Join our host Dr Joe Kosterich with endocrinologist Professor Tim Davis, general surgeon Dr Bill Gong and chemical pathologist Dr Johan Conradie to learn more about how predictive biomarkers are changing our understanding of diabetes-related kidney disease and oesophageal cancer.
Understand how blood-based predictive tests can be used for identifying patients with diabetes at risk of declining kidney function, including those with currently normal renal markers, and also for assessing progression risk in patients with Barrett's oesophagus.
Learn more about the clinical assessment of chronic reflux, appropriate investigation and referral, and the potential role of biomarkers in guiding surveillance of your patients.
Activity ID: 1756740, claim your CPD time here.
------------
This podcast is CPD verified.
------------
This podcast is exclusively for Australian medical practitioners. The content discussed is not appropriate for non-medical professionals. Please listen to our disclaimer provided at the end of the episode.
The information in this podcast is intended to be general information only and is no way intended to be medical advice, treatment or diagnoses. You acknowledge and agree that nothing in this product may be taken to be medical advice, treatment or a diagnosis by us or our podcast guests, nor are we or they intended to be a substitute for consulting a medical practitioner for your particular circumstances and needs. We do not accept any liability for injury, loss or damage incurred by the use of or reliance on the information provided in this podcast. This podcast may include the views or recommendations of third parties and does not necessarily reflect the views of Medical Forum - we assume no legal liability or responsibility for the accuracy, currency, or completeness of the information. If you require immediate medical attention, please contact 000 immediately or seek alternative and appropriate medical services.
[00:00:00] Welcome to the Medical Forum podcast. This is general information for health practitioners only and you can claim the time as part of your CPD. Welcome to the Medical Forum podcast with me, Dr. Joe Kostrich, GP and Clinical Editor of Medical Forum. And today, very pleased to be joined by three guests, Dr. Joe Kostrich, Dr. Matt Clifford and Dr. Johan Conorati.
[00:00:27] Gentlemen, welcome to the podcast and thanks for your time today. Thank you. Thanks, Joe. So we're going to walk across two quite separate areas, but there's a common theme, which is new markers which enable early detection. But before getting into that, perhaps in turn, if you could just for the listeners who may not be familiar with each of you, just tell us a little bit about your background and how you've come to doing what you're currently doing. I might start with you, Tim.
[00:00:55] Yeah, I'm a professor of medicine at UWA and also an endocrinologist and general physician. I work mainly at Fremantle Hospital, but also in private. I'm here, I guess, because I run a research program called the Fremantle Diabetes Study based at Fremantle Hospital that generated the samples and data for one of the tests we're going to talk about today. My name is Bill. I'm a general surgeon.
[00:01:22] I work mainly in private, but I still do a little bit of work at Fremantle Hospital. I see a lot of patients every week with reflux problems. I do a lot of anti-reflux procedures. And yeah, one of the things I'm really excited about is this market that might help us to triage some of these patients. Thanks, Joe. Yeah, I'm a chemical pathologist by train. I used to be a GP a long time ago.
[00:01:47] I work at Western Diagnostics Pathology, which is part of the Helios Pathology Network. And my main interest in this is over the years, because we're doing all the tests, we developed CPD programs for GPs.
[00:02:04] And one of the interesting things in emerging themes is the ability to try and predict the patients that will develop a chronic condition sooner rather than later. And if we can try and identify those patients, we should be able to actually focus the treatment on those patients. And these tests actually helps us do that. Well, that's a really good segue because one of the things we've never really had in medicine is crystal balls.
[00:02:34] We know people with conditions, and we'll start with you, Tim, about diabetes, are at risk that we've never really been able to stratify. Look, who is at greater risk and who is at lesser risk? These markers, and I'll put my own hand up until doing the prep for this. I've not heard of either of them, so I'll be really interested to learn. But, Tim, diabetes-related kidney disease, I mean, that's a major issue still, even in 2026.
[00:03:02] And from what I gather, not getting any less, what does sort of, if like, routine kidney surveillance, if you like, mean? And are we still missing opportunities? Yes, you're absolutely correct. I think that whereas we've seen a reduction in cardiovascular events over the last few decades, what we're seeing is an increase in chronic kidney disease because people are living long enough to get it now.
[00:03:32] And it's an expensive complication. So if you don't detect it and manage it appropriately and you head to renal replacement therapy, you're looking at tens of thousands of dollars a year per patient for the taxpayer to pick up.
[00:03:47] So it's very important to try and get in there and identify the patients very early because even if you're sitting with a patient who has a normal EGFR and urinary ACR, they may be at significant risk of having a rapid renal decline. And you need to know about it in advance so that you can optimally manage that patient. And this test, the ProMarkerD that we're about to talk about, fills that gap.
[00:04:16] So you said even if they've got normal EGFR, they may still be at risk. Do we have any ideas as to why? I suppose we don't. Any ideas as to why that is the case? We know from our free metal diabetes study that I mentioned earlier, longitudinal data, that there are a group of patients who do start with normal renal indices and then have a rapid decline in their EGFR.
[00:04:44] And that spawned the test in that we were wanting to know, is there a blood-based test that can identify these patients early? So we linked with Proteinomics, a local company in Perth, and we screened our patients who did and did not progress rapidly to renal impairment. And we found some proteins in the blood that we could measure, as well as some very simple clinical variables.
[00:05:10] And we generated a risk score, which basically identifies people in that group that, although you might be lulled into a false sense of security, are okay, may go bad in the next four or five years. Well, you touched and mentioned ProMarkerD. So, Johan, tell us a little bit about that as a blood marker and what it can tell us or predict.
[00:05:35] So, with ProMarkerD, as Tim mentioned, they had a whole menu of proteins that they evaluated during the studies and they sort of landed onto the CD5L and ApoE4 proteins, which is not normally measured in general medicine. But we do know that they are affected by renal deterioration.
[00:06:02] So, we will take those two proteins and we add in the renal function, the EGFR, as well as the patient's age. And then that goes into an algorithm that was developed by Proteinomics International and Tim's team. And that will then generate the risk score. But what this risk score basically does is it indicates the patients that are at higher risk.
[00:06:24] And a good example will be what we've just talked about is where a patient might come up with a normal risk based off the EGFR and album creatinine ratio.
[00:06:36] A good proportion of those, and this is where ProMarkerD works quite nicely, it indicates it's got a high accuracy rate and try and indicate those patients that's going to develop renal failure and not renal failure, but the deterioration in the EGFRs in the next four years. So, it's a predictive tool to try and identify the patient that probably needs a bit more attention than your average patient. Okay.
[00:07:02] And the marker itself, you mentioned as a protein, is it produced by the kidney or where does this protein come from? Pardon my ignorance. No, no. These proteins play varied roles in the body and they're usually produced in the liver or when necessary. So, a good example would be that the ApoE4, its role in kidneys is affected by kidney function.
[00:07:30] And it looks like it sort of changes much earlier than the normal would see. So, because that's one of the things that we sort of think about when we think of these patients with normal EGFRs. There's always some pathology going on underneath that we're not aware of and our current methods of using to try and discriminate these patients isn't that sensitive at this point. So, you've got these additional sort of markers that can help identify these patients that could potentially go forward.
[00:07:58] So, I suppose that leads to the next question. So, these markers can help identify who's at risk. Who do I identify that should have this test or is it potentially for everybody with diabetes? Yeah.
[00:08:14] Just to extend some of your hands remarks, the APO lipoprotein compound and also CD5L, they do have some influence on progression of kidney disease themselves. So, for example, CD5L, it's involved in apoptosis within the kidney. So, if you can prevent that, obviously, you preserve kidney function.
[00:08:41] So, they are biologically plausible markers of risk factors for renal decline. Going back to your question, I mean, we – the Cadego, which is the international body that makes recommendations for kidney management, has come up with this so-called heat map, which you may have seen. No, I haven't. Okay. And it plots EGFR against UACR.
[00:09:08] And it's a sort of a square with compartments in it going at top left from green where your EGFR and UACR are okay, the normal, to red down the bottom right-hand corner where both of them are bad. And obviously – and in the middle, there's a whole lot of orange zones which are intermediate dysfunction.
[00:09:37] So, obviously, in the orange zones and the red zones, especially the red zone, you miss the boat. People have already developed complication and there's pointless doing a prognostic test. You know that they need to be managed appropriately.
[00:09:51] The people of interest are in the green and yellow zones in that heat map where you've got the opportunity to identify them early and make sure that their risk factors are managed and they're on renal protective therapies like STLT2 inhibitors and ACE inhibitors and so on.
[00:10:07] So, although in general practice the heat map isn't used, it's certainly one that the specialists, especially renal specialists, hang their hat on in terms of basing their assessment of patients on. Well, I suppose that leads into the next consideration.
[00:10:28] As a GP, is this a test that I should be initiating or should I be talking to patients about it and suggesting they talk with their specialists about it? Well, my suggestion would be is to use that in conjunction with the normal guidelines. I think the important thing is to just state that this is not a replacement for normal. Well, this is to, it's an adjunct.
[00:10:56] So, this is to help identify patients that's at higher risk than the normal population. So, my personal opinion is that where I would like to see this used more often than not is in your patients in the green because those are the ones that still has good kidney function and we can actually prevent if we start identifying these patients earlier and we can actually initiate more aggressive treatments earlier.
[00:11:22] So, the best people to talk to you would be able to talk to a GP and the GP can then either speak to the specialists or give chemical pathologists a WDP at call and we can give them talking through that. So, it can be straight up ordered by a GP or any doctor at this stage. You mentioned at the start, Johan, that there's the Promarker D and it sort of goes through an algorithm.
[00:11:51] So, in terms of the result that I might end up getting, does it give me a number or does it give me a, like say for example, with the CAC scores, a low, intermediate, high risk percentile, what do I end up getting? Yeah, it's a bit of a, a little bit of both. The report itself will give you a number and it's usually in a percentage but it should not be interpreted as that patient. It's a scale from zero to a hundred. Yeah, yeah.
[00:12:18] And basically based on that, they've also then delved into green, orange and red. So, low risk, intermediate risk and high risk. The fact that you're low risk does not mean to say you have no risk. There's always a risk. And the important thing is to realize that these tests are to try and identify patients that you want to act on earlier. But, yeah, so what the doctor will get is they will get a visual interpretation.
[00:12:45] So, green, orange and red or together with a score. The higher the score out of a hundred, the higher the risk to the patient of developing. Oh. And assuming somebody comes back at a low risk, is there an indicative time when you might repeat it? Sort of be like how many, how far apart would one do it if it was low risk? Yeah, yeah, absolutely.
[00:13:11] I mean, so, and with Promocca D, if it's a low score, again, follow the normal guidelines as you would for any diabetic patient. And then, sort of, the intervals would be every four years. Okay. If there's something that concerns you, then obviously consider bringing it sooner. But at the moment, the normal predicts are for every four years.
[00:13:34] And the other thing, I suppose, in my mind is a low score, I suppose, a little bit to sensitivity and specificity. Low scores, one can be – nothing's perfect, I get that. But low scores, one can be reasonably confident in as best we can be in sort of the late 2020s. Because I know things change. Yeah, yeah.
[00:14:00] I mean, it's – for Promocca D, there's a reasonable sensitivity and specificity, right? Absolutely. So, a low score, although, again, it does not exclude progression, early progression, but it does sort of gives you a bit more confidence. I think, just to add, these risk scores, they're for the benefit of the practitioner, but also the patient.
[00:14:29] So, you can show them where they sit. Yeah. And that gives you – it provides a powerful message for them adhering to your recommended method. Absolutely. It can be a motivation tool. And the reason for the four years is that the validation, we had epidemiologically good four-year data. After that, people start to die and not come back to the study and so on.
[00:14:55] And so, it's a pragmatic thing, but you would say every four years is what you should do. Yeah. So, that's sort of based on the data we've got in the models that was developed. And look, if we're sitting here in one or two or three years' time, those recommendations might be different. But I was thinking more, it's not in four years as against annual law or every six months or so. Look, so changing gears, Bill, reflux, they're obviously very, very common.
[00:15:22] I mean, it's one of the most common GI presentations that you see a GP for, right? Yeah. And again, ranging from the very mild to get a bit of heartburn once every couple of months through to the chronic. In terms of management and prediction, tell us a little bit about where we stand with that, I suppose, particularly with Barrett's. And then also, that'll lead us into another program. Yeah, that's right.
[00:15:47] So, I mean, chronic reflux or longstanding reflux, especially patients with kind of refractory symptoms, symptoms non-managed by PPIs, which is the standard care really these days, then those patients are at increased risk of getting baris esophagus with greater than five years' exposure to chronic reflux. I think, you know, managing standard reflux or, you know, patients that you see every day with reflux,
[00:16:14] you know, need to manage their lifestyle kind of risk factors. So, obesity being one of the major problems we see in our society, in our population these days, you know, helping patients with obesity management will definitely improve their reflux control and also lifestyle factors, reducing alcohol. But we do know there is a particular group of patients that are at a more higher risk of getting complications from reflux, like Barrett's, or esophageal cancer.
[00:16:44] They're usually the 50-year-old plus males that have chronic reflux. So, they're the particular population that, you know, we need to screen for. And obviously, reflux symptoms, patients coming in with dysphagia or, you know, aspiration and things like that. So, they're the people that are obviously more attention to be paged. So, before we go into the pro marker, I mean, again, I assume we're in a similar situation in terms of predicting
[00:17:12] and you've touched on a couple of sort of risk factors, I suppose. But one step before that, I mean, who are the candidates that are most useful for me to send to you for scopes, for gastroscopes? Yeah, I think patients who are not responding to treatment. So, I think we need to work out why they're not responding to treatment. Definitely reflux symptoms. Those people should have an endoscopy early. We do a lot of endoscopy to also check for anatomy.
[00:17:42] So, really, our hiatus hernias as a cause for that reflux or other reasons for endoscopy. If we think there is a higher chance of barates in that subpopulation, then we need to get biopsies and samples. So, endoscopy, not just diagnostic, but actually, or diagnostic in the sense that we can get information about the mucosa. But sometimes, actually, the therapeutic as well. We do a lot of interventions from endoscopic means as well.
[00:18:09] So, Johan, we've got Promarka SO. I presume it's about SO, not ESO. So, again, as a predictive tool, tell us a little bit about it and what it does predict in terms of barates, I suppose, and reflux and potential progression towards esophageal cancer. Yeah. So, ProMarker ESO has got a few more proteins that they use.
[00:18:33] We're using four, the alpha-1-antichymetripsin, alpha-1-antitripsin, calicrine, and the C9 complement. Now, all these are proteins. They're not produced. They produce everywhere else. But it's the subclinical processes that sort of changes these proteins slightly. And that's what we pick up is the subtypes of some of these proteins. So, we take that together with the BMI of the patient as well as their gender, sex.
[00:19:02] And we put that into a model that was developed by Progenomics International. And that, again, comes up with a risk score for these patients. So, what we found with it, also, it will give you a low, intermediate, and high risk. And what we found with, specifically, the Promarka ESO is in terms of predicting esophageal adenocarcinoma specifically, it was quite good at, with the specificity of 98% and a sensitivity of 91%.
[00:19:29] So, we're quite good at trying to identify patients as at low risk of having that. What we also found is a secondary finding was that a lot of these patients, as they go through the different stages of Barrett's, they all be also able to pick a lot of those up. So, patients that's on the early stages of going through Barrett, we will be able to identify those patients at high risk of them going to be further.
[00:19:56] So, in terms of a risk score, again, this sort of helps give additional data to the doctor. That's, at the early stages, not to replace anything. If a patient is already at a high risk and already indicated for endoscopy, then it's not going to be useful. Yeah. But for the patients that's in the gray zone where you sort of ensure, that that might be, give you some additional information.
[00:20:22] So, are we in a similar situation to the ProMarket D where we had the green, the amber or orange and the red, the people already read, where we already know they're at risk or they've got a problem. It's the others that are most suitable for this test? Yeah. I mean, I see it for this ProMarket SO test to be really great for patients who are diagnosed with Barrett's, but they might have no dysplasia or low-grade dysplasia.
[00:20:52] But I think it creates a lot of anxiety. This is a precancerous condition. This is something that might lead to esophageal cancer. And esophageal cancer is, you know, not a great cancer to get. You know, we've got about 1,800 patients diagnosed with esophageal cancer in Australia every year. And 1,400 people die from esophageal cancer every year. So, it's not a cancer you want to get. So, when I see patients who are diagnosed then with Barrett's, they're actually quite anxious about the results.
[00:21:19] So, to have a marker to kind of reassure them and to kind of help them to share that decision-making in terms of, actually, you're going to be on a surveillance program. With your kind of changes, because you've got no dysplasia, then we can do one in five years or three years or something like that. You actually kind of... Often the patient goes, isn't that too late? You know, five years. I guess anything can change in five years.
[00:21:46] So, having this marker might help us to say, actually, you have a low chance of developing any progression to your Barrett's. Or a three-year or five-year follow-up program is very reassuring for them. All right. So, this can really help guide when do you come back for your next scope. Yeah. That's right. And a little bit like with the previous marker, if we do get a low reading that says you're at fairly low risk,
[00:22:13] I think you said, Johan, the specificity was about 91%. So, yeah, nothing's 100% in this world. But people can... We can all rest as easy as it's reasonable to do on a low result. I mean, reading the result, and maybe Johan can correct me if I'm wrong, the negative predictive value for this result when you put in the test result plus the age, sex, BMI,
[00:22:38] where the algorithm calculation is 99.9 negative predictive value for this test. So... 99.9 is pretty good. I'll take that. I'll take that too. So, I think how I read it is that if the test is low, really low score, then I think the chance of them, you know, finding a positive, like a... I think that's going to be unlikely, given that negative predictive value of 99.9.
[00:23:03] So, it's very reassuring for patients that they're unlikely to progress into a higher stage or unlikely to progress to a sobriety of cancer. And turned on its head for the person who may come up at high risk, what's the sort of conversation with them? Um... If they are at high risk, then they need to go on... I mean, they will be on a surveillance program anyway. So, we might need to intensify that. It is much better to pick up a sobriety of cancer early.
[00:23:33] The treatment is vastly different these days compared to picking up stage 3 disease. Stage 1 disease can be often treated endoscopically. Um... And I guess one question, and I guess in the future, what I'd like to maybe consider with the pro markers is, can there be a surveillance tool for post-treatment? You know, after they have their esophageal cancer resected endoscopically, could that be a surveillance marker as well?
[00:24:02] So, I don't know the answer to that, but something maybe Jochen might be able to do. Yeah, I mean, at the moment, we're in the beginnings of using this as a tool in industry. So, for now, it's just a risk. But there's ongoing research, and that's definitely one of the opportunities that we would like to explore further, is to actually, if that's going to benefit patients even further, by using it as a surveillance tool. And also mentioned before with the pro marker D,
[00:24:32] that as we said at the moment, it'd be a four-year before reassessing. Is there a timeline for reassessing the pro marker ESO as well? Um, no. So, once it's done, it's done? Well, once it's done in the purple... And these are usually for patients that's pre-scope. So, once you've done a scope, it's usefulness sort of deteriorates unless there's a clinical indication for it. All right, so really a once-off test, and that guides surveillance.
[00:25:00] And after that, the surveillance guides what happens next. Yeah, I mean, what you could do is, if the patient is, for instance, low risk, and your clinical intuition is sort of also sitting the patient at a low risk, but he sits in a population group that needs ongoing surveillance, it was worthwhile then considering doing it in 12 months time. Again, this is not... This is just sort of what we suggest. It's not based on any ongoing research at the moment.
[00:25:29] With time, we'll be able to give a more accurate prediction of that. So, both of these tests, I'm assuming, are relatively sort of new. I'm guessing there's no Medicare funding at this stage. Roughly, what are people looking at if their patients are wanting to get it done? For the pro-market disease, it will be a few hundreds. We will... And then the pro-market disease is a little bit more expensive
[00:25:58] because of the techniques that's required to actually do the analysis. I mean, to me, and maybe people have different opinions. I mean, that's not stupid money if you're talking about something, particularly with the pro-market reset. But with both of them, that potentially is going to save you a lot of problems down the track. Not everybody sees it that way, but... Yeah. With a little bit of a crystal ball, and I'll throw this to all three of you,
[00:26:26] are there more of these pro-markers under development? I mean, in a decade, might we have a lot more of these tools? Yes. So, Proteonomics and Cannesial has been busy for a long time, 20-plus years already, in connect working with UWA and people like Tim Davis, to develop these tests. And one of the ones that they're busy with that should become available next year
[00:26:55] is the pro-market endo. And that one's pro-market endos is to try and identify patients early on that might have endometriosis. What these tests need is a good population that's serially followed with valid and reliable data. And they don't grow on trees, unfortunately. So, for example, in the diabetes space,
[00:27:21] we're developing a similar pro-marker for diabetic peripheral neuropathy. Right. So, again, trying to predict those people that are going to go bad with their feet and hoping to prevent things like amputations and revascularization. So, that's in a fairly advanced stage of development at the moment. A bit like Johan said, there are more proteins involved, not two for kidney disease. We've got six at the moment we're looking at.
[00:27:51] So, where you've got samples and longitudinal data from a good, reliable sample of patients, you can develop these models. Right. So, they don't come out of the woodwork. A lot of investment comes in up front to try and get the conditions you need to develop predictive models. And then you need to validate them. So, pro-marker D, for example, has been validated in type 1 diabetes
[00:28:18] in indigenous patients in a large clinical trial in the US of one of the SGLT2 inhibitors. So, you have to then make sure that not only is the test a good one in your population, but that it can be translated beyond that. So, you know, that's why there's a cost associated with these tests, is that the R&D costs are not small. Sure. I mean, with... I was going to say, you mentioned about numbers.
[00:28:48] I mean, is there international experience in bigger populations? Are these being used in Europe or North America, for example? Proteonomics does have a presence in North America. So, for both pro-marker ESA and pro-marker D is available in North America currently. Maybe just also just to add on to Tim. So, pro-marker D has originally been developed for type 2 diabetes only, but there's increasing evidence as I suggested for type 1.
[00:29:18] And then we sort of, as time goes on, we will, especially the population groups that might not have been part of the regional population group, is going to be included. But, and then longitudinal information is going to be essential for this marker to actually develop into a more sophisticated tool. I mean, it is... We live in an era where there are lots of tests and predictive things available. Yes. So, you've got to make a call as a practitioner about which ones you're going to choose.
[00:29:48] But I would argue that things like solopageal cancer and chronic kidney disease are pretty important to be able to predict. I think these are kind of complicated disease process, but esophageal cancer barats are not easy decision-making trees for the GP. So, it would be good to have the specialists involved early with that. All right. Oh, look, this has been fascinating. Interesting. I'll put my hand up and say, I've learned a lot.
[00:30:14] And it makes me think that we're really on the dawn of a new era. I could imagine in 10 or 20 years' time, we have a suite of these tests, and our ability to tailor or personalize our approach will make what we do today look a little bit Stone Age-ish. So, before we close today, just like to ask each of you for maybe a key takeaway for the listeners. We'll go in reverse order. We'll start with you this time. Yeah, all right.
[00:30:43] My key takeaway and message would be to, these are tools to help. It should not replace current guidelines or the role of the specialist or the GP's clinical intuition. So, I think reflux is a very common disorder that you treat day to day. I think patients with reflux symptoms, patients with chronic long-standing reflux,
[00:31:08] and then they have a higher risk factor like male over the age of 50 with a high BMI, I think those people should be flagged and should be seen as specialist or consideration of endoscopy early. And then this marker is so help in that triage process. And I think for primary care practitioners who see a lot of diabetes patients these days, it's an increasing issue. And as I say, they're living longer and getting chronic kidney disease
[00:31:37] and the end-stage ramifications of that, that these tests will become more important in future. Tim, Bill, Johan, thank you very much for your time today and sharing your knowledge with the Medical Forum listeners. Thanks, John. Thanks for listening to the Medical Forum podcast. Please follow and encourage your colleagues to do the same. And if you have information and knowledge you'd like to share and be on this podcast, contact the Medical Forum team. Thank you.


