you know, to be able to pay, then, you know, clearly we need the MCED legislation to go through. So there's various pieces that, you know, can get us there, you know, get us partial reimbursement beforehand. You know, obviously, CMS and the Medicare laws is an important one. If that was not to be there, then the USPSTF pathway would always be available as well, you know, although the timeframe on that is extended. Josh, you wanna add anything in there?
And then maybe Harpal, jump in on the UK, UK pricing.
Yeah, just on the US, you know, we've had numerous discussions with large national payers. As Bob mentioned, we have some progressive payer coverage, including a Medicare Advantage plan and others who are providing coverage for the test today, but that's very few. Think about this in terms of there's kind of the short game and the long game. The short game is the game we're in right now, trying to prepare the market for the eventual broad adoption, reimbursement, and FDA approval of MCED test, kind of population scale. So in the short game, Bob enumerated that we've got self-insured employers, we have some progressive payers, we have a lot of self-pay, and we have the life insurance channel and some other enterprise channels that are working for us.
But the large national payers, have told us that they're gonna really be looking for FDA approval. And, so, you know, we, we feel like we have a good line of sight to that, and we feel like we have a high probability of success. So that's what we're looking at there. So that will give the, the payers, the national payers in the US, the opportunity to provide coverage and serve the unmet need that they have in their populations. Medicare will be contingent upon the legislation, otherwise, there's the USPSTF pathway, as Bob mentioned, which is a little bit longer, but obviously accessible. And then Harpal, for the UK?
Yeah, sure. Happy to, happy to do that. So look, first of all, to say, we do have a negotiated price with the NHS. But we're not, we're not releasing the details of that, so that's a commercially sensitive conversation that, but we have—we do have an agreed price. I think what I want to say is that, excuse me, that is set at a level that would give us a healthy operating margin in the UK. And the reason that I can say that is because we know that the way the NHS would run this, and Aaron alluded to this earlier, the way the NHS would run this is through a sort of centralized national program.
So the need that we would have to invest, for example, in significant sales and marketing infrastructure, is substantially lower than we need to do in the US, and that means that even with a lower price in the UK, we're confident we can get a good, healthy operating margin.
Got it. That's helpful. And then, if I could just briefly touch on the primary care sales force. You know, the size of the sales force needed in the US, you know, if we compare it to a current peer company that has a product in CRC screening markets, those numbers are reaching into low hundreds or close to almost 1,000. That would be a significant OpEx. So maybe a question for Aaron: How are you thinking about the broader sales force expansion, I would say potentially after 2026? And potentially doing that in a more capital efficient way. What are the approaches that you're thinking about on that front? Thank you.
Maybe if I jump in a little bit first, and I'll hand it over to Aaron. You know, one of the things, if you look at, you know, Galleri relative to, you know, to what you were referencing, is just thinking about, you know, the difficulty and the level of effort required for something that's kind of in the normal process of an annual process of a physical versus something that's not. So being a blood-based, you know, blood-based test, we think it gives a significant advantage that, you know, at a time of an annual physical with full reimbursement, you would, you know, as you're getting your blood drawn, instead of your four or five tubes, you get an initial two tubes drawn.
We think that is a much, much easier sale than, you know, an asynchronous to a physical sale, as well as, you know, different, you know, different sample type. Just makes, it makes a lot more friction in the system. So we, we think the overall, you know, the kind of the term, low cost of sales, as compared to revenue, is gonna be favorable for Galleri versus that. But Aaron, anything to add?
No, I mean, that's right, Puneet. I mean, you know, you can look at all the different types of tests out there, you know, blood-based and NIPT, an NIPT test, stool-based. You can look at cholesterol testing, like it's, you know, there's not really apples to apples in any of those. But, you know, we expect to be more effective, more efficient than things that aren't blood-based, for the reasons that Bob said.
Got it. Thank you, guys.
Thank you. Our next question comes from Tejas Sawant, from Morgan Stanley.
Tejas, if you'd like to unmute yourself, please. Thanks.
Hey, guys, can you hear me okay now?
Yes.
Yep.
Perfect. So maybe to kick things off, Josh, one for you. You know, there's been a little bit of a lack of clarity on how to handle a positive screening result, specifically for MCED. Can you just share any feedback from clinicians and patients who have used Galleri so far on that aspect? And then how much of test volume to date is patient-prompted versus physician-recommended?
Great question. So, you know, the feedback we're getting thus far from all of our studies and commercial use is very positive and really relates to our ability to predict the signal origin. For the most part, in most of our studies, this has been published, you know, people get to diagnostic resolution very quickly. You know, a cancer signal is detected, and there's a prediction that's given, and that tells the doctor where to go look right away. There are some uncommon cases where something may not be seen. And we've already seen the different approaches that doctors take. They can get a free Galleri retest, they can actually wait and re-image the CSO prediction in three or six months, or they can just assume it's a false positive and retest the patient in the following year.
So we're seeing a little bit of all of it, but I think the feedback thus far has been very positive. The majority of cases that have a signal, and end up having cancer get diagnosed pretty quickly. We're not seeing a lot of these very long diagnostic odysseys that many predicted would occur. We're just not seeing that, and the feedback thus far about our accuracy has been very positive.
Got it. That's super helpful. And then one on, just the NHS Galleri study, if I may. So, you know, the study is looking at incidence of late-stage cancers as a proxy for, for stage shift, as, you know, with MCED implementation. But there's been some data recently that showed, you know, the correlation between the drop in cancer-specific mortality and the drop in cancers, like 3, 4 cancer detection, is not even across cancer types. So in light of some of this, you know, confounding data, in a sense, what are the potential caveats to that stage shift endpoint that we should be aware of?
Yeah, Harpal, you wanna take that?
Yeah. I mean, so I know the paper you're referring to, and, I mean, I would make a couple of points. First of all, you'll see some other publications coming out quite soon that actually set out that with well-run large studies, actually, the correlation between reduction in late-stage incidence and mortality is very well established with a very strong correlation. One of the things that perhaps wasn't so clear in that paper that was recently published is, that included some very small studies, which were not very well-powered, and actually, if you look more closely at the data, the larger studies that were very well-powered actually showed a very significant correlation.
So we feel very confident that a reduction in late-stage incidence does correlate very well across many cancer types. Of course, we can't prove that it's true for every cancer type, because the work hasn't been done. But we feel very confident that a reduction in late-stage incidence will correlate very well with mortality, and that therefore, it's an appropriate surrogate for, you know, for that metric that people are looking for. But I would also make one or two other quick points, if I may. One is that, you know, the balance that we have to strike here is, do we want to wait 20 years for mortality data, you know, at a time when, as Josh showed earlier and Bob showed earlier, we're losing millions of people to cancer every year.
To the extent that late-stage incident reduction is a good proxy, we could start saving lives much sooner. The second point I want to make is that, you know, with the sort of timescales we're talking about here to get to long-term mortality data, you know, the pace of technology change now, with the sorts of approaches that we're using with genomics and machine learning, means that, you know, what we can't do is wait until you know 20 years from now, by which time, the current technologies we're using may be obsolete. You know, we may have other technologies by then. So, you know, there are lots of things to balance here, the most important of which is we're losing a lot of people every day. And, you know, to try and to make ...
To bend the mortality curve sooner, actually, we have a very strong level of confidence that late-stage reduction will be a good proxy. Josh, I don't know whether you want to add anything or whether you think that covers it.
No, I think that totally covers it. Just to reiterate, you know, we have technology today that's well-validated in the intended use population, that can find cancer in people who have no idea they have cancer. Why would we withhold that for decades, from the population when cancer is about to become the number one killer of humans worldwide? That would just be unconscionable, in my view.
Got it. Appreciate the color, guys. Thank you.
Thank you. Our next question comes from Jack Meehan from Nephron.
Thank you. Good morning, guys. First, just want to go back to the past to try and understand the future a little bit. Like, if you go back and read the S-1 from September of 2020, I think the plan was to use the Pathfinder data and some other sets to submit for PMA in 2023. At the time, Pathfinder 2 wasn't part of the plan. Now you're targeting the first half of 2026. Can you just update us since then? Like, what has changed and why the pathway to FDA has shifted course?
Sure. You know, a couple things. You know, you're referencing the S-1, which was really early on. And, you know, that was, you know, prior to us launching our LDT Galleri. It was prior to the NHS study, and as you mentioned, it's prior to the Pathfinder 2 study. You know, we've had, you know, very significant discussions with the FDA, you know, multi-cancer is a new field, and people have had to determine what's the right way to, you know, understand how to approve what to approve. And, you know, we've made, you know, really substantial progress there, and, you know, that's where we've really landed on the two pivotal trials that provide the, you know, provide the evidence required between Pathfinder 2 and the NHS Galleri study.
So, you know, a fair amount has evolved, you know, from the very early days when the S-1 was filed. But we have, you know, a high degree of confidence based on our discussions with the FDA, that those are the key registrational trials. We have a lot of confidence in the, you know, Galleri, you know, Galleri product performance. So, you know, those two coupled together is what gives us a high degree of confidence. Josh or Paul, anything like to add on those?
No, just that I think the opportunity, you know, again, as Bob mentioned, this was, you know, before we even had a product in, in the market, that, that original plan. What we've decided to do internally is, is launch our PMA version on the, on the new version, which actually was an important element to extend the timeline. So this wasn't all FDA driven. This was, in many respects, an internal decision to launch our PMA using the version of the product that we could scale most effectively and most efficiently. And that, combined with, you know, after our, original S-1 submission, we decided to do the NHS study, which again, is the biggest randomized clinical trial ever conducted in the whole space, and one of the biggest screening trials ever conducted. So of course, we would want to use that for our PMA as well.
That'll be the most rigorous data ever collected in the field. So those two things really are what pushed the timeline, and that was really of our doing. But again, the early S-1 was before we even, you know, had finished Galleri, so it was quite a long time ago.
Great. And then this question, either for Josh or for Bob. You had a really interesting presentation at AACR this year around folks that came back for another Galleri test. I think there were 47,000 folks that didn't come back and another, you know, about 6,000 who did repeat testing. So my math was about 11% came back. I know this is only, like, a year of follow-up, but I was just curious, like, what you're seeing about folks returning to get a test and, you know, like, where that ratio should land over time and why?
Yeah, so I'll take a crack then, Josh, you can go specifics into the study that was presented. You know, we right now, in the pre-reimbursing space, you know, we have a number of different channels, whether it's, you know, life insurance, employers, you know, clinicians, health systems, and each of those you know, we would anticipate would have a different, you know, retention rate, return rate, for that. You know, each channel has different, you know, certainly different economics. If you look at an employer where it's, you know, where it's a paid-for test, so it's like being reimbursed versus some of the self-pay, people are gonna make a lot of different decisions on that.
So we're still in the process of really understanding how we think about, you know, that return rate as well as, you know, what the metric should be that we'll, you know, end up pushing for and driving to. So, you know, kind of early days on that, but, you know, the data, you know, Josh can go through the study of what the paper was actually trying to demonstrate.
Yeah, no, it's a great point. I think the key point is what Bob just made, is that in the short game, in the pre-reimbursement world, we don't know what we should expect in terms of the repeat testing. But we know in the long game, when this is a reimbursed, FDA-approved test, we expect the majority of people to get this test annually. And the reason we recommend annual testing is because of the rate of change and the natural history of aggressive invasive cancers. And the poster you're referring to, and the presentation at AACR, really showed that annual testing, even in people who had a negative test and then followed, were finding additional cancers. So that it really supported the annual testing paradigm, which was very positive for us.
But in the short game, you know, in the pre-reimbursement world, I think we're already well above 11%, and it's gonna be growing as we get more market adoption, market penetration and the education. But as you know, you know, we are not doing direct-to-consumer advertising in the pre-FDA approval area. So driving awareness and education is what we can do right now, and we should expect to see those numbers improve over time.
Perhaps if I can just add one quick point to that, which is to say that, you know, until now, our studies have been a single test. Our clinical studies have been a single test. What we're going to have from NHS-Galleri and from the Galleri Medicare study is sets of data where we are doing annual testing with the participants in those studies, and hopefully, you know, those studies will show the value in well-conducted clinical trials of doing annual testing. And so that also will help us to drive, you know, a much higher level of repeat testing on an annual basis in time.
Thank you all.
Thank you. Our next question comes from Dan Leonard, from UBS.
Sorry about that. I didn't recognize there was an unmute button. So I have two questions. My first question on the burn: Can you help me understand the difference between that $250 million burn forecast for the second half of 2024, you know, how that compares and how that's such a much lesser number than the $210 million cash burn that was reported for Q1 in the Form 10-Q?
Aaron, do you wanna jump in and take that?
Yeah, yeah, for sure. So I think I highlighted earlier, you know, we have a cash-based LTI program or that's supposed to mirror an equity-type program, given the hold separate. Illumina could not give us shares as part of a stock-based compensation program. So, you know, the stock-based surrogate pays out in cash on certain vesting dates. One of those dates is at the beginning of the year. There's a LTI-based cash burn in there that you'll see. I think you can find that on the cash flow statement, called out separately in the Form 10-Q for Q1.
Also, in the first part of the year, we pay bonuses and all that stuff, so the burn of the first half of the year is going to be larger than the back half.
Okay. Thank you for that. And then my follow-up question, I wanna make sure I understand what to expect from the NHS in the summer. Is there gonna be some kind of an announcement from the NHS on whether they're going to expand the pilot program or not? And is it possible you could frame the associated revenue opportunity if they were to expand the pilot? Thank you.
Harpal, you wanna take that one?
Yeah. Obviously, you know, any announcement by the NHS will be in the hands of the NHS. Our anticipation would be, if they move forward with the pilot, there will be some sort of announcement. I couldn't tell you exactly when that's going to be, but it will be, you know... I can be confident in saying it will be this summer. You know, in terms of the scale of it, and the sort of opportunity, as I touched on earlier, we're not getting into pricing. That's a sensitive pricing, but I talked earlier about the margin impact of that.
And so, you know, a lot will depend on how they view the information that's given to them, and, you know, the degree to which they want to move forward quickly, rather than waiting for the final study results. But I can't get into specific numbers here.
Okay. Thank you.
Thank you.
Yeah, Dan, just-
Our final question today. Sorry, please continue, thanks.
Sorry, just saying, Dan, just to remind you, the guidance numbers that we gave on the revenue growth was U.S. only. It didn't really include anything for NHS. So it'd be any opportunity would be upside to that.
Understood. Thank you.
Great. Thank you. Our final question for today is from Kyle Mixon, from Canaccord. Thank you, Kyle.
Hey, can you hear me?
Yes.
Yes. So yeah, first part of the question is gonna be that USPSTF. I think Bob, you mentioned that. What's the path and the process like to get into those guidelines? Would that be, like, fully MCED, or would that be maybe reliant on a single cancer kind of guidelines or recommendations that are already kind of in place and that we might see some Galleri-based test get into in the next few years or so? And then maybe for, like, Harpal, how do you think about, you know, these new versions of Galleri that you're gonna be, I guess, creating over time? Would those be just targeted methylation consistently, or would you think about integrating other analytes, like protein or other omics? You know, just curious what you thought about that.
Josh, maybe you wanna take the USPSTF question?
Sure, sure. I mean, first of all, it's really important to remember, we are not gonna pursue single cancer testing with our... It's not how our test works. It's not how our technology works. You may be referring to the fact that many other test developers have already moved away from what they said they were gonna do, which is develop MCEDs, and are now really focused on single cancer screening tests. That's not because of USPSTF. That's because the technology that they're using was not really, you know, tailor-made to look for multi-cancer early detection tests. They're just applying the technology they have, and it's better suited to single cancer. So we will not be doing that. So if the Medicare legislation is not passed, and Medicare does not have authority through legislation to cover MCED tests, then we'll use the USPSTF pathway.
We'll use it with our full data set that we're deploying for our FDA approval. The heart of it will be the NHS Galleri trial, and we would expect to do a multi-cancer detection application to the USPSTF to try to get an A or B rating and get into the guidelines that way. But we're very optimistic in the broad, bipartisan, and bicameral support and the incredible stakeholder support that the MCED legislation has, and we're hopeful we'll get that passed this year. As you remember, for those in the audience, the reason Medicare can pay for mammograms and PSA testing and colon cancer screening is because special legislation has been passed specifically to give them the authority to pay for Pap smears and mammograms. So we're following that tried-and-true pathway.
... Okay, shall I take your second question? So, I'd probably make three points in response to your question. The first is the point I already made, which is, you know, as part of our unbiased discovery effort, we did already look at whether adding other approaches, for example, you know, single nucleotide variants or copy number changes. We already looked to see whether that added performance improvement to what we were seeing with Galleri, and the answer was no. You know, and we are very happy with the way that Galleri is currently performing. You know, it's giving us a good level of performance that we feel very confident about.
So, that's the first point I'd make, that, you know, it is something that we, of course, will explore, but so far, we haven't seen anything that would add significantly. The second point I would make is that, in contrast, you know, what we will see over time is as we generate more and more data, which we are doing, you know, as we're doing hundreds of thousands of tests now. That data we can use to feed into our machine learning models to give us further performance improvement over time.
One of the advantages of being, you know, so, so far ahead in this space is that we're generating vast quantities of data that we would expect to be able to use a sort of flywheel effect, to constantly improve our algorithms and our classifications, to give us more performance from the same, if you like, the same raw material. But the third point I'd make is, you know, it's something that we'll continue to look at. And, you know, I suppose to give you one specific example, and, you know, we've talked about this in some of the data we've presented. You know, we know, for example, those of you who, who will have looked at, at Galleri's performance will know that for some urological cancers, early stage, early stage sensitivity is not as, not as high as we might hope.
That's because these early-stage urological cancers don't shed much DNA. So we've already, you know, we've already presented some data looking at urine as a possible analyte, and, you know, and that might be an area that might look promising for the future. So, yeah, I guess those would be the points I would make, but I wouldn't underestimate the importance of my second point, which is the value of generating more data that will enable us to continually improve over time.
Okay, that was great. Thanks so much, guys. And a quick follow-up for Aaron. I guess, how much of your revenue or volume is coming from primary care or PCPs versus the concierge medicine? And I saw the stat with the enterprise for self-pay revenue, I think, but maybe just expand on that topic. And then secondly, you know, back in early 2023, Illumina said GRAIL would reach breakeven five years from that point, so by 2028, maybe. Like, does that still stand? If you just talk about that a bit.
Yeah. So on the clinical sales, so we kind of group MDVIP and concierge in with, you know, the physician orders. And so the majority of our sales come from the clinic sales channel, which would include things like MDVIP and so on. But don't have it really disaggregated any more than that. You know, your second question on cash burn. You know, 2028, it's what? 4 years from now. So we've got cash for 2.5 years. We're confident that we've got enough cash to get through some of these big inflection points coming up, and we've got levers in the business that we can titrate up or down, depending on successes as they come.
So, I don't have an update on whether we think it's about 4 years from now, 5 years from now or not. But we're comfortable with the 2.5 years of cash that we have, $1 billion, that we can make that last, depending on what's coming up, for a bit.
Yeah. Great. Thanks, guys.
Thank you. This concludes the Q&A portion of the webcast. I will now turn the call back to Bob Ragusa for any closing remarks. Thank you, Bob.
Well, we'd like to thank you for joining us today. It's certainly a very exciting time for GRAIL as we advance our vision for population scale, multi-cancer, early detection, and we look forward to providing you with additional updates in the future.