Scroll Top

Why AI-based computational pathology detects more cancers

Chances are you or someone you love has had a biopsy to check for cancer. Doctors got a tissue sample and they sent it into a pathology lab, and at some point you got a result back. If you were lucky, it was negative and there was no cancer. But have you ever wondered exactly what happens in between those steps? Until recently, it’s been a meticulous but imperfect manual process where a pathologist would put a thin slice of tissue under a high-powered microscope and examine the cells by eye, looking for patterns that indicate malignancy. But now the process is going digital—and growing more accurate.

Harry’s guest this week is Leo Grady, CEO of, Paige AI, which makes an AI-driven test called Paige Prostate. Grady says that in a clinical study, pathologists who had help from the Paige system accurately diagnosed prostate cancer almost 97 percent of the time, up from 90 percent without the tool. That translates into a 70 percent reduction in false negatives—nice odds if your own health is on the line. This week on the show, Grady explains explain how the Paige test works, how the company trained its software to be more accurate than a human pathologist, how it won FDA approval for the test, and what it could all mean for the future of cancer diagnosis and treatment.

Please rate and review The Harry Glorikian Show on Apple PodcastsHere’s how to do that from an iPhone, iPad, or iPod touch:

1. Open the Podcasts app on your iPhone, iPad, or Mac.

2. Navigate to The Harry Glorikian Show podcast. You can find it by searching for it or selecting it from your library. Just note that you’ll have to go to the series page which shows all the episodes, not just the page for a single episode.

3. Scroll down to find the subhead titled “Ratings & Reviews.”

4. Under one of the highlighted reviews, select “Write a Review.”

5. Next, select a star rating at the top — you have the option of choosing between one and five stars.

6. Using the text box at the top, write a title for your review. Then, in the lower text box, write your review. Your review can be up to 300 words long.

7. Once you’ve finished, select “Send” or “Save” in the top-right corner.

8. If you’ve never left a podcast review before, enter a nickname. Your nickname will be displayed next to any reviews you leave from here on out.

9. After selecting a nickname, tap OK. Your review may not be immediately visible.

That’s it! Thanks so much.

Full Transcript

Harry Glorikian: Hello. I’m Harry Glorikian. Welcome to The Harry Glorikian Show, the interview podcast that explores how technology is changing everything we know about healthcare.

Artificial intelligence. Big data. Predictive analytics. In fields like these, breakthroughs are happening way faster than most people realize.

If you want to be proactive about your own health and the health of your loved ones, you’ll need to learn everything you can about how medicine is changing and how you can take advantage of all the new options.

Explaining this approaching world is the mission of my new book, The Future You. And it’s also our theme here on the show, where we bring you conversations with the innovators, caregivers, and patient advocates who are transforming the healthcare system and working to push it in positive directions.

Chances are you or someone you love has had a biopsy to check for cancer.

Doctors got a tissue sample and they sent it into a pathology lab, and at some point you got a result back. If you were lucky it was negative and there was no cancer.

But have you ever wondered exactly what happens in between those steps?

Well, until recently, it’s been an extremely meticulous manual process.

A pathologist would create a very thin slice of your tissue, put it under a high-powered microscope, and examine the cells by eye, looking for patterns that indicate malignancy.

But recently the process has started to go digital.

For one thing, the technology to make a digital scan of a pathology slide has been getting cheaper. That’s a no-brainer, since it makes it way easier for a pathologist to share an image if they want a second opinion.

But once the data is available digitally, it opens up a bunch of additional possibilities.

Including letting computers try their hand at pathology.

That’s what’s happening at a company called Paige AI, which makes a newly FDA-approved test for prostate cancer called Paige Prostate.

The test uses computer vision and machine learning to find spots on prostate biopsy slides that look suspicious, so a human pathologist can take a closer look.

So why should you care?

Well, in a clinical study that Paige submitted to the FDA, pathologists who had help from the Paige system accurately diagnosed cancer almost 97 percent of the time, up from 90 percent without the tool.

That translates into a 70 percent reduction in false negatives. At the same time there was a 24 percent reduction in false positives.

I gotta tell you, if I were getting a prostate biopsy, I’d really like those improved odds.

And it’s a great example of the kinds of AI-driven medical technologies that I write about in The Future You, which is now available from Amazon in Kindle ebook format.

So I asked Paige’s CEO, Leo Grady, to come on the show to explain how the test works, how Paige trained its software to be more accurate than a human pathologist, how the company got the FDA to give its first ever approval for an AI-based pathology product, and what it could all mean for the future of cancer diagnosis and treatment.

Here’s our conversation.

Harry Glorikian: Leo, welcome to the show.

Leo Grady: Hi, Harry. Glad to be here.

Harry Glorikian: Yeah. You know, I’ve been watching the company for some time now, and the big story here seems to be that we’re really entering the area of digital pathology, also known as sort of computational pathology, and it’s funny because I’ve been talking about digital pathology since I think I started my career back when I was 25, which seems like a long time ago at this point. But for a lot of laboratory tests that we use, like it’s usually done by eye, and now we can get a lot from sort of AI being assistive in this way. So keeping in mind that some of the listeners are professionals, but we have a bunch of sort of non-experts, could you start off explaining the term maybe computational pathology and summarize where the state of the art is, which I assume you guys are right at the cutting edge of it?

Leo Grady: Yeah, so I think it actually might help just to jump back a level and talk about what is pathology and how is it done today? So today, so pathology is the branch of medicine where a doctor is taking tissue out of a patient through a biopsy, through surgery and making glass slides out of that tissue, looking at it under a microscope in order to make a diagnosis. And today, all of that process of taking the tissue out, cutting it, staining it, mounting it on slides. Then gets looked at under a microscope by a pathologist to make a diagnosis, and that diagnosis the pathologist makes is the definitive diagnosis that then drives all of the rest of the downstream management and care of that patient. When pathologists are looking through a microscope, sometimes they see something that they’re not quite sure what it is. And so they may want to do another test. They may want to do another stain. They may want to cut more out of the tissue, make a second slide. Sometimes they want to ask a colleague for their opinion, or if they really feel like they need an expert opinion, they may want to send that case out for a consultation, in which case the glass slides or are put in a, you know, FedEx and basically shipped out to another lab somewhere. All of those different scenarios can be improved with digital pathology and particularly computational pathology and the sort of technology that we build at Paige. So in a digital world, what happens instead is that the slides don’t go to the pathologist as glass. They go into a digital slide scanner, and those slide scanners produce a very high resolution picture of these slides.

Leo Grady: So these are quarter-micron resolution images that get produced of each slide. And then the pathologist has a work list on their monitor. They look through those those cases, they open them up and then that digital workflow, they can see the sides digitally. When they have those slides digitally, if they want to send them out to a second opinion or or show them to a colleague, it’s much easier to then send those cases electronically than it is to actually ship the glass from one location to another. Once those slides are digital it, it opens up a whole other set of possibilities for how information can come to the pathologist. So if they want additional information about something they see in those slides, rather than doing another stain, doing another cut, sending for a second opinion, what we can do and what we do at Paige is we we identify all the tissue patterns in that piece of tissue, match those against a large database where we have known diagnoses and say, OK, this case, this pattern here has a high match toward to something that’s in this database. And by providing that information to the pathologists on request that pathologists can then leverage that information, integrate it and use it in their diagnostic process. And this is the product that the FDA just approved. It’s the first ever AI based product in pathology that is specifically aimed at prostate cancer and providing this additional information in the context of a prostate needle biopsy.

Harry Glorikian: Well, congratulations on that. That’s, you know, that’s amazing. And I’m. You know, the fact that the FDA is being more aggressive than I remember them being in the past is also a great thing to see. But, you know, we’ve been talking and quote digitizing things in pathology for for quite some time, let’s say, separate from the AI based analytics part of it moving in that direction. What was the kind of technology advance or prerequisite that you guys came up with when you started Paige that that took this to that next level.

Leo Grady: Well, as you’re pointing out, Harry, most slides are not digitized today, single digits of slides in a clinical setting get digitized. And the reason for that has been you need to buy scanners, you need to change your workflow, you need to digitize these slides. They’re enormously large from a file size and data complexity. So then you have to store them somehow and you make all of that investment and then you get to look at the same slide on a monitor that you look at under a microscope. And so pathologists for years have said, why? Why would we make this investment? Why would we go through all of that expense? And that trouble and that change and learn a new instrument when we don’t really get a lot of value out of doing so? And furthermore, there was even a question for a long time, do you get the same information on a digital side that you get on glass through a microscope? Yep. There have been a number of things that have been changing that over time. So one is the maturity of the high capacity digital side scanners. There are now a number of hardware vendors that produce these. Storage costs have come down. And one thing that we offer at Paige is is cloud storage, which is really low cost because we’re able to effectively pool costs with the cloud providers from multiple different labs and hospitals, so we can really drive those prices down as far as possible.

Leo Grady: So that lowers that barrier. And then back in 2017, the first digital side scanner got approved, which demonstrated there was equivalency in the diagnosis between looking at the slide on a monitor and looking at it under a microscope. And that is something that that we also replicated with our digital side viewer, demonstrated that equivalency between digital and glass. But all of those barriers were barriers just to going digital in the first place. And now, really, for the first time, because of the maturity of the scanners, because of the FDA clearance of just the viewer, because of lower cost storage, many of those barriers have come down. Now what has not happened is still a major clinical benefit for going digital in the first place. Yes, you can share slides easier. Yes, you can retrieve slides easier. Yes, you can do education easier. It’s still a lot of cost and a lot of changed your workflow, so I really think that that the introduction of the kinds of technologies that that the FDA approved, which we built with Paige Prostate, that actually adds additional information into the diagnostic workflow that can help pathologists use that information help them. You get to a better diagnosis, reduce false positives, reduce false negatives, which is what we showed in the study that for the first time is is going above and beyond just going digital and some of these conveniences of a digital workflow to providing true clinical benefit.

Harry Glorikian: Yeah, I mean, whenever I look at this from an investment perspective, like if you take apart something and break it into its first principles, you know, levels, you have to have certain milestones hit. Otherwise, it’s not going to come together, right? And I’ve, you know, looking at digital pathology, it’s the same thing. You have to have certain pieces in place for the next evolution to be possible, because it’s got to be built on top of these foundational pieces. But, you know, once you get there, the exponential nature of of how things change, once it’s digitized and once you’re utilizing it and prove that it works is sort of where you see the, you know, large leaps of benefit for the pathologist as well as, you know, ultimately we’re doing this for better patient care. But you know, your product was I think the FDA called it the first ever FDA approval for an AI product in pathology, which is a big deal, at least as far as I’m concerned, because I’ve been doing it for a long time. But because it was first, it must have been a one hell of a learning process for you and the FDA to figure out how to evaluate a test like this. Can you sort of explain maybe a little bit about the process? You know, how did you win approval? What novel questions did you have to answer?

Leo Grady: It was a long process. You know, as you point out, this is this is the first ever technology approved in this space. And I think you saw from the FDA’s own press release their enthusiasm for what this technology can bring to patient benefits. Fortunately, we applied for breakthrough designation back in early 2019, received that breakthrough designation in February of 2019. And as a result, one of the benefits of breakthrough designation is the FDA commits to working closely with the company to try to iterate on the study protocol, iterate on the the validation that’s going to be required in order to bring the the technology to market. And so because of that breakthrough designation, we had the opportunity to work with the the FDA in a much tighter iterative loop. And I think that they are they were concerned, I mean, primarily about the impact of a misdiagnosis and pathology, right? Which is really understandable, right? Their view is that, yes, maybe in radiology, you see something and maybe aren’t totally sure. But then there’s always pathology as a safety net, you know, in case you ever really need to resolve a ground truth. You can always take the tissue out and look at it under a microscope. But when you’re dealing with a product for pathology, that’s the end of the road. I mean, that is where the diagnostic buck stops. And so anything there that that was perhaps going to misinform a pathologist, mislead them, you know, ultimately lead to a negative conclusion for the patients could have more severe consequences.

Speaker2: The flip side, of course, though, is that if you get it right, the benefits are much greater because you can really positively impact the care of those patients. So I think they they, you know, appropriately, we’re concerned with the exacting rigor of the study to really ensure that that this technology was providing benefit and also because it was the first I think they wanted to be able to set a standard for future technologies that would have to live up to the same bar. So there were a lot of meetings, you know, a lot of trips down to Silver Spring. But I have to say that that the FDA, you know, I think in technology, there are a lot of companies that are are quick to, you know, malign regulators and rules. I frankly both at Paige and my previous experience at HeartFlow, at Siemens, I think the FDA brings a very consistent and important standard of clinical trial design of of, you know, technology proving that is safe and effective. And I found them to be great partners to work with in order to really identify what that protocol looks like to be able to produce the validation and then to, you know, ask some tough questions. But that’s their job. And I think, you know, at the end of the day, the products that get produced that go through that process really have met the standard of of not only clinical validation, but even things like security and quality management and other really important factors of a clinical product.

Harry Glorikian: Oh no, I’m in total agreement. I mean, whenever I’m talking to a company and they’re like, Well, I don’t know when I’m going to go to the agency, I’m like, go to the agency, like, don’t wait till the end. Like there, actually, you need to look at them as a partner, not as an adversary.

Leo Grady: Yeah. And a pre-submission meeting is is easy to do. It’s an opportunity to make a proposal to the FDA and to understand how they think about it and whether that’s that’s going to be a strategy that’s going to be effective and workable for them. So I always think that that pre subs are the place to start before you do too much work because you generally know whether you’re on the right path or not.

Harry Glorikian: Yeah, I agree. And it’s funny because you said, like, you know, they’re concerned about the product, but it’s interesting. Like from all the College of American Pathology studies where you send slides to different people, you don’t always get the exact same answer, depending on who’s looking at it. So I can see how a product can bring some level of standardization to the process that that helps make the call so uniform, even across institutions when you send the slides. So I think that’s moving the whole field in a really positive direction.

Leo Grady: Well, only if that uniform call is correct, right? Or better? Great. I mean, if you bring everybody down to the lowest common denominator that that standardization, but it’s not moving the field forward. So. Correct. One of the curses of of bringing that level of standardization is that you have to really meet the highest bar of the highest pathologists and not not just the average. That said, you know, we’re fortunate to come from Memorial Sloan-Kettering and to have the opportunity to work with some of the the leading pathologists in the world to really build in that level of rigor and excellence into the technology.

Harry Glorikian: Yeah. So that brings me to like, you know. The algorithms are built on a fairly large training set would be my assumption and of pre labeled sort of images, where do you guys source that from? Is it you have like a thousand people in the background sort of making sure that everything is labeled correctly before it’s fed to the to the algorithm itself?

Leo Grady: Well, what you’re describing is very common where you have pathologists or in radiology radiologists or other experts really marking up images and saying this is the important part to pay attention to. This part is cancer. That part’s benign. Our technology actually works differently. Our founder, Thomas Fuchs, and his team at Memorial Sloan-Kettering actually really made a breakthrough not only in the the quality of some of the the AI systems that were building, but also in the technology itself. And what what they did, this was all published in Nature Medicine a couple of years ago, is basically find a way to just show the computer a slide and the final diagnosis without having a pathologist, you know, mark up the slide, but just show them the final diagnosis. And when you show the computer enough examples of the slide and the final diagnosis, the computer starts to learn to say, OK, this pattern is common to all grade threes. This pattern is common to all grade fours. Or whatever it is. And the computer learns to identify those patterns without anybody going through and marking those up. Well, this technology is important for a few reasons.

Leo Grady: One, it means we can train systems at enormous scale. We can not just do thousands of cases, but tens of thousands, hundreds of thousands of cases. Second, it means that we can really build out a portfolio of technologies quickly that are very robust and not have to spend years annotating slides. And third, it allows us to start looking for patterns that no pathologists would necessarily know how to mark up. You know, can we identify which tumors are going to respond to certain drugs or certain therapies? You know, no pathologists are going to be able to say, OK, it’s this part of the the tumor that you need to look at because they don’t really know. But with this technology where we we know these tumors responded, these tumors didn’t it actually helps us try to ferret out those patterns. So that that’s one of the real key benefits that differentiates Paige from from other companies in this space is just the difference in the technology itself.

Harry Glorikian: Yeah. I mean, it’s funny because I must admit, like when we talk about stuff like this, I get super excited because I can see where things can go. It’s. It’s always difficult to explain it where somebody else can envision what you’ve been thinking about because you’ve been thinking about it so long, but it’s super exciting. So let’s jump to like the most important benefits, like if you had to rank the benefits of the technology, I mean, I’ve I read on your website that in the clinical study you guys submitted to the FDA, pathologist used using the Paige Prostate were seven percent more likely to correctly diagnose the cancer. Is that the major innovation? Would that by itself be enough to justify an investment in the technology? I mean, I’m trying to. You know, if you were to say God, this is the most important thing and then go down the list, what would they be?

Leo Grady: Yeah, that’s right. So so the study that we did was like this. We had 16 pathologists. They diagnosed about six hundred prostate needle core biopsy patients and they they did their diagnosis. They recorded it and then they did it a second time using Paige so they could see the benefit of all this pattern matching that that Paige had done for them. And what we did is we compared the diagnosis. They got the first time and the second time with the ground truth, consensus diagnosis that we had from Memorial. And what we found is that when the pathologists were using Paige, they had a 70 percent reduction in false negatives. They had a 24 percent reduction in false positives, and their interest in obtaining additional information went down because they had more confidence in the diagnosis that they were able to provide. And what was interesting about that group of 16 pathologists is it it included pathologists that were experienced, that were less experienced, some that were specialists in prostate cancer, some that were not so specialized in prostate cancer. And among that entire group of pathologists, they all got better. They all benefited from using this technology. And what’s more, is that the gap between the less experienced, less specialized pathologists and more experienced, more specialized pathologists actually decreased as they all used the technology. So it allowed them to, like we were talking about before, actually come up to the level of of the better pathologists and even the better pathologists could leverage the information to get even better.

Harry Glorikian: So as a male who you know who’s going to age at some point and potentially have to deal with, hopefully not, a prostate issue, we want them to make an accurate diagnosis because you don’t want the inaccurate diagnosis, especially in in that sort of an issue. But what about the speed? I mean, you’ve you talk about that, you know, it helps streamline the process and reduce reduce turnaround time for for patients. What does that do to workload and and how quickly you’re able to turn that around compared to, say, a traditional method.

Leo Grady: Our study was really focused on clinical benefit and patient benefit. We were not aiming to measure speed and the way in which the study was designed and the device is intended to be used is that the pathologist would look at the case, decide what they they think the result is, and then pull up the Paige results and see if it changes their thinking or calls their attention to something that they may have missed. So the focus of the the product was really on the the benefit to the the clinical diagnosis and the clinical benefit to patients by providing more information to the doctors. And the result of that information was, you know, clearly demonstrated benefit. Now if they can get to that result by looking at the Paige results and they don’t need another cut, they don’t need another stain, they don’t need another consultation, then that’s going to get the results back to the urologists faster, back to the patient faster and will ultimately enable them to start acting on that diagnosis more quickly. But the intention of the study, the intended use of the device is not around making pathologists faster. It’s really around providing them this additional information so that they can use that in the course of their diagnosis and get the better results from patients.

[musical interlude]

Harry Glorikian: Let’s pause the conversation for a minute to talk about one small but important thing you can do, to help keep the podcast going. And that’s to make it easier for other listeners discover the show by leaving a rating and a review on Apple Podcasts.

All you have to do is open the Apple Podcasts app on your smartphone, search for The Harry Glorikian Show, and scroll down to the Ratings & Reviews section. Tap the stars to rate the show, and then tap the link that says Write a Review to leave your comments. It’ll only take a minute, but you’ll be doing us a huge favor.

And one more thing. If you like the interviews we do here on the show I know you’ll like my new book, The Future YouHow Artificial Intelligence Can Help You Get Healthier, Stress Less, and Live Longer.

It’s a friendly and accessible tour of all the ways today’s information technologies are helping us diagnose diseases faster, treat them more precisely, and create personalized diet and exercise programs to prevent them in the first place.

The book is now available in Kindle format. Just go to Amazon and search for The Future You by Harry Glorikian.

And now, back to the show.

[musical interlude]

Harry Glorikian: So I asked this out of naivete because I didn’t I didn’t go looking for it. But have you guys done a health economic analysis of the system?

Leo Grady: We have one. It certainly it’s, as you know, it’s really key to be able to look at that we have a model that we’ve built. We’re still refining it with additional data. There was a study that was announced in the U.K. a couple of weeks ago where the NHS is actually funding a prospective multicenter trial that includes Oxford, Warwick, Coventry, Bristol to be able to evaluate the the health, economics and clinical benefits of using this technology in clinical practice prospectively. So that’s something that we engaged with NICE [the National Institute for Health and Care Excellence] on in order to try to get the design correct that will help feed in real world data into the model. But we have a model that we’ve been using internally and are continuing to build and refine.

Harry Glorikian: So. Again, incredible that you guys got FDA approval, I think the company was founded in 2017, if I’m correct. Can you talk about, you know, the founders and yow you guys, you know, built this so quickly, I mean time scale wise, it’s a pretty compressed time scale, relatively speaking.

Leo Grady: Well, yeah, it isn’t, it isn’t, …so the company started in 2017, our first employee was actually middle of 2018 and we had our first venture round and in early 2018. However, the work that went into the company that spun out of Memorial Sloan-Kettering started earlier. So there is a group of really visionary individuals at MSK that back, I want to say, 2014, 2015, actually had started this push toward digital pathology, computational pathology, really seeing where the puck was going and building this technology. They formed something called the Warren Alpert Center, and the Warren Alpert Center provided some initial funding to really get this going and to hire some of the founders and to really move this technology in the right direction. And it was really because that technology started to show such promise that MSK made the decision that that was at a point where it could be better, you know, more impactful to actually go outside of MSK into a company where where we could industrialize the technology and really bring it to hospitals and labs around the world. So the technology started earlier, 2014, 2015. Paige was really launched in, I would say, 2018, although technically it was incorporated earlier and and then from that point I personally joined in 2019. And so I’m not I’m not a founder, but when I joined in 2019, you know, we we really spun up a significant team and and brought to bear some of my own experience and industrializing AI technology and bringing it out to clinical benefit.

Harry Glorikian: Well, you know, most founders don’t take the company all the way. It’s a rare breed that’s able to get it that far. So you know this a great story, but let’s step back here and talk about like now you have to like, get people to accept this technology right, which is the human factor which I always find much more confounding than the the the the computational factor. So you’ve got to get, you know, somebody inside a hospital or pathology lab. Do you run into resistance or pushback from the technology, I mean, are they skeptical about the algorithm? How do you get a human to sort of buy off on this? I remember when we were presenting this, oh God, again, 25 years ago, they hated it. I mean, just hated it. And as time has gone by, you’ve seen that that digitization is slowly taking effect and where you know, it’s assistive as opposed to something, I remember when we first launched this, it was, “This is going to be better than” or “take your job,” which is a great way to make an enemy on the other side. And I see that the two actually being better than one or the other per se on on its own. So how are you guys approaching this? And do you have any anecdotal stories that you might be able to share?

Leo Grady: Yeah, and so I think there are two elements are one is, you know. Are people resistant by the nature of the technology because they feel threatened by it, and then the other is how does market adoption start with this sort of technology to just the first point? You know, I tend to be very careful about the term AI. I feel like it know it often introduces this concept of, you know, people think of a robot doctor that’s going to run in and start doing things. And it’s just it’s not. I mean, AI is a technology that’s been in development for four decades. I did my PhD in AI, in computer vision, 20 years ago, and it’s just a technology, right? It’s like a transistor. It can be used to build many different things. At its core, it’s just complex pattern matching, which is what we how we leverage that technology. In the case of Paige Prostate was to help provide that information. I think, you know, the better frame to think about this technology is as a diagnostic. This is just like a diagnostic test. You validate it with a standalone sensitivity and specificity. The information gets provided the doctor. You have to do a clinical trial that samples the space effectively of the patient population and the intended use.

Leo Grady: And you have to make sure the doctors understand the information and know how to use it effectively. It’s before my time, but I heard that when immunohistochemistry was first really introduced in pathology, that there is a discussion that this was going to take all the pathologists’ jobs. And who needs a pathologist if you can just stain with IHG and get get a diagnostic result out of it? Well, you know, 20 years, IHT is an essential component of of pathology, and it’s a key element of of the diagnostic workflow for pathologists. So, far from replacing any pathologists, it’s empowered them. It’s made there the benefit that they can provide to the clinicians, even more valuable and even more important. And I think we’re going to see a similar trajectory with this computational technology. Now your first question about market adoption, how people adopting this, I would say that, you know, last week I went to the College of American Pathology meeting, which was in person in Chicago. It’s my first in-person meeting since COVID, so a year and a half ago. And I noticed–and this was this was right after the announcement by the FDA of of the approval for Paige Prostate–I noticed there was a market shift in the conversations I was having with pathologists.

Leo Grady: It was a shift away from “Does this technology work? Is it ready for prime time? What does it really do?” Toward, “Ok, how do we operationalize this? How do we bring it in house, how do we integrate this into a workflow and how do we how do we pay for it?” You know, those are the conversations that we were having in Chicago at CAP. Not does this work? Is it ready for prime time? So I do think that there is a market understanding that the technology is real, that it works, that it can provide benefit. Now it’s just a question of how do we operationalize and how do we get it paid for? Because today there’s no additional reimbursement for it. But you know, again, with market adoption, you’re got your Moore adoption curve for anything. You get them and you get your innovators and early adopters, your early majority, late majority and your laggards. And you know where I think we’re at a stage where we’ve got innovators and early adopters that are excited to jump in and start leveraging this technology. And I think, you know, we’re going to get to your early majority and the late majority over time. It’s always going to be a process.

Harry Glorikian: Yeah, no. I mean, you know, reflecting on your IHC [immunohistochemistry], that’s where I started my career. Like, I think I taught like two hundred and fifty IHC courses over the first, say, three or four years that I was in the in the business. Three or four years. And you know, I agree with you. There’s no way that any one of these technologies takes the place of [a pathologist]. They’re additive, right? It’s just a tool that helps. Make the circle much more complete than it would be in any one component, all by itself.

Leo Grady: Could you ever hear when you were teaching these classes? Did anyone ever say that like, are we even going to need pathologists anymore?

Harry Glorikian: No, it was when the is is when imaging systems came out that said the imaging system would then replace the pathologists. The IHC was was really the cusp of precision medicine, where I remember when I first started because we were working with ER and PR and, you know, when I first learned, you know about like, you know, the find and grind method, I would always be like, OK, it’s x number of femtomoles. Like, What does that really telling you, right? Compared to this stain over here where I can see, you know, the anatomy, I can see where the cells are. I can see. I mean, there’s so much more information that’s coming from this that lets me make a better call. I will tell you selling it was not that hard to a lot of people, they they could see the benefit and you could you could really sort of get them to adopt it because they saw it as a tool.

Leo Grady: Was that post-reimbursement?

Harry Glorikian: Uh, even pre-reimbursement.

Leo Grady: Really interesting. Yeah, there’s there’s a lot we can learn from you then.

Harry Glorikian: Yeah, it was. It was. It was an interesting ride back then. I mean, I remember my first day at work. My boss comes to me and says. By the way, you’re going to give a talk in Arizona in two weeks, and I was like, What do you mean I’m going to go? Who am I going to give a talk to you? He goes, Oh, you got to give a talk on the technology and how to use it. And I said, who’s in the audience? And he said histo techs, and there’ll be some pathologists. And I was like, Are you kidding me? And he goes, You got two weeks to get ready. Oh my God, I was cramming like crazy. I was in the lab. I was doing all the different types of assays that we had available. And you know, it was you went out there and I learned very quickly like, the show must go on, like you got to get out there and you got to do your thing. But it was it was a great time in my career to be on that on that bleeding edge of what was happening. So quickly, like, why did you guys start with prostate cancer, though like? It’s not the most common cancer, although it’s high on the list, so. Or maybe it’s the second most type of cancer, but why did you guys start with that and where do you guys see it going from there, I guess, is next.

Leo Grady: Well, the the decision of how to rank the different opportunities for, you know, ultimately we believe this technology can benefit really the entire diagnostic process, no matter what the question is in pathology. However, we did have to prioritize right and elements of of where to start, right. The elements of prioritization had a few factors. So one factor was how how prevalent is the disease? I mean, as you know, prostate cancer is one of the big four. Second, is there are a lot of benefit that we can provide today with prostate cancer. You know, man of a certain age goes in, gets a PSA test. It’s high, they go and they get 12 cores, 14 cores, 20 cores out of their prostate and that produces. You know, it can be 30 slides, it can be 50 slides, I mean, it really depends, and this can take the pathologist a long time to look through. Most of those cores are negative. In fact, most of those patients are negative, but the consequence of missing something is really significant. And so we felt that this was a situation where there was a big need. There’s a lot of there’s a lot of screening that goes on with prostate cancer. Prostate cancer is prevalent and the consequence of missing something is really significant. So that’s where we felt like we could provide maximum benefit, both in terms of the patient, in terms of the doctor, and also that it was a significant need across the space.

Leo Grady: We also had the data and the technology that we could go after that one well. But that said, you know, we announced that we have a breast cancer product that is got a CE mark in an enabling clinical use in Europe. We’re doing a number of investigational studies with that product in the US right now and and working toward bringing that one to market. You know, after our our recent funding round, we spun up a number of teams and a number of of verticals that were we’re going after in other cancer types and ultimately even beyond cancer. So there’s more to come. We wanted we really take seriously the quality, the regulatory confirmation as well as the deployment channel. I mean, we built the whole workflow to be able to leverage this technology throughout the workflow in a way that is meaningful to the pathologist. So the development is is maybe a little bit more heavy and validation than some other companies where you have a PhD student that says, Oh, you know, I won some challenge and I went to go bring this to market building real clinical products, validating them, deploying them, supporting them is a real endeavor. But prostate was just the first, breast is second, and we have a whole pipeline coming out. So stay tuned.

Harry Glorikian: So before we end here, I want to just tilt the lens a little bit towards the consumer and say, like, you know. Why would consumers show interest or at least be aware that these things are coming? Because I always feel like they’re almost the last to know, or they just don’t know at all. But, you know, in the future, you know, with technologies like this, do you see it identifying tumors sooner, faster, more accurately? Or, you know, will it will it help increase survival or help us find better drugs? I mean that that’s I think, what people are really… If you went down one level from us of the people that are affected by this. Those are the sorts of things they’d want to know.

Leo Grady: Well, I think, you know, a useful analogy is what happened with the da Vinci robot. You know, when it was necessary for a patient to get prostate cancer surgery, they often chose centers that had the da Vinci robot. Why? Because they believed that they were able to get better care at those centers. And it’s not because the surgeons at the other centers were no good. It’s because the the da Vinci added elements of precision and standardization and accuracy that could be demonstrated that would enable the the patient to feel more confident they’re getting the best treatment at those centers. So as I think about Paige Prostate and and ultimately the other technologies that we’re bringing to market behind that, I would imagine that from the standpoint of the patient, they would want the diagnosis done at a lab where they had access to all of the available information, all the latest technology that could inform the pathologists to get the right answer, right? So would you want to go to a lab where the pathologists had no access to IHC? Would you want to send it to a lab where the pathologist had no ability to do a consultation? Do you want to send your your sample to a lab where the pathologist doesn’t have access to Paige? I think in the future the answer is going to be no.

Leo Grady: And I think that we’re going to see ultimately, insurance companies and Medicare recognize that those labs are able to provide better care to patients and are going to encourage them and incentivize them to adopt these technologies. So, you know, ultimately from a patient standpoint, they they want to choose centers where they’re going to get the best care, they’re going to get the best diagnosis. I think one of the exciting elements of digital technology is that not everybody is able to go to Memorial Sloan-Kettering, not everyone’s able to go to MD Anderson or Mayo Clinic. I think the opportunity with digital technology is to really increase the accessibility and increase the availability of these diagnostic tools that can really empower and enable pathologists in many parts of America, as well as beyond to really get to better results for their patients. And ultimately, you know, every patient cares about getting those those results accurately for themselves and for their loved ones.

Harry Glorikian: Yeah, I mean, I’m always explaining, you know, to different people like once you digitize it, there’s so many opportunities that may open up to make things better, faster, easier, more accurate and even start to shift the business model itself of what can be done and where it can be done. So it’s it’s a super exciting space, and thanks for taking the time. It was great to talk to you. I mean, I don’t get to talk to people in pathology all the time anymore. I’m sort of all over the place, but it’s it’s near and dear to my heart, that’s for sure.

Leo Grady: Well, thank you so much, Harry. We’re so excited by these recent developments with the first ever FDA approved technology in this space and, you know, really excited to help roll this out to labs and hospitals around the country and around the world to really benefit those doctors and patients.

Harry Glorikian: Excellent. Well, I look forward to hearing about the next FDA approval.

Leo Grady: Working on it. Look forward to telling you.

Harry Glorikian: Thanks.

Leo Grady: All right. Thanks so much, Harry.

Harry Glorikian: That’s it for this week’s episode.

You can find past episodes of The Harry Glorikian Show and MoneyBall Medicine at my website,, under the tab Podcasts.

Don’t forget to go to Apple Podcasts to leave a rating and review for the show.

You can find me on Twitter at hglorikian. And we always love it when listeners post about the show there, or on other social media.

Thanks for listening, stay healthy, and be sure to tune in two weeks from now for our next interview.

Related Posts