Heartflow Inc Aktie

Good day, and thank you for standing by. Welcome to the HeartFlow First Quarter 2026 Earnings Call. [Operator Instructions] Please be advised that today's conference is being recorded.

I would now like to hand the conference over to your speaker today, Nick Laudico, Vice President of Investor Relations. Please go ahead.

Good afternoon, everyone, and welcome to the HeartFlow First Quarter 2026 Earnings Conference Call. Joining me today are John Farquhar, HeartFlow's President and Chief Executive Officer; and Vikram Verghese, our Chief Financial Officer.

Today, we will walk you through our Q1 performance, share updates on our commercial momentum, innovation pipeline and clinical programs and provide financial guidance. A live Q&A session will follow. The earnings release accompanying today's discussion is available on our Investor Relations website at ir.heartflow.com. During this call, we will refer to certain non-GAAP financial measures. Reconciliations to the most comparable GAAP figures can be found in today's earnings release.

I'd like to remind everyone that certain statements made on this call are forward-looking within the meaning of federal securities laws. These statements are based on management's current expectations and beliefs, involve risks and uncertainties, and actual results may differ materially.

Please note that both this live call and a digital replay will be available shortly after the call concludes. With that, I'll now turn the call over to John Farquhar, our CEO.

Thank you for joining us. HeartFlow entered 2026 with the strongest momentum in our history, and the first quarter demonstrated how quickly that momentum is translating into results. Our AI-powered platform, our deeply embedded commercial footprint, the world's largest database of annotated CCTA images, which is now over 200 million combined to create an advantage that widens with every quarter.

In the first quarter of 2026, revenue was $52.6 million, up 41% year-over-year on 67% global case growth. Four factors drove the quarter. First, FFRCT utilization remains strong across the installed base. Second, the record class of new accounts that we added in 2025 is performing very well and ramping on schedule. Third, our Plaque launch is outperforming expectations. And lastly, the broader CCTA market continued to expand, supported by guidelines and strong reimbursement.

We also reached a meaningful milestone this quarter. HeartFlow has now helped physicians manage care of more than 650,000 patients worldwide. This scale of real-world commercial experience is unmatched in our category, and we continue to extend our leadership position with each quarter.

The strength of our first quarter performance gives us confidence to raise our full year outlook. We now expect to deliver total revenue of $228 million to $232 million, representing 29% to 32% year-over-year growth. We're also increasing our plaque-specific revenue outlook to $19 million to $21 million.

Moving down the P&L. We're raising our non-GAAP gross margin guidance to approximately 81%. This is the high end of our prior range, driven by ongoing AI efficiencies, volume leverage and a higher mix of high-margin plaque revenue. At the midpoint of our revenue guidance, this implies year-over-year non-GAAP gross profit growth of approximately 37%. And finally, we remain committed to our midterm non-GAAP gross margin target of 85%.

Now turning to our 3 strategic pillars: commercial adoption, innovation and clinical evidence. I'll walk through each, starting first with commercial adoption. Within this pillar, there's 3 themes that played out in Q1. First, as expected, our FFRCT business is healthy and performed well. Second, our emerging Plaque business is accelerating ahead of expectations, and third, our platform strategy is taking hold. We're seeing it translate into deeper integration and increasing product utilization.

Now relative to FFRCT, the performance of our core FFRCT business remains remarkably strong, characterized by durable utilization across our established base of existing accounts as well as our new additions. Our cohort of legacy accounts is large and growing, delivering consistent, predictable volumes month-over-month. This is a result of our unique product differentiation as the most accurate noninvasive test for CAD and the only product with lesion-specific FFRCT values, our base business remains as strong as ever. Additionally, I'm very pleased with the performance of the accounts that we added into our installed base in 2025. This cohort of 340 accounts was the largest in HeartFlow's history and is ramping nicely in line with our expectations. Together, these factors reinforce my confidence in the long-term health and durability of our FFRCT business.

Now turning to Plaque, I'm pleased to report that adoption is ahead of our initial expectations. Activation of new Plaque accounts is ahead of schedule, and we're seeing strong early utilization trends. At a high level, the recently published ACC/AHA scientific statement on Plaque, the increasing payer coverage and the outstanding clinical and economic evidence that we shared at ACC in March, all contribute to growing awareness and demand for our HeartFlow Plaque Analysis. Closer to the front lines, our recently expanded territory account manager sales force is trained, focused and doing great work. These reps call directly on high-volume general cardiologists to raise awareness of our Plaque technology and clinical data as well as enable targeted medical education efforts.

Everything we do in this regard underscores our proven accuracy and our clinical effectiveness and is an important step in the journey of driving widespread adoption of this new technology. The scale of our medical education efforts is significant. Over the last 12 months, we've held over 1,000 medical education events focused on Plaque generating over 100,000 physician impressions. As our customers gain experience with Plaque and see firsthand the value it can deliver for their patients, we're highly confident in the early trends we've seen in Q1 and that they'll translate to even better results longer term. Given this confidence, we've raised our 2026 Plaque revenue guidance, and we now expect our Plaque installed base to reach approximately 1,200 sites by the end of the year. Finally, we see a clear path to pricing expansion beginning later this year with more meaningful upside in 2027.

Finally, a couple of comments on our HeartFlow platform, which we believe to be the most intelligent AI operating system in cardiovascular care and a very important component of our strategy. We're proud to be the only clinically validated AI platform to span the full continuum of care across coronary artery disease, from detection to diagnosis to management and treatment planning. Our platform is comprised of 4 tools: Roadmap, Plaque, FFRCT and now PCI Navigator. All powered by AI to independently provide precision and unique clinical insights and all working together as a unified solution.

Now I can't underscore this enough. Our customers are busy and operate in complex environments. In many cases, they're forced to make decisions and share information across settings of care and subspecialties. Simply stated, their job is hard. They do not want separate, discrete, ad hoc point solution. For true adoption, they demand what our platform uniquely delivers, a unified clinical solution that provides both precision and seamless coordination across the continuum of care. Further, our platform strengthens as it expands. With each new technology we introduce, the flywheel of adoption accelerates. This is happening right now with the launches of Plaque and PCI Navigator. And lastly, I'd be remiss if I didn't also mention that we integrate seamlessly into our customers' workflow, effectively becoming the operating system of record for CAD across their hospitals, clinics and health systems.

Now turning to our second pillar, innovation. In April, we launched PCI Navigator and the feedback from interventional cardiologists has exceeded our expectations. For the first time, ICs can walk into the cath lab with a fully informed procedural plan already in hand, a level of preprocedural certainty that simply did not exist before. The tool runs on any computer and reduces procedural planning to minutes. We're executing a phased rollout through 2026 with broader introduction in 2027.

Now innovation doesn't just impact the top line. Our autonomous processing initiative, which we highlighted on our last earnings call, is progressing well and is on track. We entered the pilot phase in the first quarter and continue to expect a gradual rollout through the back half of 2026 and a multiyear impact beginning in 2027. The program underpins our midterm non-GAAP gross margin target of 85%. And finally, all of our innovation is enabled by a proprietary data set, an asset unmatched in our category. As of early May, we reached the defining milestone. This data set now exceeds over 200 million annotated CCTA images. It's large, diverse and precisely annotated, and serves as the foundation for everything we do with respect to innovation. We've spent over 10 years training algorithms and it supports our leading accuracy and reproducibility. It also enables our consistent cadence of innovation as evidenced by our track record of delivering at least one major launch each year.

Clinical evidence is our third pillar, and it remains central to our leadership in this category. We've built what we believe to be the deepest and most rigorous evidence base in this space with over 625 peer-reviewed publications and more than 200 clinical studies and trials. The breadth and depth of clinical validation, spanning accuracy, clinical utility cost effectiveness and outcomes, shapes how we develop and commercialize our technologies and continues to strengthen the trust we've earned with the physicians we serve. Our leadership was on full display at the American College of Cardiology Annual Scientific session this year. Our educational dinner was standing room only. Our mobile CCTA + HeartFlow program, where physicians received a live personalized analysis was oversubscribed. We view this level of engagement as part of the broader rising tide of interest in CCTA, with HeartFlow increasingly being at the center of the discussion.

Building on our body of clinical evidence, our 5,000-patient NAVIGATE-PCI registry is now ramping with sites activating and enrollment building. This study will generate one of the first large-scale prospective data sets, evaluating how an AI-driven, preprocedural planning tool influences PCI strategy and cath lab efficiency and it will extend our impact beyond diagnosis into treatment planning. Beyond symptomatic disease, we're now advancing into the asymptomatic population, which we view as the next frontier of this category. To lead this expansion, we remain on track to initiate 2 randomized controlled trials in the second half of 2026, one for patients with prior MI or PCIs and another for patients with elevated calcium scores. The third RCT will focus on patients with prior plaque and will initiate in the first half of 2027.

These studies are designed to expand our U.S. addressable market by approximately $6 billion bringing our total addressable market to $11 billion over time. This clinical path not only builds on the depth of evidence for our platform, but it also puts us on a clear trajectory to begin accessing the incremental TAM before the end of the decade.

In closing, the first quarter of 2026 translated our platform vision into financial results. FFRCT and Plaque Analysis working together are driving sustained volume growth, deeper clinical integration and more durable account level adoption. Plaque remains early, but it's ramping ahead of our expectations. With PCI Navigator now in the market, HeartFlow spans the full continuum of coronary artery disease from diagnosis through treatment planning, positioning us as an indispensable partner to the cardiovascular service line. This performance supports our increased guidance and reinforces our confidence in the trajectory of the business.

We remain focused on our mission to transform the detection, diagnosis and management of coronary artery disease and I'm grateful to the HeartFlow team for their continued dedication to the patients we serve and for all of their hard work.

I'll now turn it over to Vikram to review our financial results and guidance.

Thanks, John, and good afternoon, everyone. Unless otherwise noted, my remarks reference the quarter ended March 31, 2026. All financial metrics I refer to other than revenue, will be non-GAAP, unless otherwise noted, and all growth rates will be year-over-year. Reconciliations to the comparable GAAP measure are in today's earnings release.

Total revenue for the first quarter was $52.6 million, up 41%. U.S. revenue grew to $48.3 million, up 42%, inclusive of $3.2 million of Plaque revenue. OUS and other revenue grew to $4.3 million. Total global revenue cases for the quarter were 67,443, representing 67% growth driven by continued strength in U.S. FFRCT, strong contribution from accounts added in 2025, better-than-expected Plaque adoption and continued expansion of the CCTA market. We continue to see strong utilization in the quarter at both existing and new accounts, consistent with historical ramp dynamics we have described previously.

New accounts continue to take about a year to ramp to near full FFRCT utilization, while existing accounts continue to demonstrate durable and consistent utilization patterns. As a reminder, FFRCT is applicable in approximately 33% of CCTAs, which means maximum FFRCT utilization in an account is approximately 1/3 of CCTA volume. We again saw particular volume strength in the clinic setting, which remains a rapidly growing and strategically important segment of the market, along with continued adoption of our volume-based rebate pricing structure.

Finally, we continue to expand our installed base at a rapid pace during the first quarter, driven by strong execution from our commercial organization and Plaque Analysis adoption that was ahead of expectations. As a reminder, we provide installed base metrics on an annual basis only.

Turning to gross margin. First quarter gross margin reached 80.5% compared to 75.3% in the first quarter of 2025. The year-over-year improvement reflects volume leverage, increased production efficiency and continued progress in AI-enabled automation, supported by ongoing training on our proprietary CCTA image database.

Operating expenses reflect disciplined investment behind our highest priority growth initiatives. First quarter SG&A expenses were $38.3 million driven by targeted investments in our commercial team to further drive adoption of the HeartFlow platform. Research and development expenses were $19.5 million as we continue to fund the innovation cadence John described, along with the clinical evidence required to support new product adoption and expand our addressable markets. Non-GAAP operating expenses were 110% of revenue versus 116% a year ago. Non-GAAP operating loss was $15.5 million compared to $15 million last year, demonstrating improving operating leverage while we continue to invest for growth.

Non-GAAP net loss was $13.3 million or a loss of $0.16 per share compared to a non-GAAP net loss of $19.2 million or a loss of $3.11 per share in the first quarter of 2025. On a GAAP basis, net loss was $27.4 million or a loss of $0.32 per share. GAAP net loss includes a $7.5 million noncash impairment charge associated with our facilities optimization and headquarters relocation to San Francisco. Weighted average basic and diluted shares outstanding were 85.6 million in the quarter.

Turning to the balance sheet. We ended the quarter with $254.9 million in cash, cash equivalents and investments. We continue to have high confidence that we are well capitalized to fund operations through profitability while continuing to invest in R&D and commercial expansion.

Turning to our updated outlook for 2026. Given the momentum in our core business and acceleration in Plaque, we are raising our expectations for the full year. We now expect total revenue of $228 million to $232 million, representing 29% to 32% growth. We're increasing our Plaque-specific revenue outlook to $19 million to $21 million. We continue to expect more material adoption in the second half of the year as clinicians gain clinical experience and broaden adoption. Based on our 1Q performance, we're also raising our full year non-GAAP gross margin guidance to approximately 81%, up 400 basis points year-over-year. The drivers of our gross margin outlook include continued volume efficiencies, increased AI-enabled automation and a higher mix of high-margin Plaque revenue. The midpoint of our revenue guidance implies approximately 37% gross profit growth in 2026.

Finally, as John mentioned, we remain committed to our midterm non-GAAP gross margin target of 85%. This reflects our confidence in continued AI-enabled production efficiencies through our autonomous processing initiatives, scaling Plaque revenue and sustained volume leverage as the platform grows. We also remain on track to achieve cash flow profitability within 3 years of our IPO.

I would now like to turn it back to John for summary closing remarks.

Thank you, Vikram, and thank you all for joining us today. We appreciate your continued interest and your support as we work to advance the HeartFlow AI platform as the new standard for care detecting, diagnosing, managing and treating coronary artery disease. We're excited about the start to 2026, and we continue to expect a milestone year.

With that, I'll turn the call over to the operator for Q&A. Operator?

[Operator Instructions] Our first question comes from the line of Robbie Marcus with JPMorgan.

This is [ Allen ] on for Robbie. Just want to start off with Plaque and the updated guide. Clearly, you began the year with a much stronger-than-expected quarter, I think, relative to our expectations, sort of came in double what we had been forecasting on a revenue perspective, and you raised the guide by roughly $4 million. So it looks like you're not pulling forward all of the upside. How much of that reflects maybe a little bit of a faster start to pull-forward to the first quarter relative to the rest of the year? And what are you seeing in second quarter that supports the rest of the guidance?

Yes, sure. So thanks for the question. I'll give you some thoughts, and I'll let Vikram sort of speak to how that plays throughout the subsequent quarters. I would say, from an adoption standpoint, the way we think about there's really 3 drivers here, okay? The first is coverage, and we talked about this coming into the year, but we've made progress since then. As of the end of March, we're now at 75% total covered lives. So that's a really good place to be, and we're happy with that. And this is, again, a function of us taking our DECIDE data which is the largest prospective data set of its kind. We're taking that to payers to open and expand coverage. And we've still got work to do. There's a fairly long tail of payers, but I feel good about the team's execution and where we're sitting from a coverage standpoint.

The second piece is we need to get Plaque into our installed base. And I mentioned in the prepared remarks, we're now upping our guide relative to installed base. I've got high confidence we'll be at about 1,200 by the end of the year. And this is a function of just, again, really strong market demand. I'm in the fortunate position where I can see the funnel moving forward. I think the funnel looks really good. So I feel good on where we're taking Plaque relative to our installed base.

And then the third piece and probably the most important is we need to get out with cardiologists, and we need to help them understand how to use plaque with their patients. Far and away, probably the #1 question I get is, "This technology is amazing. It's really interesting, but how do I actually use it with my patients." So we lean very hard into medical education. We've got a great stable of physician partners that speak on our behalf. Since last year, we've done over 1,000 events, and we continue to do that. And so I would say across all of these dimensions, everything is performing really well. I think that sort of underscores the confidence in raising our guide. Now I will say in spite of the optimism, we're still really early in the adoption cycle.

So you think about the components of that, first, you need to get contracted with the account, you need to get the workflow up and running, installed, et cetera. Then you've got to get confidence in the reimbursement. So accounts don't open the floodgates right away, they want to trust that payment is coming through. There's no denials, et cetera. And then ultimately, physicians need to use it with their patients to help manage them medically and then those patients need to come back after a period of time and hopefully show some improvement.

So we're working through that adoption cycle right now. I'm really happy with the team's execution. And I've got really -- my optimism on the future of Plaque is as strong as it's ever been. Now with that, I'll let Vikram kind of speak to kind of how this plays out throughout the rest of the guide.

Yes. Thanks, John. I'll start with comments on Q1. It was a strong quarter for us. We did north of $3 million in revenue, and that was really driven by better utilization as well as stronger new account additions. Our pricing was in line with expectations. Keep in mind, the category 1 CPT code, for Plaque went live in early Jan., which supported the sequential step-up that we saw from 4Q to Q1.

Now relative to phasing of revenue, I wouldn't characterize this as a pull-forward of any kind. It really shows sustained and strong momentum in the business. All the metrics we track point to continued adoption momentum, similar to our last guide. It's still early in the year, and this is a high conviction guide that sets us up well to outperform. And then finally, I do want to reinforce, we continue to expect Plaque to be back-half weighted, and that really reflects the adoption curve that John described.

And then I just had a quick follow-up on the spend side. Good to see gross margin ticking up to the higher end of the range supported by that strong Plaque number. But I think operating spend did come in a touch higher than we had been expecting. I think previously, we had been looking to $223 million to $224 million in OpEx for the year. Based on the first quarter run rate, it looks like you're going to be coming in a step above that. What really drove that spend in the first quarter? And should we use that as the right run rate for SG&A and R&D for the rest of the year going forward?

Yes. Thanks for the question there. The first quarter OpEx, I would say, was increased sequentially, and that really was a reflection of the full quarter impact and investments were made in the fourth quarter. We also made the conscious decision to front-load certain investments early in the year, particularly around medical education, given our expectations on Plaque and the importance of supporting that effectively through the year. Importantly, I would say, even with those investments, EBIT came in favorable to our expectations, which reflects the continued discipline across the broader cost structure.

Now to your question on 2026 OpEx. For the full year '26, we continue to expect OpEx as a percentage of revenue, which is an important metric we track to decline year-over-year. There's some puts and takes within the cost base. On the clinical front, we'll have some favorable sequencing of costs, DECIDE-related spend would phase out. We'll see that will create capacity for the investments we want to make in the asymptomatic RCT. We're also absorbing some incremental legal spend, but that's partially offset by the savings from the sublease of the facility.

So net-net, we expect OpEx to increase modestly on a sequential basis through 2026. But combined with the continued margin expansion that should drive a meaningfully narrower loss -- operating loss versus '25. All that being said, given the underlying drivers of the business, we continue to expect cash flow profitability by that mid-2028 time frame.

Our next question comes from the line of Matthew O'Brien with Piper Sandler.

This is Samantha on for Matt. Could you talk a little bit more about how the distribution of Plaque account is trending across the installed base? Did you target large centers first and now going broad, just a little bit more depth there.

Yes, sure. And thanks for the question. I think bottom line, I would say the distribution is fairly even across any cohort you want to cut it. The biggest difference at this point that I see is more tilted towards our existing accounts, being the first ones to get it and then the new accounts are just now coming online. And the dynamic that drives that, just so you know, is these were all by and large contracts that were put in front of customers last year. And so then as they move through the funnel, the first ones to go live are within our existing accounts.

Now moving forward, I would say, from Q2 and beyond, we'll see more and more new accounts come in that will have the full platform from the get-go. At the end of the day, we go where the demand is. And so I'm not reading too much into it one way or another on sort of an existing versus new standpoint. I think your question was probably a little bit more towards the size of the cohort, large, medium and small. I see that spread pretty evenly at this point, and I wouldn't expect that dynamic to change. As I mentioned, we've got good visibility into the funnel, and it looks pretty relative that way.

We still got -- as I mentioned, we've got really good confidence that there's more accounts to go add. And as we add more, as I mentioned, there'll be more new accounts coming in. But that 1,200 kind of [ IV ] call for plaque by the end of the year. That's a new one, and we've got good confidence in that. And I'd also just say relative to adding new accounts. The total hunting ground, just as a reminder, is about 3,200. That's about 300 accounts more than there were last year. So I think there's still a really good runway out there to continue to bring on to the franchise.

Our next question comes from the line of John Young with Canaccord.

Congratulations on the quarter. I just want to also touch on Plaque, if I can, on ASPs. Could you just speak -- add a little bit more additional color on Plaque ASP expectations for this year? Generally, how long are existing contracts in terms of timing for ASP increases? Have you been able to take price? And how should we think about the price point new accounts are being on board at relative to Plaque reimbursement today?

Yes. Thanks for the question, John. I'll provide some commentary, high level on how we look at pricing. Just by the way of background, many of our early Plaque contracts were really structured with attractive pricing for principally 2 reasons: first, to accelerate account activation and really drive early adoption. And I think we're doing a great job there. And then the second, because reimbursement was more limited at that time. It was really principally Medicare reimbursement. So it's also worth noting that within our contracts with our customers, we do have mechanisms for price increases to step up over time.

You may recall, in the March earnings call, we had already increased our Plaque ASP expectations for 2026. Now with the visibility we have, we continue to see upside to current consensus with really more favorable pricing becoming visible in 2027 and beyond.

Okay. Great. And then just as a follow-up, too. Our survey work has been suggesting some strong unawareness in the clinical community still about Plaque reimbursement. Do you see that as well? And are there any plans for education to overcome the hurdle?

Yes, I think you said you heard that in your survey. That doesn't surprise me. And depending on where you are in the market, I think that's fair that, that could be the case. I think in general, I would describe awareness of Plaque as low. I mean, that's part of this kind of category creation effort under which we're undertaking. A large part of that is clinical education, as I mentioned. But another part of that is the reimbursement that comes with it. I mentioned this dynamic, customers need to understand the category 1. They need to understand the payment dynamic in their ZIP code. They need to trust that the payment is going to come through after a submission and there's not going to be a denial, what have you. All of that is part and parcel of the work we do to help to develop the category. So my expectation is that was probably the case earlier this quarter. It will probably still be the case in future quarters, just to a lesser degree as we continue to do our work.

Our next question comes from the line of Rick Wise with Stifel.

John, congratulations on the really another outstanding, if not brilliant quarter. And in that spirit, you've given us -- you beat street numbers, you beat our numbers and you raised by more than beat. So I feel slightly embarrassed but not so embarrassed, I'm not going to ask. When I think back and look back over the last couple of years, you've been growing 40%-ish kind of character or better. The midpoint of the guide, the new guide is I think I'm right in saying sort of 30% or so. Maybe just help us think through in a little more detail, I'll call it your conservatism in light of, not just the past performance, but really impressive continued rollout of innovation, as you said, adding new accounts, expanding reimbursement, Plaque moving faster. Maybe just help us put that in context. It's clear that the -- even looking into '27, the outlook is extremely positive. And maybe just talk about how we should think about it? And maybe where the drivers of upside could come from?

Yes. Sure, Rick. And thanks for the question. So I think you're probably familiar with our guidance philosophy at this point. But when we give guidance, we want it to be a high-conviction guidance, okay? And I would certainly characterize this one within the same light. I've got really high confidence in this raise. When I think about sort of the biggest drivers of it. Number one is our FFRCT business continues to be strong and durable and predictable. We talked about in 2025, we had the biggest year ever in terms of bringing on new accounts. We brought on 340 new accounts. That -- those accounts are really starting to pay off well, okay?

They are ramping on schedule, and we've got -- I've got really high confidence that they'll get up to sort of that higher utilization range that we historically see. So I feel good in that regard. Then you mentioned Plaque. Plaque is this sort of next layer of the cake growth, so to speak. Every metric we track internally is ahead of plan, okay? It doesn't mean we're where we want to be, but I'm very, very pleased with it. And that's both relative to volumes as well as new accounts coming through. And that underscores our -- the guide that Vikram mentioned that underscores that 1,200 account installed base by the end of the year. We're still really early, but I feel better now than I ever have relative to Plaque. And I should also say qualitatively when I'm out in the market with customers, the feedback is outstanding. It's just working through this kind of adoption cycle that we've discussed, and that's hard to rush, okay.

And then the last piece is just the wind at the sales of the overall CCTA category. I think we like to say we're on the right side of history here, as this structural shift from older modalities for noninvasive testing towards a CT first pathway. We think that's a multiyear tailwind and that's certainly a component as well. So from that standpoint, I've got good confidence in this guide. And relative to the guiding philosophy, Vikram, I don't know if you have anything to add on that.

Yes. Thanks, John. I'll just reinforce from my perspective, again, to what John said. This is a high-conviction starting point that really preserves room for operational upside as the year unfolds. And relative to the FFRCT business, we meaningfully outperformed our expectations on both volumes and revenues in 1Q based on that performance and the visibility we have. We have meaningfully increased our volume and growth assumptions versus the March guide. We also have a clear path here to further strengthen the growth outlook as the year progresses.

Now second on Plaque, I would characterize this as a derisked guide based on the early trends we're seeing in Q1. It is still early in the fiscal year, and we continue to believe Plaque revenues will be back-half weighted. John talked about the conviction behind the 1,200 Plaque accounts by the year-end. And that's really supported by the funnel and the high visibility we have on the commercial front. And finally, we also did outperform expectations on new site additions, which we believe is a strong indicator of continued interest in the HeartFlow pathway.

In sum, I would characterize this guide as a durable base case with really multiple paths to drive performance above the top end of the range.

Yes, it's exciting to hear. And just another topic I'm hoping you're going to expand on a little bit is what I -- sort of my reaction was that tsunami of annotated CCTA images, you're accumulating and rapidly building, now over 200 million. Back in January, I think you were talking about over 160 million. And I don't know whether this number makes any sense, but it feels like you could be at, I don't know, 300 million by the end of this year or early next year. Regardless of that number, can you talk through the implications for HeartFlow, the multifaceted implications from this really rapid accumulation of these images relative to the speed of the platform, to the adoption, to the implications for your COGS, your margins. Maybe just any updated thinking and color and perspective there.

Yes, sure. Thanks. Good question. So yes, so we've passed the 200 million annotated image threshold. So we're now just a bit north of that. And that's really a function of the model that we've put in place. So for every CCTA that one of our customers does, that comes into our cloud, we annotate it. It's got precise annotations, and we can use that to drive innovation, both that drives customer-facing innovation with physicians and new clinical insights and gross margin expansion. So we absolutely see this as a competitive advantage that is getting stronger each and every day. And it really is the bedrock for all we do relative to innovation.

We've talked about, historically, we've got a very strong track record of putting at least one material new launch into the market every year. We've got plaque tracking coming out next year. Obviously, this year, we have a PCI Navigator. Obviously, we've got more in the Roadmap beyond that, that we'll share with you on future calls. So I feel great with this as an asset, so to speak.

More broadly, we try to innovate across 3 vectors, okay? We go deep, and this is all about making our existing products better. So just last year, when we launched our next-generation Plaque algorithm, that's a good example of that. That came from that same data set of annotated CCTA images. We innovate around breadth. We expand the platform, so we can do more. A couple of years ago, we launched the Roadmap. This year, we're launching PCI Navigator. Next year, we're launching serial plaques. So that's another example, real-world example of that. And then we use the same data set for efficiency. A lot of that efficiency is expanding our gross margins. We talked about our autonomous processing initiative that's now underway. And we continue to use that same data to do that.

So this is core to who we are as a company. It's core to our engine that helps drive growth. And I think as we've discussed, it's only getting stronger as each quarter goes on.

Our next question comes from the line of Larry Biegelsen with Wells Fargo.

This is Nathan Treybeck on for Larry. Can you talk about how accounts with Plaque Analysis are utilizing it? Is it broad across the addressable patient population? Or is it in specific patient cohorts?

Yes, sure. Thanks, Nathan. Good question. I think it's still early to tell. I think anecdotally, we have a direction, but we don't know objectively at this point. I think most of our early users are using it in patients with more disease first. I think that's not surprising to hear that feedback. I think as they get confidence that folks -- that patients that they have with a higher plaque burden come back with better adherence, better outcomes and LDL lowering, et cetera, I think that will broaden over time.

My only comment on that is I put an asterisk on it, that's an anecdotal kind of pulse of the market that's not based on any data. I'm sure there's instances where patients with lower plaque burden are using it or getting it as well.

And then in terms of how Plaque utilization is trending, as we think about new accounts, how is it trending in the new accounts relative to how FFRCT trended on utilization when it was launched?

Well, I think good question, and thanks for asking it. I think we're kind of talking about 2 different parts of history. I mean, when FFRCT was launched, that was pre guidelines, that was an older base of CT capital that was in the market. The awareness was quite a bit lower. This is a new world. I think CT plus AI is much higher awareness right now. So I don't know if we can kind of compare the utilization to the same degree. I will say we're very confident in FFRCT's ability to and our understanding of ramping over time, FFRCT will go to the full extent of the range. We have not seen that yet in Plaque. We're too early in the adoption curve. I think physicians need to use Plaque, have patients come back, understand how it's helping them manage it. So we're not at that point yet.

Our next question comes from the line of Kallum Titchmarsh with Morgan Stanley.

I'll be unoriginal and stick with Plaque. You described the upcoming DECIDE 1-year outcomes as a potential inflection point for Plaque. Can you maybe just give us a sense of what specific endpoints or metrics you expect will resonate most with payers and physicians? And are you still on track for the readout in H2?

Yes, sure. So we use DECIDE, obviously, in a lot of what we do relative to educating the market. We use that with both payers to get the coverage as well as more importantly, with physicians to help them understand. The 90-day readout showed a 51% change in management, meaning when a physician uses -- have a Plaque Analysis, they change their management plan with patients and a 19-point reduction in LDL, okay? So the 1-year data that comes out later this year, we'll see each of those. Obviously, I don't know what the -- what that would be and we'll have about 13,000 patients in that readout. So we certainly expect that will help with Plaque adoption. But I think the most important thing that's going to help with Plaque adoption is this first round of patients that get it are going to come back to see their cardiologists? And those outcomes will become very personal. And I think as that cardiologist and physician experiences that, that's when the flywheel, I think, will just get stronger.

Great. And now that the PCI Navigator has been in the market for about a month, again, I realize it's early days. But any early observations on adoption patterns? How many sites have activated it? Have you seen any early sign of incremental FFRCT pull-through from the kind of interventional cardiologists? And I guess, if not now, then when are you expecting that to take place?

Yes, sure. So Navigator, we're super excited about it. And the feedback is really strong. So there's lots and lots of interventional cardiologists asking for it, asking how they can get it, et cetera. We are going to be -- like everything we do, we're going to be very planful in the rollout. What's most important is the cardiologists have a good experience with it. And we believe the ability to have a plan for a PCI before the patient -- before the physician is in the cath lab, is really important, okay? And this is going to allow physicians to plan ahead. They can plan both relative to the complexity of the lesion as well as the approach that they're going to take. They'll be able to hopefully select devices more intelligently. They can walk in with a clear procedural plan.

I've had an interventional cardiologist tell me recently, the cath lab should be for executing the plan, not deciding the plan. And that's exactly what happens with TAVR and mitral and left atrial appendage, and we think one day PCI will go there. So we're really excited about that. We think it makes our platform more stickier, more compelling. We're -- right now, we're targeting the largest PCI hospitals first. That's where -- given our strategy, that's where we're targeting. We'll roll through that, and we'll make sure that all initial experiences are strong. At this point, we're not disclosing any metrics more specifically for that, but I will say I'm very excited about Navigator and what it does for our platform this year.

[Operator Instructions] Our next question comes from the line of Brandon Vazquez with William Blair.

Congrats on a nice quarter here. John, maybe I'll start with a higher-level picture. You used this phrase that there was a flywheel effect going on as you start to see adoption of kind of all the different solutions within the portfolio here at HeartFlow. Maybe expand on that a little bit. What do you mean by you're seeing a flywheel effect where -- what metrics give you excitement and optimism to come on the call and start talking about flywheel effects and how does that build over time? What metrics do you think we'll see that in the model?

Yes, sure. So thanks, Brandon, for the question. So I think I link it back to 2 things: one, the power of the platform and then two, the power of the database that I mentioned earlier on the call. Relative to the platform, we believe the stronger the HeartFlow platform, the stronger the business. And we back everything we do with clinical data from detection to diagnosis to management and treatment. So the more we expand our platform, and that's core to our strategy, the more we see and we believe utilization and adoption will occur. So this year, introducing Plaque, introducing Navigator is obviously a big year for the platform.

Next year, we'll have a Plaque tracker out there. And as I mentioned, we've got other ones down the road that we'll share. So that's -- that provides stickiness that provides utility. Everything we do across the platform is co-registered, you get concordance across all of it. So wherever you are in the continuum of taking care of a patient with coronary artery disease, we have our clinically backed AI technology to help you with that. Now relative to the other data set that I mentioned that gets bigger and bigger as well, right? And so for every patient we serve, for every account we open, we're adding more into our proprietary database. The more we have in our database, the more diverse it is, and the more we'll be able to develop new innovation along the lines of what I previously spoke about.

Got it. And as a follow-up, maybe I'll ask 2 here, one, John, for you and one for Vikram. And part of this flywheel, John, is it fair to say do you guys see in accounts that are adopting the entire portfolio, are they seeing higher utilization levels across the board? I think that's part of what I'm trying to understand a little bit, too, is there a flywheel or a halo effect of the entire portfolio in the accounts? And then Vikram, maybe talk to us a little bit about -- I know you guys don't give specifics on ASPs and volumes every quarter, but talk a little bit as you think about the increased guidance relative to what you had in the prior guidance at how much of that is coming from volume versus better-than-expected pricing?

Short answer right now, Brandon, it is too soon to tell. But we anticipate over time when successful, certainly, that could be the case. I think the more components of the platform, the more value we're delivering. And so with that, you would expect stronger utilization. But at this point, it's too soon to tell.

And Brandon, to your question, it's really driven by volumes. We expect on the Plaque side, ASPs to be in line with consensus in 2026. On FFRCT, as we have talked about previously, we do have the dynamic of -- 2 dynamics specifically, and both of them are volume-enabling dynamics. First is our use of volume-based rebates. This was to our advantage in Q1. Our volume growth in Q1 approached 50% and as volumes come in ahead of plan, the ASPs obviously reflect that level of outperformance. And then second is the mix shift to the clinic side of service, which carries with it lower ASPs, but is, again, fundamentally volume enabling. Against that backdrop of ASP shifts, we continue to drive towards profitable growth. Gross profit in Q1 increased about 10 points faster than revenue to about 51% and was, in fact, higher than unit growth as well. With respect to the guide, we assume these dynamics will continue through the balance of the year on FFRCT.

And I'm currently showing no further questions at this time. Thank you all for your participation. This does conclude today's conference. You may now disconnect.

Good day, everyone, and welcome to HeartFlow Fourth Quarter 2025 Earnings Conference Call. [Operator Instructions] Please note, this conference is being recorded. I would now like to turn the conference over to the Vice President of Investor Relations, Nic Laudico. Please proceed.

Good afternoon, everyone, and welcome to the HeartFlow Fourth Quarter 2025 Earnings Conference Call. Joining me today are John Farquhar, Heartflow's President and Chief Executive Officer; and Vikram Verghese, our Chief Financial Officer. Today, we will walk you through our Q4 and 2025 performance, share updates on our commercial momentum, innovation pipeline and clinical programs and provide financial guidance. A live Q&A session will follow. The earnings release accompanying today's discussion is available on our Investor Relations website at ir.heartflow.com.

During this call, we will refer to certain non-GAAP financial measures. Reconciliations to the most comparable GAAP figures can be found in today's earnings release. I'd like to remind everyone that certain statements made on this call are forward-looking within the meaning of federal securities laws. These statements are based on management's current expectations and beliefs, involve risks and uncertainties, and actual results may differ materially. Please note that both this live call and a digital replay will be available shortly after the call concludes. With that, I will now turn the call over to John Farquhar, CEO.

Thanks, Nic, and good afternoon, everyone, and thank you for joining us. We're pleased to host our fourth quarter call and close out a record year for HeartFlow. We've achieved outstanding financial performance driven by the sustained adoption of our HeartFlow platform and strong execution across our commercial, innovation and clinical initiatives. Total fourth quarter revenue was $49.1 million, representing more than a 40% year-over-year growth. Global cases grew nearly 53% year-over-year, driven by record installed base growth, consistent FFRCT utilization and strong CCTA market growth. We also achieved a record non-GAAP gross margin of nearly 80%, reaching our long-term target just over 7 months from our IPO.

Based on this momentum and our confidence entering the year, we're initiating full year 2026 revenue guidance in the range of $218 million to $222 million. This represents a year-over-year growth of approximately 24% to 26%. Embedded in this outlook is our expectation that full year 2026 plaque revenue reaches approximately $15 million to $17 million, reflecting continued progress across reimbursement, installed base expansion and utilization.

From a gross margin perspective, we're initiating full year 2026 non-GAAP gross margin guidance of 80% to 81%, representing approximately 300 to 400 basis points of expansion year-over-year, driven by continued volume leverage, AI-driven efficiencies and the increased contribution of higher-margin plaque revenue. We're also increasing our midterm non-GAAP gross margin target to 85%, up from our prior target of 80%. This represents sustained volume leverage, increasing plaque revenue contribution and continued gains in AI-driven efficiencies.

Our financial targets reflect the strength of the underlying business, a solid foundation for continued growth and a high degree of confidence in our ability to execute consistently. Now turning to our strategic pillars. As we look ahead, we remain focused on our 3 strategic pillars: Driving commercial adoption; advancing our plaque innovation pipeline; and extending our clinical leadership.

Starting with driving commercial adoption. 2025 was a record year of installed base expansion. We added 340 new accounts, closing the year with 1,465 U.S. accounts, the strongest new account growth in the company's history. FFRCT utilization remains strong and durable. New accounts ramp to steady state within the first year and maintain those order patterns consistently over time.

Turning to plaque. We drove strong early momentum in plaque account additions through 2025, closing the year with an installed base of 489 accounts. Feedback from customers has been very positive. They increasingly recognize HeartFlow plaque analysis as the most accurate AI-driven plaque assessment technology available. It remains the only AI-powered plaque solution supported by prospective published clinical evidence demonstrating 95% agreement with the invasive gold standard of IVUS. The reimbursement landscape has also materially strengthened.

As of January 1, 2026, the Category 1 CPT code for plaque is officially in effect, formally assigning RVUs and enabling physician reimbursement for the first time ever. And now with Aetna recently joining UnitedHealthcare, Cigna and Humana in covering our analysis, Plaque now has coverage across approximately 75% of U.S. covered lives. These payer wins are a testament to the clinical power of our DECIDE registry. As the largest prospective study of its kind with over 22,000 patients, DECIDE demonstrated that the heart flow plaque analysis changed physician management plans 51% of the time compared to CCTA alone.

With respect to plaque adoption trends, we're encouraged by the strong initial plaque volumes so far in the first quarter. We continue to expect plaque revenue will become more meaningful in the second half of the year as newly activated sites scale and physicians build clinical experience. Our conviction is high that plaque will be a strong contributor to both the top line growth and margin expansion this year and beyond.

Our second pillar is the continued advancement of our innovation pipeline. We continue to leverage our proprietary database of 160 million annotated CT images for innovation. As you'll remember, in late 2025, we launched our next-generation plaque algorithm and evolved an already market-leading product into one that's even more precise. Now in 2026, our proprietary database is again unlocking innovation, this time with the launch of PCI Navigator, the first and only AI-driven planning tool to integrate anatomy, plaque burden and lesion-specific physiology in a single interface.

PCI Navigator extends our AI platform deeper into the interventional suite as interventional cardiologists will be able to plan complex interventions with unprecedented confidence. They can anticipate and plan for lesion complexity, selecting devices more intelligently, thereby enabling them to enter the CAD lab with a clear procedural strategy. This is highly analogous to the preprocedural planning that's already standard for TAVR, mitral and left atrial appendage. We anticipate PCI Navigator will not only strengthen engagement with interventional cardiologists, but deepen provider support for the heart flow pathway and further increase the stickiness of our AI platform across health systems.

We're also pleased to announce that our product development team has executed ahead of schedule. We're pulling forward the PCI Navigator launch to April of 2026. This is earlier than the original plan that we had communicated for the second half of this year. In addition to customer-facing innovation, we're also applying AI in our proprietary database further down the P&L to expand our gross margins. And we're excited to announce a major AI-driven efficiency initiative, HeartFlow autonomous processing.

Built on over a decade of proprietary algorithm training, autonomous processing transitions our case processing into a highly automated single-step verification model. With autonomous processing in place, our algorithms will autonomously manage intake, analysis and output with the final human-in-the-loop quality check to preserve our gold standard accuracy. This initiative underwrites our confidence in our new midterm non-GAAP gross margin target of 85% and it's another proof point of how we convert our database and AI into long-term financial scale. Importantly, the rollout will follow a deliberate phased approach.

We'll start the initial rollout later this year, followed by a multiyear expansion beginning in 2027. Our third pillar is clinical leadership, and we continue to make great progress in this area. The goal of our strategy has always been the same, to create a new standard of care. In order to accomplish this, high-quality clinical evidence is paramount.

In 2026, we'll accelerate our evidence generation through prospective trials and peer-reviewed publications. Earlier this week, we announced the first patient enrolled in our NavIigAT- PCI registry. This prospective 5,000-patient study is designed to evaluate how PCI Navigator influences clinical strategy, enhances procedural efficiency and bolsters physician confidence in the CAD lab.

To our knowledge, this is the industry's first prospective registry that evaluates an AI planning tool for PCI. Now turning to plaque. At the American College of Cardiology meeting later this month, we're presenting real-world data from a 15,000 patient registry for Mass General Brigham. This real-world evidence demonstrates that the heart flow plaque analysis is the most powerful CT-based predictor of MACE.

Importantly, it also solidifies our plaque staging system as the most clinically validated system for personalized risk stratification, enabling clinicians to identify high-risk individuals beyond conventional metrics. And finally, in the second half of the year, we plan to report 1-year outcomes from approximately 13,000 patients within our landmark DECIDE registry. We believe that the addition of longitudinal outcomes data to DECIDE will shift the conversation from clinical utility to clinical impact and meaningfully expand the evidence base for plaque-guided care.

Now moving to our expansion into the high-risk asymptomatic population, which we announced earlier this year. From the beginning, we've been methodical in our journey of bringing our technology to more patients, starting in the $5 billion symptomatic patient population first with FFRCT and then moving to plaque. We view the high-risk asymptomatic population as the next logical step in this journey. We estimate this market represents an incremental $6 billion opportunity in the U.S. alone, expanding our total market opportunity to $11 billion.

Over the next 12 months, we plan to initiate 3 randomized controlled trials across targeted high-risk asymptomatic subpopulations, patients with prior heart attack or PCI, patients with coronary calcium and patients with prior symptoms and documented plaque. Targeting these subpopulations represents a derisked approach that prioritizes a diagnostic pathway, focusing on patients already in the health care system who have known disease.

We're excited about these TAM expansion efforts, and we'll share more details as the year progresses. It's also worth noting that from a capital allocation standpoint, this is a very efficient approach. We believe we can achieve our clinical objectives with approximately 1,400 total patients combined across all 3 trials. In closing, we've executed extremely well against our core strategic pillars in 2025. And in 2026, we expect this execution to continue with meaningful catalysts on each pillar on the horizon.

Our confidence in our 2026 guidance is high. It's supported by continued growth in our core FFRCT business, a meaningful second half ramp in plaque strong installed base expansion and a rapidly growing CCTA market. We look forward to advancing our mission and transforming the standard of care in CAD. I'll now turn the call over to Vikram for his financial review.

Thanks, John, and good afternoon, everyone. Unless otherwise noted, my remarks reference the quarter ended December 31, 2025. All financial metrics I refer to other than revenue will be non-GAAP, unless otherwise noted, and all growth rates will be year-over-year. Reconciliations to the comparable GAAP measure are in today's earnings release.

Total revenue for the fourth quarter was $49.1 million, up 40%. U.S. revenue grew to $44.8 million, up 41% OUS and other revenue grew to $4.3 million. Total global revenue cases for the quarter were 57,776, representing 53% growth, driven by continued strength in our U.S. FFRCT business. We also expanded our installed base at a record pace during the year and the fourth quarter, bringing our year-end total to 1,465 accounts. This reflects strong execution by our commercial organization and continued growth in interest around Plaque Analysis.

As a reminder, we provide installed base metrics on an annual basis only. We continue to see strong utilization at both existing and new accounts, in line with historical trends. New accounts continue to ramp to steady-state FFRCT utilization in about a year, while existing accounts consistently maintain that utilization. We again saw particular volume strength in the clinic setting, a rapidly growing segment of the market as well as continued adoption of our volume-based rebate pricing structure. Turning to gross margin.

Fourth quarter gross margin reached nearly 80% compared to 75.3% in the fourth quarter of 2024. The year-over-year margin improvement reflects better-than-expected volume leverage and an increase in AI-driven efficiencies enabled by continuous training on the company's proprietary CCTA database. Operating expenses reflect disciplined growth investments. Fourth quarter SG&A expenses were $34.6 million, driven by investments in headcount and the expansion of our TAM sales force to further drive adoption of the HeartFlow platform.

Research and development expenses were $17.1 million, driven by investments in technology to advance our innovation pipeline and in clinical research to expand our evidence base. Operating expenses were 105% of revenue versus 114% a year ago. Operating loss was $12.5 million compared to $13.5 million last year, demonstrating improving operating leverage while we continue to invest for growth. Non-GAAP net loss was $9.8 million or a loss of $0.12 per share compared to non-GAAP net loss of $18.6 million or a loss of $3.15 per share in the fourth quarter of 2024.

On a GAAP basis, net loss was $24.4 million or a loss of $0.29 per share. The GAAP result includes a $9.3 million noncash charge from the remeasurement of our common stock warrant liability driven by higher share price during the quarter. In October of 2025, the warrant holder net exercised all warrants, so Q4 will be the last quarter with any warrant remeasurement impact. Weighted average basic and diluted shares outstanding were 84.8 million in the quarter.

Turning to the balance sheet. We ended the quarter with $280.2 million in cash, cash equivalents and investments. We continue to have high confidence we are well capitalized to fund operations through profitability while continuing to invest in R&D and commercial expansion.

Looking ahead to 2026, I'm pleased to provide our guidance for the full year. We are initiating total revenue guidance for the full year 2026 to be in a range of $218 million to $222 million, representing approximately 24% to 26% growth. This outlook includes Plaque revenues of approximately $15 million to $17 million weighted towards the back half as clinicians build clinical experience.

We are also initiating non-GAAP gross margin guidance of 80% to 81%, representing 300 to 400 basis points of expansion. Drivers of our gross margin guide include volume efficiencies, increased AI-driven efficiencies and high-margin plaque revenues in the second half of the year. The midpoints of our revenue and gross margin guidance imply approximately 31% gross profit growth in 2026.

As John mentioned, we are also increasing our midterm non-GAAP gross margin target from 80% to 85%. This reflects high confidence in further AI-driven efficiencies driven by our autonomous processing initiative, scaling plaque revenues and continued volume leverage. We also remain on track to achieve cash flow profitability within 3 years of our IPO. I would now like to turn it back to John for summary closing remarks.

Thank you, Vikram, and thank you all for joining us today. We appreciate your continued interest and your support as we work to advance the HeartFlow AI platform as the new standard of care for detecting, diagnosing, managing and treating coronary artery disease. We're excited about the progress we've made in a record 2025, and we look forward to a milestone year in 2026. With that, I'll turn the call over to the operator for Q&A. Operator?

[Operator Instructions] Our first question comes from Robbie Marcus with JPM.

Congrats on a good closing quarter. Two questions from me. I wanted to start first with 2026 guidance came in above the Street, and I believe that Plaque number is also above where consensus sits. So I'd love to just hear the confidence in the building blocks, particularly any extra commentary you have on Plaque so far in first quarter and feedback on the launch and the assumptions underpinning FFRCT? And then I have a follow-up.

Yes, sure. Thanks, Robbie. Great to hear your voice. So yes, so thinking about '26, I mean, I'll start by saying I've never been more confident in this business and what we're setting up to deliver here. I think there is fundamental demand out there, and I could not be happier with our team's track record of execution. And I want to compliment the team for continuing to deliver. The way I would kind of categorize this guide is I'd put it as a healthy starting point. And we've talked about our guidance philosophy previously.

We want to provide very high conviction guides that set us up well for quarters down the road, okay? So that's a high level. First, around FFRCT and the base business, I've got great confidence here, okay? And there's really kind of 3 drivers underwriting it. We have proven our ability to go out into the market and sign up new accounts, okay? We've got -- we just did the biggest year ever. We did 340 last year, as you're aware of. And I think our go-to-market model has proven our ability to go get it, okay?

And that number incidentally grows. The number of accounts we can go get grows by about 300 every year, okay? So one, we can go get the account. Two, once they're in, we've got very predictable performance relative to utilization. We see physicians ramping and using our FFRCT technology very predictably. So they become healthy in that regard. And then lastly, there's this kind of underlying factor out there relative to CCTA in and of itself.

And that -- I'd like to think of that, that genie is not going to get put back in the bottle. CCTA as a frontline test for diagnosing coronary artery disease is in the guidelines, not just in the U.S. but around the world. And I think we're on the right side of history here, and that's going to continue to sort of fuel growth moving forward. So that's the core business. And then what makes '26 really exciting is plaque. And I think we did a great job last year. We signed up close to 500 accounts by the end of '25. we're well into Q1 here right now, and we've had some strong activations that have continued. I have visibility, obviously, into the funnel.

So I've got high confidence that by the end of this year, we'll be in 1,000 accounts. And again, just as a frame of reference, it took us 8 years to get into 1,000 accounts with our FFRCT business. We'll do it in less than 2 with flat. And then the accounts that are live now, the early trends are very positive relative to volume, okay?

So I'm very happy about that. Now I will say, I don't think this is going to be a light switch moment. I still think, as Vikram mentioned, the material volume is going to come in the back half. We've got some catalysts out there. We've got the 1-year DECIDE data reading out in the second half of this year. Physicians really need to experience it. experience outcomes with their individual patients to get into a more kind of higher utilization rate. But all that being said, very confident. And again, I'd say this is a high conviction forecast that we're giving you.

Yes. I'll probably add some commentary to that. Again, this is a high confidence baseline that really sets a solid foundation to allow for quarterly progression. Additionally, there's pockets of incremental pockets of potential upside. I'll start with FFRCT. Again, a high conviction guide. We've not factored in any utilization tailwinds from the Navigator launch. So that will be upside. The feedback from the physician community has been incredibly positive.

Now focusing on plaque, and you had a very specific question there, Robbie. I'll point out that, again, this is highly derisked given the early volume data that we're seeing in Q1. If the adoption curve were to steepen more quickly, it provides an additional catalyst for growth. Second, we referenced the 1,000 accounts by the end of the year. They've got strong funnel with good visibility.

So any outperformance there would also be beyond the range we provided for guidance. So stepping back, the framework here is straightforward. We've been conservative around optionality and realistic around execution assumptions. So the overall SKU in our view is to the upside.

Great. Then just as a follow-up, a lot of software companies have come under pressure from fears around AI that they could do it better and faster. And obviously, there's a lot of moats you have with your data and tech and med tech are very different sectors for a reason. I'd love to get just on record, your view of your moat around the business versus some of the AI companies that are hitting software and how you feel your defenses and things you're doing to stay ahead of all the competition. Appreciate it.

Yes. Sure, Robbie. And I think that question makes a lot of sense given some of the recent headlines. I'll say, I mean, I would characterize our moat as highly defensible, okay? And I think there's a couple of components that sort of to make it up here. The first is, I think as everybody is aware, an AI model is only as good as the training set and the data that, that model is trained on. And we have, and we're very proud of it, what we believe to be the world's largest proprietary CT database of annotated CT images, okay?

We've got over 160 million images. This data is effectively ground truth data, and it does not exist in the public domain. So there's no large data set out there that somebody very quickly could go acquire and start training a model on, okay? We've been training this model, and we've been building this data set for over 10 years. So not only is it very large at 160 million annotated CT images, it's also super diverse. because it's got all sorts of different types of anatomy and types of CT capital feeding into it, okay?

The last thing I'll say on it is right now, we're already connected, and this is sort of at the end of -- or at the start of this year, we're connected to 60% of the market's CCTAs. So every day, as the category grows, we're growing disproportionately with it, okay? So this is -- this would be a very difficult catch-up game if somebody were to try to start a race here. So that's the data that the data -- that the algorithms are trained on, but that works hand-in-hand with clinical evidence. And we like to say you can't code your way into clinical trust. clinical trust with a physician comes by delivering high-quality clinical data.

And HeartFlow as a company has been investing in this regard for over 10 years. We have over 600 peer-reviewed publications. We've got a couple of randomized controlled trials. This is high-quality clinical data, and you can't fast forward that. There's no way to rush that. And so this is an incredible lead that we have, and we'll continue to invest and build on that.

Now that clinical data works hand-in-hand with the fact that we're med tech. And in med tech, you can't have an AI hallucination the way you can in consumer technology. We are a regulated medical device. You need clinical data first to get FDA clearance, but then you need to operate within a regulated quality system. And we have just that. And of course, we have our human in the loop that ensures our accuracy remains incredibly high. And then the last thing I'll say is we are software.

And software is useless if it's not integrated into a doctor's everyday practice. And we are effectively the proven operating system, if you will, in over 15 hospitals and clinics right now, okay? So we work really hard to integrate in and meet our physicians where they are, so we help we help do their job easier. We don't ask them to adapt and do it with us. That can be a tough road to sort of hope initially, it takes over a year to get into some of these hospitals. But once you're in, we're pretty sticky, okay? And being the kind of operating system of choice, we think is a great advantage for us. And then lastly, underscoring all this, we've got a great strong global patent portfolio that we think is very defensible as well.

We have a question from the line of William Plovanic with Canaccord Genuity.

Just I'd like to ask just on the product itself. I mean you guys are innovating rather quickly. And so you've had significant changes to both FFRCT and plaque, I think, over the last 3 to 6 months. Just I was wondering if you could just kind of maybe highlight some of those changes and how they've impacted the workflow to help us understand. And then my follow-up is on the PCI Navigator. Just initial feedback, future data, product innovations. We saw the first patient go into the registry. Just love to hear more on that. It sounds like you're using that as one of your competitive moats.

Yes, sure. So good to hear your voice, Bill. So on kind of the product innovation piece, last year, we introduced our next-generation plaque algorithm, okay? And that was, again, we're going into this proprietary database that we have. We're training the algorithms to get smarter. And we took what was already a really good plaque algorithm, and we enhanced it, okay? From a customer standpoint, there's no changes to the workflow per se. We're always doing behind the scenes work that I don't think is really -- it's not material enough for a call such as this to take out steps and improve functionality of our product.

Two years ago, a little over 2 years ago, we launched our new -- another example is we launched our new user interface, which incorporated both Plaque and FFRCT co-registered with our risk profile or nomogram information into a singular user interface and now physicians can use that to interrogate the entire coronary tree. This year, and we're really excited, and we're launching it actually next month. The team did a good job pulling it forward. We're launching our PCI Navigator, and we're really excited about that. And we've talked about the strength of the platform matters and PCI Navigator expands our platform, okay?

And we really think for interventional cardiologists, this can be a pretty attractive tool for their toolkit. And what allows them to do is really plan ahead before a PCI procedure, so they can walk into that procedure knowing the right the devices to select, the complexity of the anatomy, et cetera, that's very analogous to what they're already doing with TAVR and mitral, et cetera. So we're excited to that. We have, and Vikram can speak to this if there's interest.

We haven't necessarily baked in FFRCT upside as a result of this, but we do think it's possible. We think having an interventional cardiologist who's an important part of the CV ecosystem, if you will, really advocate strongly for CT and heart flow pathway that helps us. and certainly using HeartFlow on every case that comes in helps us as well. So we're super excited about that. Vikram, I don't know if you want to talk about that.

Yes, I'll quickly underscore our current forecast really does not assume any incremental upside from PCI Navigator. To the extent Navigator drives incremental activity, it will really flow through FFRCT volumes as each case is anchored to an FFRCT order. And then we're highly encouraged by the early feedback we've received on the product and the early potential we see, but we would frame PCI Navigator as a longer duration growth vector for us than a real near-term driver of financial performance.

And the last thing sorry, go ahead, Bill.

I was just going to say, from my understanding, you're not charging for PCI Navigator. And I just wanted to circle up with just on the consensus on Robbie's question, Q1 consensus, I think it's $46.9 million. That's up 26% year-over-year. You're guiding 24% to 26%. How do we think about that first quarter? I mean, you're 2 weeks away from closing it. Are you extremely comfortable with that current number? Do you think it will be higher? Or how should we think of Q1?

Yes. In terms of phasing, generally, Bill, what I would say is our full year guide reflects a midpoint of 25% year-over-year. For the first quarter, and this is specific to your comment there, we expect growth in excess of 30% year-over-year, and that bakes in about a sequential improvement of 1% to 2% full cognizance of where we are relative to the end of the quarter.

Moving into Q2 for what it's worth, we typically benefit from a little bit of seasonality, and we expect that pattern to hold true in 2026 as well. And then as Plaque starts to really materialize in '26 towards the back half of the year, you'll see more sequential growth relative to where consensus is at. And again, Plaque, should Plaque outperform, that's the principal upside to the back half phasing assumptions.

Bill, just to round out your question on innovation a little bit. Obviously, Navigator is what we're discussing today. I would describe our pipeline of new products is really strong, and we'll have more news in future quarters of other innovations that are coming.

One moment for our next question please. It comes from Matthew O'Brien with Piper Sandler.

Just for starters on the plaque side, should we continue to assume $350 per case for plaque here in '26? And is there any upward mobility to that revenue per case going forward?

Yes. Maybe I'll say a couple of comments on plaque and then Vikram, you can speak to the specifics of the model. I mean, first thing, Matt, I'll say, I mean, I've said this before, but it's worth saying again, I could not be more excited about what plaque is going to mean full stop. I think it's going to mean a lot for patients, most importantly, but I also think it's going to transform this business. Now over time, we've got to be a little judicious in how we call it, given we're still in the early innings. And as we monitor uptake every quarter, we'll be able to adjust going forward. Part of that uptake, to your question is pricing, and I'll let Vikram speak to that.

Yes, happy to, John. So Matt, relative to your question, given that we have now commercial payers coming online, Aetna being the latest one that gets us to about 75% covered lives in the U.S. We now have clearer line of sight to pricing improvements over time. Our contracts with our customers have built in mechanisms that enable better pricing with broader coverage. And so consequently, in our 2026 base plan, we have underwritten modest ASP upside in the '26 plan and then more meaningful step-ups in future years.

Okay. So a little above $350 to start and then maybe going up more in '27, '28. Is that fair?

That's right.

Got it. And then, Vikram, just sticking with the numbers here a little bit. I wanted to get into gross margin because that was so good, but trying to keep it to 2 here. Was the FFRCT case number per site similar to what you saw in Q3 because given the number of hospitals you added, that's really good. And then I think it's kind of to the guidance question. If you assume -- well, when I'm looking at the FFRCT revenue that's factored in when you think about the plaque growth plus OUS, it's the lowest level that we've seen from you guys or would see from you guys over the last 3 years.

Is there something in there that we should be aware of, again, given all the momentum that we're seeing in the core FFRCT business? Or is it just back to kind of what John was saying to start with like, look, we're just -- we're trying to be very judicious with how we're guiding early in the year, and then we'll kind of reflect back as things progress.

Yes. Thanks for the question, Matt, again. Relative to cases per site, we did outperform despite the heavy onboard number. We were about 15% higher on a cases per site basis in 4Q '25 relative to where consensus came at. And that's really driven by the strength of that market shift that's happening here with CCPA ultimately on its way to becoming standard of care. With respect to guidance, I'd say we've -- it's part law of large numbers as well as a part guidance philosophy. With -- we're lapping an exceptionally strong year in 2025. And when you start with that kind of elevated base the growth rates really start to normalize.

And then the second piece to underscore what John said, this is core to our guidance philosophy. The framework we put forward for 2026 still assumes very strong underlying demand from a nominal volume perspective with a fair amount of conservatism. So I would really view this guide as an early year framework that intentionally leaves sufficient room for quarterly progression.

We have a question from Larry Biegelsen with Wells Fargo.

This is [ Nathan Treybeck ] on for Larry. Can you talk about the utilization that you're seeing for plaque analysis in your FFRCT accounts? How does the utilization compare to FFRCT? And how do you envision utilization trending throughout '26?

Yes. Thanks, Nathan. So again, we're in the early innings, but I think what we're seeing so far is positive, okay? Now you need to remember the total applicability for plaque is 60% of all patients, okay? We're nowhere near 60% right now because we've just kind of got out the gates here. So over time, there's plenty of upside to ramp. But what we've seen is very positive and I think leads to sort of the bullishness that you're hearing from me today. Relative to FFRCT, FFRCT obviously is a more mature therapy once -- or sorry, more mature diagnostic. Once an account is up and running, we see FFRCT being utilized pretty close to the full range.

So that's a good news, but the full range is about half of what it is on plaque. So FFRCT is about 30% or 33% full utilization, and we get pretty close to that with most of our accounts. So I think over time, there's certainly an opportunity to see plaque really in a material way take off here. I think long term, I've got really high confidence plaque is going to be a bigger business than FFRCT. I think where we have to sort of wait and see and kind of measure it every week, every month is what does that ramp look like. And we're in the early innings there, but again, so far, pretty positive.

Okay. Can you talk about how you're helping your accounts to better understand how to utilize plaque analysis? And I guess, what's next for plaque from a clinical data standpoint?

Yes. Well, we lean very heavily into education. And I think as we've talked about before, one important unlock is coverage, and we're making great progress there. That's super easy to measure, so to speak. Medical education is a little more nuanced, okay? In the FFRCT journey, it was a much simpler story. You just needed to prove accuracy versus invasive FFR and then everybody knew what to do with whatever, 0.7 FFR value.

Plaque is much more nuanced, and we are leaning in very heavily to help physicians understand how to use plaque with their specific patients, okay? And we do a lot of medical education. I would say moving forward, and we're at AHA just this next week in New Orleans. And far and away, the lion's share of everything we're talking about there will be plaque. In the second half of this year, the big data that we'll share is the 1-year data on Decide, where we'll have 1-year outcomes there, and that will be another piece that's really important to bring to our physicians.

Ultimately, what we're trying to do is shift it from clinical utility. Question number one is, first, how do you use it to actual clinical management. And that's the journey that we're on with our physicians right now.

One moment for the next question. It comes from the line of Rick Wise with Stifel.

This is John on for Rick today. A couple of quick questions for me. First, I just wanted to ask about the broader role of CCTA for heart flow and how that's impacting growth today. I remember back in the IPO, you gave a couple of metrics about how penetrated and utilized CCTA is for noninvasive coronary test. I was just hoping for a general update on where CCTA adoption stands. How much are you benefiting from it? And how much more do we have to go in terms of uptake there in your view?

Yes, sure. Thanks, John. Appreciate the question. Relative -- so again, we're trying to create a new standard of care here. And it's well documented that the existing standard of care is suboptimal. Patients are being misdiagnosed, way too many false positives, way too many false negatives, okay? CT plus HeartFlow addresses that issue, and we're trying to create a new one. The first piece of that is CCTA being adopted.

At the end of 2021, CCTA in the U.S. came in as a Level 1A test ahead of all other alternatives. okay? We like to say we're on the right side of history here, but we have a lot of upside. It's around 10% to 12%, I think, penetrated relative to the whole standard of care. So there's lots of ways to go before we get there. But as I said earlier, I don't think the genie is going to get put back in the bottle there. I think it's just a matter of time between the guidelines, the stronger reimbursement, more and more readers are raising their hand, becoming CCTA readers.

Plaque is coming on board. So that makes the utility even that much greater. All of those trends, in my mind, are durable, and there's a high ceiling for us to aspire to go hit. Now from what I'm seeing in the data, I'm not seeing anything slowing us down. We're seeing more and more accounts join the category, so to speak. At the start of this year, there's 3,200 accounts with active CCTA programs. We expect that number will be 3,500 before the year is out. Right now, the majority of volume coming off the CT scanner is not CCTA. It's for another modality. So more and more accounts can start adding slots to it before they even have to tap into kind of a longer uptick of a capital cycle. So there's lots of tailwinds into this story, and we continue to lean into it.

That's helpful. And I just wanted to also ask about competition quickly here. I'm just curious what you're seeing in terms of win rates or when you're going head-to-head against peers. Are you getting any pushback on price if they come below you? I'm just curious about how the broader heart flow ecosystem factors into the competitive positioning when you go up against others?

Yes, sure. So first thing I'll say, and I remind the team of this often, our competition is the standard of care, okay? When we wake up every morning, we are trying to convert volume from the existing standard of care into CT plus HeartFlow. And when we do that, we will be successful and patients will be better off, and we're continuing to focus on changing those practice patterns and bringing physicians into kind of a CT plus HeartFlow pathway. Now your question, I think, is more specifically around other AI vendors. This is not a new dynamic. Some of these vendors have been around for -- since 2018, 2017.

None of that is slowing us down. I think if anything, having more competitors or players in the category is actually good for the category. We don't compete on price. We compete on the quality of our data, which we're extremely proud of. We're prospective. We're published. We lean into that. We prove our accuracy, and we compete on the quality of our product. And we continue to improve our product just like we're doing this year and expanding our platform with PCI Navigator. We don't publish win rates and things like that, but I think our results are very positive, and I feel good about the way the team is competing and how we're winning in the market.

And with that, we conclude our Q&A session and conference for today. We want to thank everyone for participating. You may now disconnect.

All right. Thanks for being here, everyone. Great to see you. Glad to have the HeartFlow team with us today. Maybe quickly before I jump in, you got to grab your research disclosures from our website at morganstanley.com/researchdisclosures. If you have any questions, please reach out to your Morgan Stanley sales representative.

So thanks for being here. Really excited to have HeartFlow with us and dig into another health care AI story.

Maybe, John, just to start, if you could just sort of hit where you guys sit in the market, what your application is, what the need is and sort of what your value prop is?

Yes, sure. So thanks for having me. Pleasure to be here. So we're a technology company. We use AI. We deploy AI to better diagnose coronary artery disease. So at our core, that's what we do. I think there's a tech conference, but you're probably all aware of heart disease in and of itself is the world's #1 killer. So we think we're deploying our technology to a very serious problem. It kills more people than all forms of cancer combined.

Every 40 seconds, somebody in this country is having a heart attack, half of all those heart attacks are a total surprise, meaning they've been through -- the patient has been through the health care system, and it hasn't appropriately been diagnosed. So that's really the problem that we're trying to solve, and we're in the very early innings, but we believe we can create a new standard of care using our technology to do that.

And everything we do is backed with clinical data. Our clinical data proves our accuracy. It proves our outcomes. We're very proud of that. What's unique in our AI and the insights that we deliver to our physicians is the use of our technology is reimbursed.

Well, first off, it's all cleared by the FDA. But it's reimbursed. So it's covered by Medicare, it's covered by commercial payers. So our physician customers and their hospitals, they make a margin by using our technology. And similarly, we have separate revenue streams for those reimbursed technologies as well. So we went public in August of -- just this last August and super excited for what the future holds with our technology.

Maybe we can spend a second sort of double-clicking on the noninvasive testing market for cardiology. What does that look like? maybe spend a second on CCTA, which is the underlying imaging modality and whatnot.

Yes, sure. And I should mention our technology sits on top of a coronary CT image, okay? So that's what we deploy our AI to. But...

You don't own that...

We don't own that. We deploy that into our cloud, and we put our technology on top of that. Now relative to the -- as you referred to it, the noninvasive testing market, if you take a step back, our current patient population that we're treating with our technology are patients with symptoms, okay? So that's -- you're mowing the lawn on the weekend, you're running to catch a flight, you're chasing your grandkids, what have you, you have some type of a symptom. That could be chest pain, shortness of breath, dizziness, fatigue.

Ultimately, that patient finds their way in most cases, to a cardiologist. And that cardiologists job is not to run them straight into the hospital and do an invasive procedure, is to diagnose those symptoms noninvasively, okay? So that's the market that we're sort of playing in currently.

In total, there's 9.5 million noninvasive tests every year in this country. If you monetize that against our technology, that's about a $5 billion TAM, okay? Now our penetration within that is just like I mentioned, the really early innings, we're about a 2% penetration against that total noninvasive testing market. But again, we believe we're on the right side of history here, and we can create a new standard of care.

You asked about the coronary CTA. So the coronary CTA, again, is what we -- that's the frontline test that patients in some cases, but in an increasingly amount of patients are getting ordered for this noninvasive test. That's about 10% penetrated, and then we're a fraction of that.

Maybe just spend a second actually, what are the other noninvasive types of tests?

Yes. So these are stress tests, treadmill tests, sometimes you call them nuclear tests. And the vast majority of the market still relies on those. And we really believe our technology is better and can improve on the inaccuracies. Just a note on those, and this is well understood in the clinical data. 55% of the time, they diagnose with a false positive, meaning you'll go and you'll get a nuclear test and be sent into the cath lab when you don't need to be there for an intervention.

More concerning, 30% of the time, they miss it altogether. And that's when these patients go home and unfortunately, have have bad events at home. Again, we have clinical data that proves we're much more accurate than that, and we can avoid those false positives and those false negatives.

So patient ends up with the cardiologist, gets a CCTA, image goes in the cloud to HeartFlow. What's the platform and the products and solutions that come off of that then at that point?

Yes, good. So first of all, our -- we have a technology platform. We're not a singular test, okay? And our platform, and we're committed to this, but our platform continues to expand year-on-year, okay? Right now, we have 4 components to it.

The first sits at the front end of the workflow. And I should say wherever that patient is in their journey, whether or not it's detecting it, diagnosing it, managing it or treating it, our goal is we bring clinically validated, reimbursed insights to our clinicians, so they can take better care of their patients, okay? The first part of our platform, we call it the road map analysis. This is a workflow tool. We have clinical data that substantiate -- it allows CT readers. So these are the physicians that are reading that CT image to read it faster with less variability, okay?

And we provide that free of charge for every CT that's sent to HeartFlow, okay? And it's a really nice component of the model, and I can speak more as to sort of how it enables some other things in a moment.

Next, sort of in between the bookends here, we have 2 reimbursed technologies. Each serve a separate insight for clinicians and each are reimbursed separately. So there's separate revenue streams for us. The first is our plaque analysis. plaque analysis measures plaque, whether it's calcified, noncalcified or low attenuation plaque down to the cubic millimeter. And so that's orders of magnitude more precise than any physician could do with the human eye.

And again, we back this with clinical data, so we know it's accurate against the invasive gold standard, okay? Our next reimbursed piece of the platform is our FFRCT analysis. What FFRCT does is it models the blood flow across any disease that's found and tells the physician whether or not the blood flow has been impeded to the point where that lesion or that disease spot of the coronary tree needs to be intervened on, okay? So separate insight than plaque and again, separate reimbursement for the physician.

And then lastly, which we're launching here shortly this year, we have what's called a navigator tool, a PCI navigator tool. So again, if you follow that patient all the way through the continuum and they end up needing to be in the cath lab with a disease that needs to be treated, we now have a tool where the interventional cardiologists can plan accordingly to take care of that patient. So that's a platform as it stands now. But again, we fully intend to continue to add components to it.

So maybe a couple of other questions on the platform. One of the things I don't think we've talked about yet is one of the unique, I think, advantages, although maybe it needs a little explanation is you've got a human in the loop still on your side. So not just the physician treating the patient, but on your side, you've got a human in the loop. Maybe just talk about how that works, John, why that is or isn't necessary? Could that human go down to 0 over time? Can it not, et cetera?

Yes. So first thing I'll say, again, we're in health care. This is regulated. This is FDA-cleared technology. This is technology that needs to operate within a regulated environment. As part of our FDA clearance, we are required to have a human in the loop, okay? That human serves a really important component. They ensure quality of the algorithm, okay? So the first step that occurs when we ingest these CT images from hospitals and clinics, it comes to our cloud.

And the first thing the algorithms do is they create a precise 3D model of that actual patient's coronary tree, okay? If that model is perfect, then all of our algorithms can run autonomously off of it, okay? But because CT images come and sometimes they have artifacts in them, sometimes they might be blurry based on a patient's heart rate while they're getting the scan, when they show up, they're not always pristine. That's where our human in the loop, we call them a quality control analyst make a couple of adjustments all the way, okay?

And the analogy I sometimes use is if you have a Waymo taxi driving around San Francisco, if you give it a perfect map of the city, they won't run the stop sign, they won't drive up or they won't drive up on the sidewalk, what have you. Similar here, what these humans in the loops is they're making that city map perfect and pristine.

From there, the algorithms take over, and that's how our technologies are sort of brought to fruition. Now if we were having this conversation, you asked kind of how long does it take to do. If we were having this conversation 15 years ago when the company was founded, that quality control analyst probably spent 20 hours doing those connections, okay? Last year, at the end of last year, they spent about 20 minutes, okay? So every...

Has that fundamentally have been a technological improvement?

Yes, exactly. It's 100% what it's been. So every unit of time that's taken out is a function of our algorithm getting smarter over time as we feed more data. So we've taken a lot of time out over the years, and you can look at our margin expansion over the same period of time and see that occurring. And I've got super high confidence that, that expansion is going to continue in the future.

To get to your last question, will there always -- can we always take a human out altogether? We never say never. I think where I've got really good line of sight on is we're going to take that time down and down and down. If you were to take the algorithm or the human out altogether, that's a new regulatory filing. And we'd have to evaluate whether or not that's a strong business case when it's appropriate to.

So you -- maybe just spend a second on the algorithm. You've talked about -- you've got 150 million-plus annotated images. You talked about the algorithm improvement over time. There's a lot of talk at the conference and the investor community right now on, how proprietary or not is the data, how proprietary or not is an algorithm or piece of software. Maybe just talk us a little bit about the training and building of the algorithm, the testing that's went into it, the proprietary nature or not of the data you have, et cetera.

Yes. So good question. So first thing I'll say is, right now, we've got, give or take, call it, 116 million annotated CT images, okay? So these are clean CT images that have been, for lack of a better word, kind of scrubbed by our analysts over the course of time. This is a data set that's incredibly diverse. We've been growing it for 15 years.

So as different anatomies come through the cloud, as different types of CT capital comes through, we've got a very diverse data set there. This data set, to the best of our knowledge, is the largest that exists, and it is proprietary to us.

And like I said, we've grown it over time. And as we speak, it's growing, okay? Every time we ingest a CT, we grow it. So that's one really important kind of component of the business model. And we use that data to create new algorithms to expand gross margins. We use that same data to create new technology that we launch back to our existing customers through those same data pipe and connections. So that's one really important piece that I think serves as a compelling moat, so to speak, as we move forward.

The other is around -- as I mentioned, this is a regulated space, okay? You need FDA clearance. You need to know how to approach the FDA in order to get that clearance. You need to know how to produce products at scale and deliver them back to physician customers underneath a quality control system that's auditable by the FDA. You need complaint handling, you need design controls. No different if you were an implantable device. So that's a skill set that we've built, and I think we're pretty good at over time.

There's another kind of data that's, I would argue, equally as important as the data we use to train the algorithm, and that's our clinical data, okay? At the end of the day, our customers are physicians. okay? Physicians, by and large, make decisions on how is this going to help me treat my patient better. The way you impact practice patterns and change behaviors. So remember, we're trying to shift behavior from this traditional standard of care to this new one is through strong clinical data. And the company has invested over time. We were founded in 2010 and credit to the team in 2010 for making the commitment to do this.

We have over 600 peer-reviewed clinical publications in the market right now. That's orders -- and that's taken us 15 years to accumulate. That's orders of magnitude more than any other -- anybody else has got. And we're continuing to invest in that. And then I think the last piece is we work really closely with customers. And again, we've learned this over the years on how do you take this technology and integrate it into a health systems ecosystem, okay? So these health systems, there's a lot of different stakeholders.

You could be an interventional cardiologist, you could be a CCTA reader, you could be a general cardiologist, you could be an internist or a primary care physician, okay? And after we kind of go live in an account, we work to integrate across that entire ecosystem.

Now that can be challenging in a lot of respects, and it's taken us a long time on how to do it well. But where we are now, we really think it's another kind of compelling piece of our moat because once you're in, the switching costs are pretty high, and we see a lot of really kind of sticky behavior. And then the last piece, I should say, everything we do is backed by really strong IP, and we've got a good IP portfolio as well.

You mentioned the sort of bidirectional data pipes. I think that's actually an exceedingly rare asset or capability across health care. I talked to Tempus earlier today, they've got that. But if you go talk to a very large percentage of our clients, they don't really have that in any form or fashion. Maybe they're passing a PDF to their client or something, but there's not a true data relationship. Can you talk a little bit about what that looks like for you, what it's been like to build that, where you're at in building those connections and where you need to go sort of in the future?

Yes. Okay. So first thing, just for kind of context here. At the start of this year in the U.S., we were connected to about 1,500 accounts roughly, okay? And those are hospitals or those are clinics, okay? Anyone that has a CTA within the building.

Once we're into those accounts, that's where this bidirectional data pipe comes in. We're in about 1,500. That number again is growing just about every day as we're adding more. The total hunting ground is around 3,200 accounts. And that 3,200 is growing by about 300 a year, okay? And I just say that to say there's a lot more pipes for us to kind of connect. And if there's 300 new ones entering the category every year, that's going to keep us plenty busy.

And this is just U.S.?

Yes. This is just U.S. So it's by virtue of that, that we set up these data pipes. Now once that occurs, the minute one of these patients that I mentioned, the patients coming in for a noninvasive test for -- to diagnose coronary artery disease. The minute that patient has a CCTA, it comes to our cloud automatically.

Once it's in our cloud, we digest the data, we create the analysis and we bring that analysis back. So by the time the physician is ready to make a decision or make a diagnosis, our technology is there waiting for them, okay? The benefits that the customers yield or glean from this model is they get instant access to our technology the minute they need it.

They get access to our road map technology for every one of their CCTAs that's irrespective of whether or not they need any of our other tests. And they get integration into their workflow, okay? So those are benefits for them.

From our standpoint, there's a number of benefits as well. We get to consume that data. It builds into our database. So that 160 million annotated CT images that's growing and growing and growing. We mine that data to create new algorithms. Those new algorithms help us expand our gross margins. Those new algorithms help us create new technologies. So as we launch new components of our platform over time, we have that data.

And then because we have those data pipes, as we launch that technology, we do it in a very frictionless way, okay? It literally is a new button on their already existing user interface. So physicians like that because they get it quickly and easily. And obviously, we like that because it's a very efficient way to bring new technology to market.

And I would assume there's at least some form or fashion of a moat there because you went through the Infosec reviews and all the things at the hospitals.

Yes. So in order to get -- great point. We've got really high confidence that we've got a good model where we can go get the next new account. And we've done that to the tune of 1,500 accounts currently. That process, however, you need to go through -- first, you need to build clinical champions.

But then beyond that, you've got to bring in administration, you've got to bring in IT, you got to bring in information security. In some cases now, you've got to kind of navigate a sort of an AI board at the health system. That process to land that next new account can take a lot boards every year, okay? So again, a little bit of a barrier going in. But once you're in, it makes for a pretty sticky customer on the other end.

Maybe we could double-click on one of the things you mentioned, which is plaque. That's a newer sort of offering, at least from a commercial perspective for you. Maybe talk a little bit more about what that is how that launch is going, sort of what you expect out of that? How important is that going to be over the next couple of years?

Yes, perfect. So just for reference for the audience, the vast majority of our current revenue is all coming off our initial test, that's our FFRCT test. And that's a durable business that we believe is growing very nicely with lots of upside relative to penetration.

The next kind of layer of cake, so to speak, of growth is our Plaque Analysis. And this is an analysis that's FDA cleared, and it's paid for, both through Medicare and a large growing number of commercial payers. And we started launching this last year. And at the start of this year, we're in close to 500 accounts just at the start of early January. We think we'll be north of 1,000 by the end of this year, okay? So we're really excited about.

Is that largely your existing account base you're going and penetrating?

The higher the majority -- the majority of that is existing, but there's some news as well. Yes, good question. So we're super excited about kind of the market interest on plaque. And plaque from a patient applicability standpoint, has doubled the applicability of FFRCT. So more patients can benefit from this reimbursed technology, okay? Now we're still in the early innings of that. So there's a lot of kind of the initial phases of this is you got to sign a contract, you got to take them live.

Now we're in this -- help them ramp and use it. Now there's a lot of questions out there on exactly what is the right way to use this analysis. That's probably the #1 question that I get that the team gets. So we're leaning into quite a bit of medical education to help physicians understand how to use it.

Now that being said, based on all the early indicators, I'm extremely excited about what 2026 has for plaque. And I think by the second half of this year, we'll see some pretty material impact by virtue of all the early demand we've seen.

You mentioned how much it expands the patient population. Could you spend another second on who is plaque applicable for? What's the exact?

Yes. So it's covered. The applicable patient population is really driven by the coverage that the physician will get paid for it. Where FFRCT is relevant to about 1/3 of all patients based on their disease burden, plaque is about 2/3. So much bigger. And so you could go get a heart flow analysis, not need an FFRCT, but very likely you're going to need a plaque.

And I should say with plaque, we're measuring disease down to the cubic millimeter. And a lot of people probably have had calcium scores. That's usually a question I get. Calcified plaque is one type of plaque, okay? There's noncalcified plaque. There's low attenuation plaque. Physicians really need to understand the full profile of what your composition of your plaque burden is in order to better treat you. And that's what we aim to do with our analysis.

Right. Much higher fidelity set of information around...

Yes, you can't get it with the human eyeball, and it's not appropriate. There's no physician out there that's going to send you into the cath lab for an invasive procedure just to understand your plaque burden that wouldn't happen.

Which is effectively the gold standard. Invasive is the gold standard. Right. Okay. So we talked about sort of plaque, how that could impact the near term. It sounds like you've got a lot of excitement there. As you maybe try to look out the next several years, what is it that excites you guys? What are you sort of focused on? What are the growth and value drivers as you play forward here?

Yes. So I think in the whatever you want to call it, near midterm, super healthy kind of base business. Our core FFRCT business has tons of runway ahead of it. And that's the beauty of creating a new category is the more we're shifting towards CT plus HeartFlow, the more runway there is.

So there's great opportunity with that in and of itself. Plaque coming online is super compelling and the top focus of the teams right now. But we still have this database, right? And we have a very long track record of using that database to introduce new technologies that we then can launch through the same bidirectional data pipes as we call them.

In 2026, the next element that we're launching is our Navigator. This is our PCI Navigator tool. This, again, is intended for downstream usage. This is interventional cardiologists, and they can use our platform to help plan for a PCI procedure, okay? We really like this for a couple of reasons.

One, that interventional cardiologist can be a great champion for HeartFlow within the health system. So if they use our technology on all of their patients, they're going to help champion a CT-first pathway across their health system. So that's a compelling piece.

The other piece is we believe in the power of the platform. The stronger the platform, the stronger the platform. And we know we can deliver value to an important stakeholder through this. So that's our '26 kind of platform expansion. In '27, we have a what we call a serial plaque analysis planned. And so that's -- you can take -- where you can track disease progression or hopefully regression from Scan A against Scan B. So if you think about this longitudinally, 2026 is really the year of patients getting their first Plaque Analysis, okay? You don't need a plaque analysis every year, okay?

But then as you go into '27, and you're working with your cardiologist and you want to understand whether or not the therapeutics that you've been on are actually impacting your disease burden, what do you need? You need your second scan, okay? So we're going to launch a serial plaque technology that will co-register against the first and allow physicians to see exactly how the disease burden is being changed, hopefully in a positive way relative to the therapeutics that, that patient has been on. So that takes us through 2027.

Beyond that, we have other technologies planned, but not yet for disclosed discussions. We're very excited about opening up a new addressable market, okay? So everything we've been doing up until now, has been symptomatic patients, patients with chest pain, that's a $6 -- excuse me, $5 billion total addressable market.

The nextdoor neighbor to these patients are high-risk asymptomatic patients. okay? And then beyond that, you have sort of at-risk and low risk. We're going to enter the high-risk asymptomatic market next. That's another $6 billion TAM, okay? We have high confidence that our existing technology doesn't discriminate whether or not you're having symptoms or not having symptoms.

What we need is we need clinical data in order to open up that market. And so we have randomized controlled trials that are planned this year against 3 subpopulations. One is patients with high calcium scores. The other patients are patients with prior heart attack. And then the third is patients with prior plaque.

And we'll have separate clinical investments against each of those subpopulations to open up those that high-risk asymptomatic market. All of that, we believe, is going to hit us or help us before the end of this decade.

What do you -- what do those trials fundamentally look at? What do you have to demonstrate to get FDA clearance or to get adoption from the market?

The 2 primary endpoints are going to be plaque regression. So again, how much does plaque change over time? And then second, LDL. It's change in biomarkers. We'll capture -- sorry interrupt. We'll capture longer-term hard outcomes, but we don't believe that's needed to open it up.

And just maybe one last question on that. You mentioned now a couple of times sort of plaque regression or comparative plaque analysis effectively. Does that over time start to get you more integrated with the pharmaceutical companies? Like is there a deeper partnership there in the way that they're delivering drugs and you're measuring success?

I think the opportunity is absolutely there. And what we are able to do is take a picture to tell whether or not the drugs are working. We'll stop. So the goal of these drugs is to prevent the disease from growing. And again, there's a lot of analogies here between what we do and how cancer has evolved over time.

But if you find a tumor, you treat it. and you take another picture of the tumor and you make sure it's getting smaller or it's not spreading. That paradigm doesn't yet exist in coronary artery disease. But we believe with our technology, we can get there.

Great. Awesome. Well, thanks again for coming, John. Really appreciate it. Exciting year ahead for the HeartFlow team.

Yes, great. Thanks for having me.

Hello, everyone. We're going to get started here. I hope you had a good lunch. I'm Robbie Marcus, the med tech analyst at JPMorgan. Very happy to introduce CEO of HeartFlow, John Farquhar. John will give a presentation followed by some Q&A. John?

Great. Thank you, Robbie. I appreciate the time today. Great to be here. Good to see everybody. So I'll jump into this. I'll give everybody an overview of HeartFlow, and then I think we've got some time at the end for Q&A.

So just to set the stage here, and these are all themes that you'll hear as I go through the presentation, but we are a software company. We use AI in everything we do. We deploy that software and that AI to the treatment diagnosis and management of coronary artery disease.

The company was founded in 2010, just down the road here from Stanford University. We had our first FDA clearance in 2014. And I've been the CEO for about 4.5 years now. So great to be here.

We currently target the symptomatic market. So these are patients that have some form of symptom for coronary artery disease, we calculate that at a $5 billion TAM.

Very pleased to announce we had tremendous progress in 2025. In the U.S., we now have an installed base of 1,465 accounts. We define an account as any hospital or clinic that has a coronary CCTA image or coronary CCTA piece of capital in it. And I'll talk through in a minute what our model looks like.

But this was the biggest fiscal year we've ever had relative to adding new accounts. It was also the best Q4 we've ever had. In total, for the year, we added 340. So it was a record for the team, really pleased with the execution there.

As I mentioned, we've been at this for a while. And from the very early days, we've had a commitment to generating clinical evidence. We have over 600 peer-reviewed publications. And this really is the tip of the spear in everything we do to help change physician behavior.

We've also, over the same time period, acquired what we believe to be the world's largest proprietary database of coronary CCTA images. We have over 160 million annotated coronary CCTA images. I'll talk about the benefits to that to us and how that helps the business.

Great financial performance. These are numbers through Q3, but we're starting to get good scale trailing 12 months of $162 million, 41% growth. Very pleased with that, but equally pleased with our margin expansion.

We're not giving Q4 results this week at JPMorgan. We wanted to keep the focus more on the strategy. But what I can tell you definitively, we had a very strong Q4, there's no bad surprises we beat consensus. I feel good on that.

And we're also right on the doorstep of this next wave of growth. So we've got this attractive base, durable business, but what is coming now is the next wave. And that next wave is coming through our Plaque analysis.

I'm very pleased to share. We are as of January 1 or actually New Year's Eve in 489 accounts live with plaque. So that's tremendous. And I'll talk about what the pipeline looks like in the future, but I'm confident by the end of 2026, we'll be in over 1,000 active accounts with plaque. And just for frame of reference, it took HeartFlow originally 8 years to get in our first 1,000 accounts, and we'll do it with Plaque in less than 2. So we're excited on that.

And then lastly, we're going to initiate some new RCTs that expand into the asymptomatic market. We're going to do that in 2026, and we're going to be into these markets before the end of the decade. So lots of great things going on for the company and most importantly, the patients that we serve.

Obviously, we're at JPMorgan's Healthcare Conference. I won't belabor the problem, but heart disease is the world's #1 killer, more than all cancers combined, okay? So it's a huge issue. Every 40 seconds, somebody is having a heart attack. And obviously, the cost to societies is staggering.

The root cause of this is really the current standard of care is inaccurate. And the reason it's inaccurate is because it does not measure the actual disease itself, it only measures surrogates, okay?

Because of this inaccuracy, the outcomes are as bad as they are half of all heart attacks occur with no prior diagnosis, okay? 30% time when a patient goes to a physician and tries to get diagnosed for having chest pain or what have you. They're sent home told they're okay. When in actuality, they're not, they need treatment. And 55% of the time, they're sent downstream into the cath lab when they don't need to be there. And obviously, there's costs and risks and waste associated with all that.

So we addressed this problem. And I think the paradigm shift that's occurring here in coronary artery disease with AI is very analogous to the shift that's already occurred in some forms of cancer. If you think about legacy standard of care for breast cancer or lung cancer, it started with sort of remedial measurements. In the case of breast cancer, it's a physical examination, lung cancer, chest X-ray, then over time, advanced imaging came online. Earlier diagnosis occurred, outcomes got better, and then more recently, all of that's being augmented with AI.

Okay, that's a similar arc that we believe is occurring in coronary artery disease, and that's happening with advanced imaging in the form of coronary CCTA and our AI platform.

This pathway is endorsed around the world in society guidelines ahead of the original legacy standard of care. There is undisputable clinical evidence that it's more accurate, that it's a better patient experience that delivers better outcomes and takes cost out of the system.

So we really like to say we're on the right side of history here. We play a leadership role in creating the standard of care, and that really is the company's mission. I'm very pleased with the progress that we've made.

This last 4.5 years, since I've been here, has been tremendous relative to the results we put up, but we are just getting started. Our mission is to create a new standard of care. We're less than 2% penetrated right now, okay? So there's a lot of growth and a lot of runway ahead of us.

Just to sort of ground everybody in our strategy. HeartFlow has had a consistent strategy over the years. We've started with the most disease patients first. These are patients that have symptoms that need our -- when we started our first technology, FFRCT, we're now launching within the symptomatic world, the plaque technology, a little less disease, a little less urgent need but still highly important. And then we're moving upstream into the asymptomatic world. And I'm going to talk about what that pathway looks like here in the minute, but we're excited to expand into the new TAM that I mentioned.

We do this by virtue of our platform, okay? Our platform is an end-to-end platform. We think we need to provide value all across the continuum of care from detection to diagnosis to management, to treatment. We use AI in every component of the platform. We're clinically proven in everything we do. And as I mentioned, that's really the tip of the spear in changing -- driving adoption and changing behaviors. And we've learned and we do this, we seamlessly integrate into EMR and PACS system, okay, because workflow really, really matters.

Okay. So I'll take you through each component of the platform here. The first relative to detection, this is our HeartFlow road map analysis. We provide this on 100% of all the CCTAs that are sent to us. The easiest way to think about it is this is a GPS for the coronary tree, okay? It helps a CCTA reader, read a CCTA faster, and we have clinical data that shows you can read a CCTA 25% faster.

It reduces variability across a pool of readers. So if you're a health system and you're looking to adopt this for your entire program, you want to make sure that you have consistent reading results that achieves that end.

And then lastly, it increases confidence, very analogous to if we walk out of this building and said, somebody get us to the Golden Gate Bridge. If we had a map, we'd have more confidence that we would end up there, okay?

So this is on every CCTA that happens and we provide this free of charge and it's our way of helping with the workflow of our customers.

Next, we have plaque, okay? So what Plaque does, it takes a CCTA image. It measures the plaque burden down to the cubic millimeters. So whether or not it's calcified, noncalcified low attenuation that measures that. We are proven to have accuracy with published prospective data to be 95% accurate against the invasive gold standard. We're the only player in the market that can say that. When you have accuracy, you've accomplished the first conversation with physicians. You've got to prove that accuracy.

We've also proven that this analysis is useful. So when you give plaque to a physician more than 50% of the time, they change their management plan by virtue of having the insights that come that come with plaque.

As I mentioned, we're the most adopted plaque platform in the market, over 400 -- excuse me, 489 live accounts right now. From a patient applicability standpoint, this is applicable to basically twice the population that FFRCT was. This is 60% of the patient population are applicable to that it's covered and paid for if a physician has any visible plaque up until a 70% stenosis.

As of January 1, very happy there's a category 1 CPT code in place. And then we're making great progress relative to driving coverage. We now have over 70% covered lives with this plaque. So this is -- we're really excited with this, and we view this as our next wave of growth.

Next, we have -- and incidentally, that's a separate revenue stream for the company, and it's a separate reimbursement stream for the customer.

Next, we have our FFRCT technology. This answer is a fundamentally different question than plaque, okay? This answer is the disease that's found, that specific lesion is that significant enough that it warrants an intervention, okay? We have accuracy proven against the invasive gold standard of FFR to be as accurate or 95% as accurate as that. And we're the only player in the market that provides lesion-specific FFRCT values, and that's very important.

We have RCT level evidence that's published that proves with this pathway physicians and health systems can find more treatable disease. They can reduce the number of false positives going downstream, and they can increase the economic profile of those cath labs. Whereas plaque is applicable to 60% of all patients that come through. FFRCT is applicable to 33%. Any stenosis between 40% and 90% is covered. This is fully reimbursed as a Category 1 CPT code in place. We've treated over 500,000 patients around the world and over 1,800 accounts. And this is really the cornerstone of our platform.

Next, we provide a staging system, okay? So not all plaque burden is related -- is created the same. There's different stages. Again, you can think of the analogy here at the cancer. If you unfortunately, if a patient were to get a cancer diagnosis, you got Stage 1, 2 or 3, et cetera, we provide the same here with our proprietary staging system. This is the largest risk stratification system on the market, over 8,000 patients. And you can see the spread of the curves that occur. If you're in stage 4, you have a 5x increase of risk, okay?

With that, we've also developed with a cohort of preventative cardiologist a personalized medical management framework that informs physicians on how to treat their various cohorts of patients in each stage. We've got 90-day data showing that when followed, LDL reduces by about 19 points.

At the end of this second half of '26, we'll have 1 year data on the same registry. And we continue to drive the education in the market on how to use this tool to better treat patients, okay?

And then lastly, if the patient has a positive FFRCT and it's known that an intervention needs to occur, we've created the HeartFlow Navigator tool. And what this does is this allows interventional cardiologists to plan for a PCI procedure before the patient is on the table. And that's very analogous to how they plan on most every other procedure in the cath lab, like a TAVR, mitral, left atrial appendage, and that takes effectively all the best things from plaque, all the best things from FFRCT, a couple of new metrics that help interventional cardiologists plan and it allows them to do that before the patient is on the table. This is going to be available for all FFRCT patients. We're going to initiate -- we'll launch it in the first half of this year. We're going to initiate a 5,000 patient registry as well. And there's also a randomized controlled trial data that's coming out in the second half of the year from the P4 study in Europe.

So in totality, the platform matters and each component of the platform provides a very complete picture and very complete toolbox for our customers, okay?

So with that, just an image or a video, quick video on how it all kind of comes together. There's a case list when a physician wants a road map, click of a button, a plaque, click of a button, FFRCT click of a button and Navigator, click of the button. And every new product we launch is literally just the next button on the case list, and all of this is done seamlessly and integrated into the EMR and PACS systems.

Okay. So now I'll talk a little bit kind of -- that's the portfolio of technology. Let's talk a little bit on how the model works, okay? So we go out, we go to accounts that are already up and running with coronary CCTA programs, and we bring them into the installed base. Once we connect, we automatically connect and bring every CCTA that they do into our cloud. At that moment in time, the algorithm starts its work.

In a couple of steps in the -- early in that process. We have a human-in-the-loop quality control analyst. They make changes to the algorithm really around the development of that 3D model to make sure that no markers are missed. So for instance, if they need to change the center line of the lumen or if they need to tag a vessel that wasn't tagged. They make those little changes.

Every change they make is recorded, and we use that for future versions of the algorithm analogy I always like to make is if you're a Waymo taxi cab driving around San Francisco and you run through a stop sign next time you come through the same neighborhood, Hopefully, you're not running through that stop sign.

Okay. Once that occurs, the algorithm takes over and the road map is delivered directly back to the -- to that same case list and then the plaque and the FFR analysis and the navigator analysis already when they order them, okay?

All of this sits in our database. Now we've accumulated over $160 million annotated CT images, and we use that to continue to refine our algorithm and produce more technology. So we believe this model is -- has a lot of benefits for our customers. One, it allows us because the minute we get the CCTA, we start the road map, we're able to give that on every CCTA.

The HeartFlow analysis is ready when the physician needs it, okay? So when they sit down to read and they want to order it the next click of the button, they get HeartFlow analysis. We don't ask them to order it, wait and then come back and pick up the work again.

We also deliver -- as we deliver our results because we have our human in the loop, the results that we deliver require no extra work at the point of the customer. They can take our analysis and make their decision. They don't need to validate it to be accurate, okay? I mentioned we integrate into the workflow. And then obviously, we don't -- this is all click as you go. We don't ask customers to sign up for long subscription models where they can't manage -- where they can't foresee costs and there's no capital required either.

From our standpoint, we think this model really helps us acquire new accounts and maintain market share within and I'll show you in a minute what that looks like.

It also allows us to -- because we've consumed all of these CCTAs, it allows us to monetize new products as they become available. And plaque is the best opportunity that we've had yet to do this. I'll share with you what that looks like. And then this $160 million database we have, we use that to expand our gross margins and produce new technology, and I'll share you what that looks like.

First, relative to the installed base, I mentioned we're on 1,000 -- excuse me, 1,465 right now. From an account penetration standpoint, that's 46% account penetration. There's 3,200 accounts out there right now in the U.S. that are doing coronary CTA, that number grows by about 10% a year.

Within our installed base, we're 85% sole vendor. On the hospital side, we're 98% sole vendor, okay? So we think we have a good thing going on that front.

We initiated this new workflow where we consume all CCTAs, we did that at the start of 2022. Since then, our process case volume has almost 100% CAGR, okay? And obviously, the build case or the monetized case is a bit less. But we've been able to expand margins in spite of the fact we've been consuming and doing the work on all these consumed cases.

And we think that's an investment that was a wise decision to make at the time, and we think it's going to start paying off very nicely in 2026.

The process cases that we've consumed right now in 2025, we've run the algorithm already on 60% of the CCTA market, okay? So 60% of the CCTA market has already done our algorithm that's already reflected in our gross margins, but we've only monetized about 1/3 of them, okay? So what that means is as we launch new technology as we farm from this $100 million -- $160 million database and produce new technology, we're launching them into this kind of captured cohort, okay?

So we're excited about this, and we think this is going to pay dividends, not just with plaque in '26, but down the road from there as we continue to bring new technology to market.

But relative to plaque, that is the opportunity that's on our doorstep right now. Great work by the team, close to 500 as of New Year's, December 31. I actually checked earlier this morning. We are now active in more than 500 because we had some that went live in the first couple of weeks of the year. Equally importantly, we have 200 other accounts that have signed and are waiting to go live. And we've got about 500 right past that in the funnel that I would describe as in a very healthy funnel position. So because of this, I've got high confidence that by the end of this year, if not sooner, we'll be in over 100 -- excuse me, 1,000 accounts with plaque. And again, by frame of reference, it took us 8 years to do that with FFRCT. We're going to do it in less than 2 with plaque.

So that's one really, really important step in the process, and I feel great about where we are. But an equally important step is we need to educate the market and keep bringing evidence. I think everybody probably saw in November, ACC and AHA put out scientific statements on plaque that's, I think, substantiate the value and the utility of plaque for cardiologists, taking care of patients.

We have now fully enrolled our DECIDE prospective registry, close to 22,000 patients. We're going to read out the 1-year data from this registry in the second half of this year. That's going to be really important relative to continuing to educate the market.

And then lastly, from a coverage perspective, that's important as well. And we're ahead of plan on where we thought we'd be on that front, but I'm very pleased it's occurred. We, over the past year have been taking the decide data to payers. And then from there, they've been bringing coverage online. Aetna was the most recent that came in just at the end of December. United and Cigna came in, in October. So we're in a really good spot there. And of course, with the CPT Category I code effective with RVUs January 1, we're really setting up 2026 to be a big year for plaque. So I'm super excited about what's in store there.

Next, relative to gross margins. So I mentioned we have the human in the loop. That human in the loop is making annotations every time we release a new gross margin -- or should be a new automation algorithm. The time that, that human is spending in the loop is coming down as that occurs, margins expand. And we have a long history of doing just that. And that's one really important driver, okay?

The next driver, as I mentioned, this plaque analysis, that work's already been done. So as plaque is starting to monetize, it's monetizing with 0 incremental COGS. That's our second driver of margin expansion. And of course, volumes help and volumes as they continue to grow, we'll also expand.

So that's a long way of saying I've got really high confidence that we can be kind of top tier from a gross margin standpoint. We're going to have an eat in front of it. We're not guiding on to when that's going to occur, but I have got high confidence that it will occur.

Okay. And then lastly, that same database, right? Not only are we using that to refine our algorithm to expand our gross margin, we're also using that database to produce new technology. And we've got a good track record of every year, if not more, delivering new technology to the market.

In 2025, we did that with our second-generation plaque algorithm. In 2026, we're going to do that with our PCI navigator tool. And then in 2027, we're going to put out a plaque tracking analysis, which will allow customers to compare scan A on plaque to scan B on plaque we'll co-register both of them. And hopefully, if treatment is going well, we'll see regression in soft plaque for those patients. So that's coming in 2027. And again, a lot of the patients that hopefully are going to get the plaque this year could be good candidates for that down the road.

All right. So that's our -- that's kind of the model as it works. The other thing I'll say is earlier, I said, standard of care is our primary competition. And we're less than 2% penetrated against that. And that is the company's north star. That is the company's mission. That is where we source volume. Literally not a week doesn't go by where I don't hear from somebody that, "hey, I tried to get my brother and my mom, a HeartFlow and I couldn't, they ordered a spec". That's the focus of the team. That's how we're going to be successful.

But I also recognize that this is a really attractive category, AI plus CCTA is attracting new entrants. I don't think we're going to be a winner take all in this market, but I think we're very well positioned to be a winner takes most, okay? And we're going to compete with clinical data first and foremost, and we stand by the accuracy, and we stand by prospective published clinical data.

We're going to win with workflow, okay, and help physicians, we're going to meet them where they need to be with their workflow. We're not going to put extra steps into their day.

And then lastly, we're going to win with scale. And we've got a team right now around the world, over 250 field-based team members, not all of those are sales and customer success, field reimbursement, medical education, obviously, sales as well that are helping customers learn how to adopt HeartFlow and grow a CT first pathway, and we're dedicated to that.

And of course, last but not least, we've got a very strong IP portfolio that we think gives us some additional barriers to entry.

Okay. So next, let's talk about TAM expansion. And then I think soon Robbie will be ready for questions. I mentioned from a near-term standpoint, everything I just talked about is happening in the symptomatic world. It's a $5 billion TAM, 8.6 million patients. We want to take a step-wise approach which we think has a lot of advantages to enter into the asymptomatic opportunity or the asymptomatic market here. And that's about a $6 billion TAM.

Now we think there's patients that are already in the health care system that have already been diagnosed with coronary artery disease that exists right now that if we design and deliver on the right RCTs, we can expand the current codes to include them, okay? And we've got three separate subpopulations here, one being patients with prior heart attacks or PCI, the other being patients with an elevated calcium score, and the third most logically are patients that have already had a plaque, okay? We're going to initiate three separate RCTs against each one of these populations. The endpoints won't be hard MACE outcomes. There will be plaque regression and LDL, okay?

And we think with those RCTs, we'll be able to take those to payers and expand the coverage for these. We spend a lot of time in 2025 getting smarter on what the right entry is. There is a path of entry where you could go and go to patients that aren't already in the health care system, try to screen them and find them and bring them in. We're not taking that off the table long term. But from where we sit, this is a much more capital-efficient way to get into this market, and it doesn't make a lot of sense to walk past these captured patient populations to go chase a screening opportunity.

So we're super excited about this. The other thing I'll note is in probably Q1, if not early Q2, we expect to see new guidelines for prevention and lipid management come out. That's really important for us to see because we'll match the control arm of each of these RCTs to that. This is all our -- with our existing technology. So we don't think there's technological risk in this strategy. And as I mentioned, what I'm most excited about is allows us to help these patients before the next decade. So this is going to happen before 2023 -- or excuse me, 2030 will be in these markets. So we're excited on that.

And then last thing I'll say, just relative to financials, these are Q3 financials. Everybody has seen it. My message on this is I look forward to sharing both full year '25, Q4 results in our upcoming earnings call in Q1. Just to reiterate, no surprises, we're going to beat consensus for sure. And I look forward to continued success on the financial side, both in the next earnings and many more to come on that.

Okay. And then lastly, and then I think we'll go to Q&A. Just from a team standpoint, I feel great about the team. We've got a good team in place. Everybody here, not just on this slide, but across the company are committed to creating a new standard of care. And we had the IPO in August. And I really think as great as the first chapter of HeartFlow's growth story was, I really think '26 and beyond is going to be even better.

So with that, thank you for everybody's time. And I think, Robbie, I'll turn it over to you for some questions.

Well, great. Since you started with fourth quarter and you beat consensus and had a good quarter. How good was fourth quarter which is really all we want to know.

I'm going to let my CFO answer -- no wait. Fourth quarter was strong. I think we shared the installed base numbers. That's the biggest year we ever had. Within Q4, it was the best Q4 we ever had. We shared the plaque, some of the plaque metrics, phenomenal work by the team in that regard. So I feel really, really good there. And most importantly, I just feel really good about where we are as we start '26.

Great. Maybe just on '26, I believe in the Street right around 23% sales growth. Obviously, you'd be fourth quarter. But any early thoughts on how you feel about what that range sits and if you're comfortable with that?

Vikram, do you want to answer that?

Yes, absolutely. So a little premature to talk about '26 in detail, but we're pleased with the setup as we approach '26, especially based on what we saw in Q4, the installed base expanded meaningfully, utilization in both existing and new sites continues to be consistent and predictable on the FFRCT front.

And then plaque is generating a lot of interest. And as John shared, there's strong momentum in pulling the plaque product through our installed base. So we're excited to get that through our installed base in 2026. We'll exit 26 with north of 1,000 accounts, as John mentioned.

All told, you've got FFRCT being the first building block, very predictable, very consistent utilization.

On the plaque side, as we've talked about a couple of the key drivers there. First is coverage getting to 70%, which you've already achieved. But the second component is really driving medical education. And so that's going to be the huge focus of the team going into '26.

We've got more data readouts coming out of DECIDE, which will help. But inherently, our expectation is utilization on plaque will ramp gradually through the course of the year and really be more back weighted from a revenue standpoint in '26.

Maybe we could touch on plaque, because you've checked the box on coverage here north of 70%. Medical education will obviously probably be the most important component of the launch in 2026 here. But I hear from a lot from investors, why shouldn't every account that's using HeartFlow today also use plaque given such an easy cross-sell at checking a box and reimbursement is favorable.

No, it's a great question. So what I'll say is it's too early to tell what that plaque utilization is going to be. I mean the full applicability of plaque is 60% of all CCTAs. It's a very reasonable assumption that as a new plaque user comes online, they're going to use it in a much narrower range. And that's the same thing they did with FFRCT. So it's probably more likely going to be the first usage pattern is patients with more plaque. And those physicians are going to get comfortable with that, and they're going to learn how to interpret it, they're going to learn how to manage patients on it. And then they're going to expand to maybe patients with a lot of plaque and medium plaque. And then ultimately, we hope they go to the full extent of the range.

It's too soon to tell what that looks like. But as Vikram talks about the need for medical education, that's what we need to do, okay?

Far and away, the #1 question I get when I talk to a physician about plaque, everybody is excited for all the obvious reasons. The first question is, how do I use this? There is no ground truth right now on what to do with a patient with 248 cubic millimeters of plaque with 170 of them noncalcified. We don't know that. The medical community needs to be educated. We'll be a part of that education, but we're not the only one that needs to educate them.

So that has to take time. I think when it gets back to the controlling what we can control, I think the team has done a good job doing that. We have gone to our existing accounts with the plaque opportunity and sold contracts into them. And as we go to new accounts, which we did, we're including plaque in virtually all of those as well. So that's step number one.

Step number 2 is you got to make sure the coverage is there, okay? And we started the journey generating the DECIDE data years ago more than a year ago. We're taking that data to payers and we're getting coverage. Okay. So that's step number 2.

Now we're in the market, and they got to try it, and hopefully, they got to use it again. And hopefully, they're going to tell their colleague about it. And hopefully, their patient that has it will tell one of their family members. And from there, we'll see a ramp. But -- and I know you're not suggesting this, but it's the last thing I want to do or we want to do as a public company is call a number and not deliver, okay? So our guidance is nothing material on plaque until the tail end of '26. We're going to do everything we can to get patients to -- or physicians to use it. But again, it's just too soon to tell.

When you think about the addressable market and the real addressable market. Do you look at just how many CAD test there are? Or are you firmly looking at CCTA? And then what percentage of those you're capturing? I guess do you look at it as just people should be using HeartFlow, and that's the bottoms-up driver? Or do you look at it as a percentage of the modality being used?

Our north star is to create a new standard of care. And there's 8.6 million patients every year that get tested, noninvasively for coronary artery disease, and we want to do everything we can to shift them to a CT plus HeartFlow pathway. So as we think long term, over the course of where the company is going to go over 10, 15 years. That's our north star. And then we're doing the same with asymptomatic as we spoke about.

Now much more tactically, every 90 days when we go out and execute, we only go to accounts that are up and running on coronary CTA. We help them grow their programs. And that's what the teams are doing on a much shorter time horizon.

How are you thinking about profitability? Obviously, your new company, you just IPO-ed last year and growing strong double digits, but everybody wants you to grow strong double digits with the drive towards profitability. So how are you thinking about when you can cross the threshold of cash flow breakeven? And what are some of the puts and takes as you think about that?

Yes. We're consistent with what we communicated before at the time of IPO. We see profitability as a choice. We have plans to get to profitability within 3 years of IPO, which puts us in the middle of 2028. We've got all of the building blocks with consistent strong revenue growth, margin expansion, courtesy the investments were made in the automation of the algorithm volume leverage as well as plaque being a tailwind, not just on gross margins, but also operating margins. And we've got a very capital efficient commercial model. Which enables long-term strong operating margins to be had.

But ultimately, we want to be growth oriented here. So we'll continue to invest and not walk past those TAM expansion opportunities that we just walked you through. But we'll be disciplined here. We'll continue to look at EBIT metrics year-over-year, ensure we're on that glide path to profitability and get that profitability in 2028.

As we think about top line, there's obviously volumes and there's pricing. You've been using volume-based discounts to help drive volumes and ultimately, revenues. How should people think about pricing going forward, I guess, both for plaque and for FFRCT?

Vikram, maybe I'll share some commentary and then you can tie it out. So first, drivers in pricing right now. We have volume-based discount, which everybody knows is very common in med tech. These are large customers that are growing. We give incentives to continue to grow. And if they hit them, there's a discount in that regard.

There's also roughly two segments to the market. There's a hospital-based segment. where I think when you look at the full value from a HeartFlow platform, there's a lot of value accruing across a CV line administrators line so to speak. But then you have outpatient imaging centers, which are much more -- they're governed by the PFS side of the house, and they're much more price sensitive. We're seeing a little bit higher growth in that segment. And because they get a little bit lower ASP, that's where what's driving the pricing mix.

Longer term on plaque, we see opportunity to raise it. Okay. Now I don't think '26 is the year to do that because these contracts are kind of baked right now. But as we think in the out years, we think there's certainly opportunity. Maybe Vikram, I let you...

Short answer on FFRCT is the more recent trends we saw in '25 will persist for the reasons John mentioned. Continued growth out of the PFS segment as well as volume-based rebates. The reason for these volume-based rebates is because CCTA is only 10% penetrated in the NIT. So that's the north star getting category conversion. We'll continue to deploy those structures to drive top line growth.

And then to John's point, longer term, given where coverage is coming in higher than expected. We do see upside in the longer term. But near term, we'll be more disciplined.

Great. We're unfortunately out of time. Thanks for a great discussion. Thanks, everybody, for coming today.

Thank you.

Good day, and thank you for standing by. Welcome to the HeartFlow Third Quarter 2025 Earnings Conference Call. [Operator Instructions] Please be advised that today's conference is being recorded. I would now like to hand the conference over to your speaker today, Nick Laudico, Vice President, Investor Relations. Please go ahead.

Good afternoon, everyone, and welcome to the HeartFlow Third Quarter 2025 Earnings Conference Call. Joining me today are John Farquhar, HeartFlow's President and Chief Executive Officer; and Vikram Verghese, our Chief Financial Officer. John and Vikram will provide prepared remarks and then open the line for Q&A. The earnings release accompanying today's discussion is available on our Investor Relations website at ir.heartflow.com. During this call, we will refer to certain non-GAAP financial measures.

Reconciliations to most comparable GAAP figures can be found in today's earnings release. I'd like to remind everyone that certain statements made on this call are forward-looking within the meaning of federal securities laws. These statements are based on management's current expectations and beliefs, involve risks and uncertainties, and actual results may differ materially. Important factors that could cause such differences are detailed in our earnings release today as well as in our filings with the SEC. Please note that both this live call and a digital replay will be available shortly after the call concludes.

With that, I'll now turn the call over to John Farquhar, our CEO.

Thanks, Nick, and good afternoon, everyone, and thank you all for joining us. We're very pleased to host our first earnings call as a public company following our August IPO, which generated approximately $332 million in net proceeds. On behalf of the entire HeartFlow team, I want to extend my sincere gratitude to the investors who supported the offering and who share our conviction in the company's mission and long-term opportunity. The third quarter of 2025 was an outstanding one for HeartFlow. Total revenue reached $46.3 million, representing a 41% year-over-year growth rate, driven by strong adoption of the HeartFlow AI platform.

Global revenue cases grew 48% year-over-year and 7% sequentially, driven by consistent platform utilization, the continued expansion of our commercial footprint and strong overall market growth. Given this performance and the visibility we have into the remainder of the year, we're initiating full year 2025 revenue guidance in the range of $173 million to $173.5 million, representing approximately 37.5% to 38% growth. For those newer to the story, HeartFlow is an artificial intelligence company transforming the detection, diagnosis and management of coronary artery disease, also known as CAD, which is the world's leading cause of death. CAD occurs when the arteries supplying blood to the heart muscle narrow. Full blockage can cause a heart attack.

And unfortunately, a heart attack happens every 40 seconds in this country, and half of these occur with no prior diagnosis. Our AI software platform enables physicians to see the actual disease itself and precisely quantify and measure it, helping physicians find disease earlier and diagnose it more accurately. Our platform is the most widely adopted AI solution for CAD management on the market today, and we're leading the charge to create a new standard of care. To date, we've served more than 500,000 patients worldwide. We're endorsed by both the ACC and AHA medical guidelines. We've been peer-reviewed in over 600 clinical publications, and our AI algorithms are powered by the world's largest coronary imaging dataset of over 110 million annotated images.

To put the magnitude of the problem we're solving in context, each year in the U.S. alone, roughly 8.6 million Americans see their physicians with symptoms of coronary artery disease. This can be chest pain, dizziness, fatigue or shortness of breath. Most of these patients are still evaluated using traditional stress-based imaging tests that have been around since the 1990s. These tests merely infer the presence of disease through surrogate markers. And unlike our technology, they do not measure the actual disease itself. The result of this is poor accuracy, and it's well documented. Up to 30% of patients with treatable CAD are sent home undiagnosed, while up to 50% of patients sent to the cath lab ultimately don't require an intervention.

The impact of this inaccuracy is severe, unnecessary death and human suffering, inefficient and inconsistent delivery of care and unnecessary costs to say the least. All of this represents a massive opportunity for HeartFlow to redefine this paradigm by creating a new standard of care. Using a single cardiac CT scan or a CCTA, we apply our AI to create a personalized 3D model of each patient's coronary anatomy and precisely measure the actual disease down to the cubic millimeter. This breakthrough positions HeartFlow as the leader of a large and expanded opportunity. In the U.S. alone, we address a total addressable market of $5 billion, and that market remains deeply underpenetrated. And despite our emerging scale, we're still in the very early innings, and we have a long runway ahead of us to support sustained and durable growth.

Now turning to our technology platform. We have 4 fully integrated products that together represent the most comprehensive AI-powered platform on the market. Our platform spans the full continuum of coronary artery disease from detection to diagnosis to management and ultimately to treatment planning. So whether you're a radiologist detecting and diagnosing disease, a general cardiologist medically managing a patient or an interventional cardiologist planning treatment for the cath lab, our platform provides physicians the power of AI in their decision-making and clinically validated insights with the click of a mouse. There's 4 integrated products to our platform.

First, our Roadmap Analysis. You can think of this as a GPS for a patient's coronary [ tree ]. Our AI identifies and color codes troubled areas of disease, quickly informing the radiologist on where to focus. This workflow tool improves the speed at which a radiologist can read a CCTA by 25% and has been shown to improve intra-reader consistency by over 40%. It's a great example of how we're using AI to bring speed and standardization into everyday clinical workflow. Next, our Plaque Analysis. It's well understood in the clinical community that both the amount of plaque as well as the type of plaque are important predictors of future heart attacks. Our Plaque Analysis quantifies the total plaque burden and type down to the cubic millimeter with 95% agreement versus IVUS, the invasive gold standard.

Knowing this information informs physicians on how to best medically manage their patients. So instead of using broad-based tools like family history, population-based risk scores or surrogate markers like LDL, we provide precise clinically validated measurements to physicians, so they, in turn, can develop personalized medical management plans for their patients. Third, our HeartFlow FFRCT Analysis. While knowing the plaque burden and type relative to medical management is a game changer, it isn't enough when it comes to deciding if an intervention is needed. To accurately make this decision, the physicians need to understand the degree to which the blood flow is impacted across the precise location of the disease.

Our FFRCT Analysis does just that. We use AI and computational fluid dynamics to measure the blood flow noninvasively with lesion level precision. It's validated for accuracy against the invasive gold standard of FFR and our large randomized controlled trial, the PRECISE RCT, proved better diagnostic accuracy and better cath lab efficiency against the standard of care. And lastly, our PCI Navigator tool. We introduced this at TCT last month to great excitement amongst the interventional cardiology community, and we plan to commercially launch it in 2026. This is an AI-driven planning tool that integrates anatomy, plaque and physiology, all aligned to optimize potential stent placement. Together, these 4 fully integrated products represent the most comprehensive AI-powered platform on the market.

We are the only company with offerings that span the full CAD continuum from detection to diagnosis to management and now to treatment planning. Now turning to our core strategic pillars and our third quarter execution. We have 3 core strategic pillars: first, driving commercial adoption; second, advancing our innovation pipeline; and third, extending our clinical leadership. With respect to driving commercial adoption, we've made great progress in the third quarter. We continue to focus our efforts on expanding our installed base of accounts. As a reminder, we target a total account opportunity of about 2,900 U.S. accounts that are currently actively operating coronary CTA programs. This target account pool is growing at about 10% annually.

In Q3, we continue to add new accounts into our installed base rapidly through strong efforts from our territory sales managers or TSMs, which are a dedicated arm of our sales team that focus solely on onboarding new accounts, and we expect to continue to make rapid progress moving forward. Now once a new account is added into our installed base, a separate arm of our sales team, territory account managers or TAMs, focus directly on the referring cardiology community. They increase awareness of our technology, and they help educate physicians on the value of the HeartFlow AI platform. Currently, we're expanding our TAM sales force and making great progress with the goal of doubling this team by year-end 2025, which is going to further extend our reach into this important segment.

Both of these efforts, expanding our installed base and educating the referral community will pay even greater dividends as we scale our Plaque Analysis, which we expect to be a major growth driver for us. Now recently, we've had several notable achievements relative to plaque. On the coding front, CMS announced that the Plaque Analysis will transition to a Category 1 CPT code on January 1, 2026. This is a major milestone that formally assigns RVUs and for the first time, enables physician payment in the clinic setting. CMS also finalized a national Medicare payment rate of $1,026 for the clinic setting and propose to continue the hospital rate at $950. From a coverage perspective, UnitedHealthcare and Cigna recently made policy updates covering HeartFlow Plaque Analysis for patients with acute or stable chest pain across all lines of businesses.

This brings coverage of HeartFlow Plaque to 57% of U.S. covered lives as of October 1, 2025. Now as a reminder, we're not forecasting material plaque adoption until we receive 70% of all covered lives. And this is the same coverage inflection point that unlocked broad adoption for our FFRCT Analysis. Now to our second core pillar, which is advancing the innovation pipeline. Our unmatched dataset of more than 110 million CCT images and our embedded infrastructure that enables bidirectional data sharing with our customers gives us a unique ability to iterate, validate and deploy new technology rapidly. In Q3, we demonstrated this. We received FDA 510(k) clearance for our next-generation Plaque Analysis and began the full commercial launch.

By retraining algorithms on thousands of new CCTA scans, we achieved a 21% improvement in plaque detection performance versus the first-generation model. We also expanded our nomogram to 273,000 patients. This is 9x larger than any other plaque quantification study. In addition to this, we also expanded our platform to interventional cardiology. In TCT 2025, we unveiled our new PCI Navigator tool, which is our CT-guided PCI planning tool. And the feedback from leading interventional cardiologists was overwhelmingly positive. Physicians recognize how CT-guided PCI can improve both efficiency and safety by enabling preprocedural planning where it traditionally has not existed before, and we're excited to bring them this innovation. Because of its clinical value, PCI Navigator will strengthen our relationships with key customers and further extend our AI technology across the full continuum of CAD care.

Our third pillar is extending HeartFlow's clinical leadership. Clinical evidence is the foundation of trust in cardiology. It determines how physicians practice, how payers reimburse and how quickly new standards of care take hold. This is why we continue to invest deeply in building the data that shapes guidelines, secures reimbursement and accelerates adoption. Our DECIDE registry is a clear example of this. It's the largest prospective study of its kind, enrolling about 22,000 patients across more than 30 U.S. sites. In July, DECIDE met its primary endpoint, showing that more than half of physicians changed their management decisions after reviewing HeartFlow Plaque Analysis compared with CCTA alone. Notably, 30% of patients with no calcified plaque also had management changes, demonstrating the technology's clinical sensitivity.

This data is now being used actively in discussions with commercial payers as well as in our medical education efforts. For PCI Navigator, which I just spoke of, new data continues to validate its potential. The PLAN CALCIUM study, which was presented at TCT, showed that using the CCTA plus HeartFlow pathway changed calcium modification strategy in more than half of evaluated lesions. This demonstrates how CT-guided PCI planning can bring new precision to interventional cardiology. In the first half of 2026, we plan to launch the NAVIGATE-PCI registry. This is a real-world study that will enroll about 2,500 patients across 30 U.S. sites and will measure how PCI Navigator affects physician confidence, procedural planning and patient outcomes. Now longer term, beyond symptomatic patients, we see a major opportunity to redefine the standard of care for those at risk but without symptoms.

Roughly 200 million individuals globally stand to benefit from earlier image-based detection rather than reliance on population-based risk scores alone. Capitalizing on this opportunity aligns perfectly with both our mission as a company as well as our core competencies. And we look forward to sharing more details of this strategy in future discussions. To close, we're very encouraged by the strength and consistency of our third quarter performance. It reflects the continued momentum of our strategy and the great commitment of our team. We're seeing broad adoption of the HeartFlow platform, sustained progress across our innovation pipeline and growing clinical validation that reinforces our leadership position.

Our focus remains clear: execute with discipline, put patients first in all we do and continue advancing the field toward a new standard of care for coronary disease. We believe this focus positions us exceptionally well for long-term growth and further impact.

With that, I'll hand the call over to Vikram Verghese, our CFO, to review the financial results and outlook.

Thanks, John, and good afternoon, everyone. Unless otherwise noted, my remarks reference the quarter ended September 30, 2025. All financial metrics I refer to other than revenue will be non-GAAP, unless otherwise noted, and all growth rates will be year-over-year. Reconciliations to the comparable GAAP measure are on our IR website. Total revenue was $46.3 million, up 41%. U.S. revenue grew 42% to $42.5 million. OUS and Other revenue grew 24% to $3.8 million. Total global revenue cases were 51,805, representing 48% growth, driven by continued strength in our U.S. FFRCT business. We continue to see strong utilization at both existing and new accounts in line with historical trends. We saw particular volume strength in the clinic setting, a rapidly growing segment of the market.

We also expanded our installed base at a rapid pace, adding new accounts above expectations. Third quarter gross margin was 76.8% compared to 75.9% in the third quarter of 2024. The year-over-year margin improvement primarily reflects volume leverage and continued gains in production efficiency. Operating expenses reflect disciplined growth investments. Third quarter SG&A expenses were $30.3 million, up 15%, driven by investments in headcount to drive further adoption of the HeartFlow platform. Research and development expenses were $16.3 million, up 44%, driven by investments in technology to advance the HeartFlow platform and in clinical research to expand our evidence base. Operating expenses were 101% of revenue versus 114% a year ago, demonstrating improving operating leverage while we continue to invest for growth.

Operating loss was $11.1 million compared to $12.6 million last year. Non-GAAP net loss was $13.2 million or $0.27 per share compared to non-GAAP net loss of $16.2 million or $2.90 per share in the third quarter of 2024. On a GAAP basis, net loss was $50.9 million or $1.04 per share versus $19.1 million or $3.43 per share last year. The GAAP results include a $32.1 million noncash charge from the remeasurement of our common stock warrant liability driven by higher share price during the quarter. As of October 2025, the warrant holder net exercised all warrants. As a result, we expect to record a final noncash warrant revaluation charge of approximately $9.3 million in Q4 '25. Q4 will be the last quarter with any warrant remeasurement impact.

Third quarter GAAP net loss results also include a $4.8 million noncash benefit from remeasurement of our derivative liability, which terminated upon conversion of our convertible notes at the IPO and a $6.4 million loss on debt extinguishment related to the prepayment of our term loan in August 2025. Weighted average basic and diluted shares outstanding were 49.1 million in the quarter. Given the mid-quarter IPO, the weighted average understates the post-IPO share base. For modeling go-forward EPS, period-end basic and diluted shares were 83.5 million. Turning to the balance sheet. We ended Q3 with $291.2 million in cash and equivalents. This reflects approximately $332.3 million of net IPO proceeds and our subsequent debt actions, mandatory repayment of $55 million and full prepayment of the remaining $60.1 million term loan, both completed in Q3.

With 0 debt and a strengthened cash position, we believe we are well capitalized to fund operations through profitability while continuing to invest in R&D and commercial expansion, consistent with our offering disclosures. Now turning to 2025 financial guidance. We're initiating total revenue guidance for the full year 2025 to be in the range of $173 million to $173.5 million, representing 37.5% to 38% growth. This outlook reflects durable adoption of the CCTA plus HeartFlow pathway and our recent commercial outperformance. This quarter's results reinforce our confidence that gross margins can reach 80% or higher and that we are on track to achieve cash flow profitability within 3 years of our IPO.

I would now like to turn it back to John for summary closing remarks.

Thank you, Vikram, and thank you all for joining us today. We deeply appreciate your continued interest and support as we work to advance the HeartFlow AI platform toward becoming the new standard of care for detecting, understanding and managing coronary artery disease. We're excited about the progress we're making, and we look forward to updating you on our continued momentum in the quarters ahead.

With that, I'll turn the call over to the operator for Q&A. Operator?

[Operator Instructions] Our first question comes from the line of Robbie Marcus from JPMorgan.

Congrats on a great first quarter out of the gate here. Two for me. Maybe first, third quarter came in far above expectations. You raised the -- you started the guide above where the Street was for the year and more than the beat. Walk through sort of what you're seeing in accounts, the level of interest in FFRCT and sort of how you're winning in the field, what's driving the great results?

Yes, sure. Thanks, Robbie. Good question. First thing, what I'll say is the confidence in raising that we have is really driven by 3 things. I'll speak to that first. If you think about the FFRCT business, that's our core business. We view this as extremely durable. And as I look at it, it's executing really well on 2 fronts, okay? One front is we need to grow our installed base. So that's out there. We're hunting for new accounts. We're bringing new accounts into the family. And then the other side of that is then we go out into the referring community, we raise awareness of the new pathway.

And both of those then translate to these different cohorts, which I think is where you're going with your question, where our existing sites are performing really, really strongly as are our newest sites. We're bringing them in. We're bringing as many as we need to. They're ramping really well. So that's kind of the fundamental driver of the confidence. Now on top of that, I've got visibility to the funnel. So I can see kind of down the road a bit. And I feel very good about where we are. And given that, I've got confidence in the moving forward trajectory there. And then I think the last piece as kind of the underlying foundation behind all this is just the strength of the overall category or the underlying category, category being coronary CCTA.

We know historically, that's grown in kind of low 20% range. I'm not seeing anything now that would make me think it's anything south of that. And I think with the reimbursement that came on, the increased reimbursement that came on in 2025, with Plaque coming on in the future, I think coronary CTA is a very durable trend moving forward as well.

Great. Maybe a quick follow-up. Everyone is very excited, I'm sure you are, about the Plaque launch next year once you hit more complete coverage. You're almost there as of the beginning of fourth quarter. How should people think about the launch of Plaque? What will it take to roll out? How easy is the adoption? And are there barriers that you're working to reduce ahead of it?

Yes, good question. So first thing I'll say is I'm extremely excited about Plaque. And I really believe this is going to be the next wave of growth for this business. There's really 3 pillars that we're working, and I'm happy with the progress we're making on all 3. First is we've got to secure reimbursement coding, payment and coverage. The second is we need to build compelling evidence, clinical evidence. And then the third is we have to take that clinical evidence and go into the market to do the real market development work through educating physicians on the value of the tool. And I would say across all 3, we're making really good progress.

I can tell you qualitatively, when I speak with customers at conferences and when I'm on the road, there's very, very high interest. So I feel good there. And then similarly, I mentioned kind of the funnel that I have visibility to. So I feel good about kind of those -- the out time line there. Now that being said, I got to also sort of answer the question relative to when we think this is going to hit in a material way. And there's 2 pieces of that. One, we've made great coverage on -- great progress on coverage so far. As of October 1, we had 57% total covered lives. United and Cigna most recently came on. We don't believe -- I don't believe we're going to see material adoption until that number gets north of 70%, okay?

And the reason I think that is we live this journey through FFRCT, and we didn't see a real inflection until FFRCT coverage was north of 70%. The second element here that I think is equally important as we think about this is the market development, physician education piece of it. The #1 question I get from customers when I'm out is how do I use it, okay? And it's a little nuanced and it's different than the education journey we went on with FFRCT, where there was a established invasive measure of FFR that already existed in the market. So once we proved accuracy against that, the adoption followed.

This, there's real teaching. And we're leaning into our DECIDE data and all the efforts that we make from -- to our medical education. And those 2 really need to work hand in glove before we think the market is going to really accelerate here. All that said, I would not advise any material adoption until the tail end of '26.

Our next question comes from the line of Patrick Wood from Morgan Stanley.

I'm going to keep it to one just for the sake of expediency. But TCT, I saw you guys at the conference that the room was pretty packed and interest on the Navigator side sounded very high. How are you thinking about how that fits into the broader portfolio? And strategically, how that's going to position you to penetrate accounts faster and lock those customers in? How are you seeing that fit into the broad offering from HeartFlow and its strategy?

Yes. Good. Thanks, Patrick. Appreciate the question. So what I'll say, I mean, when you think about our innovation strategy, an important piece of that is we want to add new clinical insights into the platform, okay? And Navigator is a great example of us doing that, okay? We're the only company with an end-to-end platform from detection to diagnosis to management and now with Navigator to treatment planning, and we believe in the power of the platform, okay? And what this allows us to do is this gives interventional cardiologists, which are an important stakeholder, as everyone knows, in hospitals and health systems, this gives interventional cardiologists information that they normally don't have until the patient is already on the table in the cath lab.

And it allows them to plan a procedure before the procedure is actually going on. And this is very analogous to how they're already used to planning with other procedures like left atrial appendage, TAVR, mitral, tricuspid, et cetera. So from that standpoint, interventional cardiologists, and certainly, I was at TCT as well, a lot of interest there. And we think it strengthens the platform and helps engage a very important stakeholder and interventional cardiologists. So we're excited for that, and that will be an important part of our 2026 launch plan.

Our next question comes from the line of Matthew O'Brien from Piper Sandler.

Maybe for starters, John, just talk a little bit about the Q4 implied guidance. Just at the midpoint, it doesn't really signal any kind of sequential improvement. And so that's not what we saw last year. Is there anything specific that you're trying to call out? Or is this just kind of part of your guidance philosophy to try to take a measured approach to things?

Matt, this is Vikram. Good to hear you. First off, I'll reinforce what John had said. We're really pleased with how the quarter came together. It was comprehensively strong. The installed base expanded meaningfully, and utilization did come in higher at both new and existing sites. Additionally, the sales team is firing on all cylinders. As we think about 4Q, just given the better-than-expected 3Q, the sequential comps do get a bit tougher. And second, our guidance philosophy here is really to -- is really anchored on a high conviction forecast, which we believe provides a reasonable baseline from which we can overdeliver. So we felt it was important to embed a measure of conservatism in our estimates, and that's what's really reflected in the guide here.

Okay. Makes sense. And then I think, John, you were kind of touching on this a little bit, but there's just a number of different components that are going to drive your growth going forward, and I know not to expect Plaque until back half of next year. But the different components, how do we think about how those can trend between new centers, utilization, et cetera, over the next 12 months? And I'm sorry to sneak this one in. But Vikram, the gross margin and EBITDA was way better than expected. Are those durable improvements? Or were there any one-timers that maybe affected those positively?

Matt, I'll take that first, and Vikram can come back. So I mean I think there's a number of growth drivers. To answer your question upfront, I think they're all durable and one of these is emerging, okay? And that's Plaque. I'll talk you through each. So the first I mentioned, the category in and of itself, the coronary CCTA category has historically had a very healthy clip, low 20%. I think it's got -- we've got every reason to believe that's going to continue on the backs of new guidelines, on the backs of improved reimbursement. We know in the market, more and more physicians are raising their hands, wanting to learn how to become a CCTA reader.

So I feel really good on that front. And I don't think capacity is going to hold us back on that, okay? The second is our existing cohorts of accounts. And we've got a large cohort of existing accounts. And in any given year, about 75% of our volume growth comes from existing accounts, okay? And we've got a good track record and good -- this is visibility that we have internally that we can count on really good consistent adoption from these accounts once per annum. So I feel good -- and that's on our FFRCT business. So I feel good in that regard, and I think that's, again, stable and durable. The next kind of component of the equation, so to speak, is installed base growth. And this is how do we go and find new accounts, bring them into our family and then grow them so one day, they obviously become existing accounts.

And there's a pool, a large pool by which we can go hunt for, so to speak. There's 2,900 accounts out there. That 2,900 adds about 300 every year. We have a long track record with our TSM, our territory sales management arm of the sales force of being able to bring them on board. And we feel really good. And as Vikram mentioned, we've had some good overperformance in that regard on the ability to go convert. And things like -- incidentally, things like Navigator that helps support that action. And then lastly, Plaque, and I talked about that. But Plaque, once it's material, I think absolutely is going to be a very strong and very important wave of growth for us.

And just as a reminder, this is going to go into the same installed base through the same bidirectional data pipes. It's the same interface, so the same user interface. It's literally a button that's going to go right next to our FFRCT button. So there's huge synergies there. Now again, like as I mentioned, very pleased with the progress we're making, but I don't want to call anything material until the tail end of '26.

Yes. And on gross margins, the short answer is we do think they're durable. These margins can move around sequentially, but a lot of the benefits that we saw will persist into 4Q. Just stepping back on gross margins, there are really 3 principal drivers there. We've got higher case volumes. Second is the automation of our algorithm. And third is the emergence of Plaque revenues. In 3Q specifically, Plaque revenues were not material. So the beat was really driven by higher volumes as well as the automation of the algorithm, which is, again, speaking to the AI flywheel that we had talked about before.

Looking at 4Q, assuming the midpoint of our revenue guide, we expect gross margins to be similar to what we saw in 3Q. And finally, on your question on EBITDA, we will see an uptick in OpEx given the headcount we're filling in the field as well as incremental R&D expenses and the revenue and gross margins we've already guided to.

Our next question comes from the line of Rick Wise from Stifel.

I was hoping, John, you'd expand on your comments about your expansion actually of the TAM sales force. It's going to double by -- I think, by the end of this year, you said, if I heard you correctly. Talk about how quickly that expansion turns into productivity and revenues? Is it have an immediate positive impact? Is it something that happens over 6, 9 months, which is sort of what I would assume? And what are your plans or thinking beyond that? Do we see -- do you expect further expansion? Is that going to be a major aspect of the growth engine?

Yes. Rick, good question. Nice to hear your voice. So we're making good progress. As we talked about, we're doubling the size of the sales force. We're on track to do that by year-end as we've set out to do. Just as a reminder, this team effectively didn't exist 4 years ago. So we've gone from nothing to where we are now to then doubling that. Moving forward, we'll evaluate probably second half of next year on the need to continue to expand. I don't have an answer to that right now, but I certainly don't want to take anything off the table. This role in our sales force is a really important one, but it is different.

This is not a traditional medtech role. It's much more of a pharma type model. They play a really important role in the referring community, okay? So they engage with referring cardiologists. And a lot of what they're doing now is around educating. So we've got the DECIDE data out there on plaque. We're helping to coordinate medical education events. We're helping to facilitate questions getting answered. So that's a real focus on that team right now. I'll let Vikram speak to the productivity.

Yes. So relative to that, again, to reinforce what John said, this is a strategic investment on our part, especially with Plaque coming down the pike. I would caution against using any kind of transactional sales metrics based on territory counts and revenue per territory as is the case with medtech. This is much more focused on referrer activation, increasing market awareness, the kind of activity you see typically in pharma. So we're continuing to be -- we are excited about this expansion. We think it's a longer-term tailwind. We're -- the trends are encouraging, but it's much more of a market development investment we're making.

Yes. And maybe last for me. I want to reflect a little further on the plaque opportunity. I hope I'm not misremembering that you've launched the product in a limited number of sites. I don't know if I should -- I'm saying it correctly, but basically for clinical trials, but beta settings, that's my impression anyway. And as you say, reimbursement is key to for broader launch. But I was hoping you talk about the early experience in this limited number of sites. What you've learned about the utility, the utilization and how it's informing your thinking as you get ready for a broader launch with more abundant reimbursement in hand?

Yes, sure, Rick. So yes, you're correct in remembering that we've launched through our DECIDE sites, and we have experience in that regard, primarily product performance and usability. And of course, we're capturing the data that's appropriate for the registry. I think we're learning what we were -- kind of what we were expected to learn going in. I mean there hasn't been a huge surprise. I mean the #1 question is how do I use this, right? And being really good to help educate physicians, and this isn't necessarily always the HeartFlow team educating, this is peer-to-peer via other physicians, and we use a lot of our DECIDE physicians in that regard.

So that's the #1 question. And then it's an evergreen need, so to speak, but workflow matters. I mean everything we know is it's got to fit in the workflow. They've got to have confidence in reimbursement being there. We've talked previously, this is a diagnostic pathway. Pathways get turned on. Patients do not get cherry picked based on their coverage and then sent down. That's not how it works at scale. So we assumed that was the case going in with Plaque, and we've certainly validated that since then. And that's really why I'm guiding to tail end of '26 on this. I think it's going to take that long to both educate the market and then get to a coverage threshold that's material enough that a health system will actually turn on that pathway.

Our next question comes from the line of William Plovanic from Canaccord Genuity.

A couple here, got and clean up. You guys raised a lot more money at a light higher price on your IPO very successful. And what has this done in terms of changing your go-to-market strategy or your time lines with the extra capital?

Yes. Thanks for the question, Bill. Overall, I'd say we are still committed to the plan we had outlined at the road show. What the IPO did provide us is a tremendous amount of operating flexibility to invest opportunistically down the P&L. We're highly confident this capital provides us enough cushion to ultimately get the company to profitability while allowing us to continue investing. And our OpEx philosophy is really anchored in those strategic themes that John had talked about in his prepared remarks, driving commercial adoption, advancing the technology platform and ultimately leading with clinical evidence.

So overall, plan stays the same, allows us flexibility to invest opportunistically, and we remain confident in getting this company to profitability within 3 years of IPO.

Okay. And then just we've seen a lot of data come out recently, and you've highlighted it through some PR. We've been in a lot of the medical meetings. But is there anything that you'd call out that you think moves the needle on adoption in any of the certain patient populations? And same question, kind of anything in the next 6 months that maybe this is better than expected or it's different or it just really opens anything up or like you said, gets payers going? Anything that came out that can really move the needle that we should focus on?

Yes. I mean -- thanks, Bill. Good question. And again, good to hear your voice. First thing I'll say on clinical, and we have a long history of investing in clinical data. We really believe it's the currency of the realm and you need to have high-quality clinical data in order to move the needle, as you say. We've invested. We've got over 600 peer-reviewed publications. That number continues to grow, and we're going to keep that commitment moving forward. You'll see focused investments to increase more clinical in our Plaque Analysis. We'll have new clinical insights. I think what we're doing with the Navigator study is an example of that.

And then ultimately, we'll expand into new indications. I think most recently at AHA, I can point to the data that we presented there. And again, this -- you can put this in the bucket of helping to educate the market on plaque. But we presented our HeartFlow Plaque staging data, and this is from the U.K., it's our FISH&CHIPS dataset. It was a late-breaking session, full house, very well attended. And this showed some pretty important insights, staging plaque, meaning breaking down plaque into the component cohorts is really important. And we believe there's a paradigm shift that's underway in managing coronary artery disease, very analogous to how oncology went through with cancers. I mean not all cancers carry the same risk and therefore, warrant the same treatment.

The same holds true for coronary artery disease. So our plaque staging data, this is 8,000 patients with a 3-year follow-up. This is the largest plaque staging data to date. It's the largest plaque staging data that exists. And it showed HeartFlow Plaque helps predict risk of heart attack and death. There's 4 stages, the extensive stage, so call it Stage 4 [ or ] extensive, that shows a 5x increase with someone with no plaque. And so what this does is this informs a physician, that patient deserves pretty aggressive medical therapy.

So gone are the days of peanut butter therapy, we can do it much more precisely and personalized now with this data. Now how market moving that is in a given month versus 2 or 3 years, change takes time. And as I mentioned, you need to get it out there and we'll get it out there and educate on it. But that's a really important one on our plaque journey.

And then if I could sneak one more in. You have had a lot of new -- surprising to us, I mean, a lot of product iterations kind of come out since the IPO. I mean you guys have really done a good job there. Anything that we should kind of hone in on? Are these quality of data improvements? Is it workflow improvements? I mean what are you really solving with some of these new products?

Yes. Well, listen, I mean, our goal and our plan is to have a really strong pipeline of innovation that we can continually and predictably bring to our physician partners. Vikram talked about use of IPO proceeds. This is certainly one of them, okay? And we've got work being done relative to new products. We have work being done relative to improving automation. This is our AI-aided automation that expands gross margins. We've got work being done relative to workflow efficiency. You can never work efficiently enough if you're in a health system. And everything we do will correspondingly have clinical data that backs it.

So that's another use of our investments. So we'll share the news of the new products as they come, but those are kind of the buckets that we expand in. And again, we've got this proprietary dataset. We've got over 110 million annotated images, CT images, and we use that. We use that for our second-generation Plaque Analysis that we discussed. We use that in a large extent to our Navigator tool that we just talked about. And so we keep leaning into that, and we're going to take the proceeds from the IPO and hopefully be an innovation machine moving forward.

Thank you. This concludes today's conference call. At this time, I'm showing no further questions. Thank you for participating. You may now disconnect.