The Market Bull recently invited Imagion Biosystems’ Ward Detwiler and Wayne State University’s Dr. Mark Haacke on the podcast to discuss how molecular MRI and quantitative MRI techniques are setting the foundation for improved imaging clarity and automation. Watch the video above or find the transcript below.

Video Transcript:

Welcome to the Market Bull podcast, where we dive into the stories shaping Australia’s finance, tech, and resources landscape. Each episode, we will be sitting down with ASX listed executives, industry experts, emerging explorers, and innovators to unpack the latest trends driving investor attention. Whether you’re a retail investor, market watcher, or just simply curious about the companies making headlines, this is your place to hear Australia’s next generation of growth stories. I’m Matthew Craig and this is the Market Bull podcast powered by the marketbull.com.au, your home for small cap coverage. Hi, I’m Matthew Craig and joining me online today is the Imagine Biosystems Chief Business Officer Ward Detwiler and the co-director of MRI research at Wayne State University, Dr. Mark Haacke. Welcome, guys.

Hi, Matthew. Thanks for having us.

Thank you, Matthew.

So, maybe I’ll just start with you, Ward, just to introduce us and give us a reminder of the MagSense technology and just what Imagion Biosystems is working on and, you know, why we’re here today.

Yeah, absolutely. So, um, imagion is a clinical stage medical imaging company really focused on changing how we detect and diagnose cancer. We’re doing that through our MagSense imaging agent technology, which is a first of its kind targeted magnetic nanoparticle, which brings molecular specificity to the over 50,000 installed MRI systems globally. Our first product focused on this improved staging of HER2+ breast cancer has successfully completed its phase one trial and we are now preparing an IND submission for our phase 2 study which we expect to submit later this year. Um but what I think is really most unique about MagSense beyond the HER2 implication is that this is really a a platform technology. While we are laser focused on our HER2 and in completing the phase two successfully, we have additional applications in prostate and ovarian cancer that have completed a lot of their pre-clinical work and are now ready to be prepared for IND enabling studies in the future. So that’s something that we’re keeping our eye on. And pretty much anything that we can find a targeting leg end um to to address, we can find a cancer that we may be able to to go after and help improve the diagnosis of.

Yeah. Awesome. And just over to you Mark maybe just give us a bit of a background on you know how you’ve got here today all the work that you’ve been doing and then just maybe give us an overview of why molecular MRI is so important.

Thank you Matthew. Well I’ve been doing MRI for 40 years now. MR imaging is uh really an exciting field. It has advanced tremendously as you can imagine over 40 years in terms of the equipment itself and many new developments in both hardware and software in MR imaging as well as many new  technical advances in how we’re able to probe pathophysiological processes in the human body. And uh although a lot of MR today is predominantly structural MRI, it’s also possible to do things like image the blood vessels of the human body. So if you are delivering a drug through the vascular system for example we can map that vascular system out very accurately. Um and along those lines we have actually used something called ferumoxytol which is an iron-based contrast agent to image the vasculature down to 50 microns. So when we have an iron-based substance uh it makes it possible to dramatically improve the detectability of small structures and many years ago I think back around 2017 I published a paper using again ferritin to see you know how accurately could we measure iron content and I believe we measured iron down to the level of six picograms of iron per cell. Uh and we also showed that um we could image with with ferritin which is I think similar to the iron-based contrast that Imagion is using that we could literally image a few hundred cells. So that’s very impressive what MR can do. Uh of course those were test tube measurements. Um now we have done something similar to that with the current doses that we were given by Imagion down to now the level of about one microgram per milliliter in terms of the dose and and we can still see that effect quite well. So having that ability to um tag these type of um changes in the body especially you know cellular changes occurring in cancer for example makes it possible to really search and tag very specific pathophysiological processes. And I I think that’s really the future for MR. um can we design you know other technologies and I don’t think anything is more powerful than iron tagged technologies to continue to enhance our ability uh to use MRI not just for structural imaging but for molecular imaging at the level that we’re talking about here with Imagion.

So can we run this for for all of these 50,000 magnets Ward’s talking about?

Well, the the ability to do this is is certainly best at high field. And by high field here, I mean maybe 1.5T and 3T where the presence of uh a magnetic contrast agent basically enhances our ability to distinguish these small structures. And although there I’m not sure how many that is out of the 50,000 but even if it’s half of that it’s certainly a lot of machines worldwide and most of the modern scanners today are high field scanners. So that opens the door for not only imaging these pathophysiological processes but now for automatically detecting the presence of these cancer cells if they can be you know monitored by such a contrast agent. And so I I think uh combining all of these three concepts together is really you know where we are today. We’re at the forefront of this technology and I think Imagion is really entering this at the right time and providing the right type of um agent for us to be able to probe this cancer.

Absolutely. It’s incredible work. I’ll maybe just go to you Ward and just let us know so what was the purpose of the recent study at Wayne State and why was it so important for the MagSense roadmap?

Yeah. Uh so first of all I think setting the stage of why we went to Wayne is that Dr. Haacke is just one of the absolute top minds in the world of MRI. I everything I know about MRI I learned from him over several years working together. So when it came time to uh take on the study, there was only really one place in my mind that we were going to go because of a lot of the work that he and his team have done in improving these sequences, developing a lot of really advanced quantitative sequences um and being able to see things you couldn’t really otherwise see with MR. So in doing that um we knew we had a few things that we wanted to to address in the phase 2. Um the first of all was reducing the dose of the Magsense agent. We knew from some of the work we had done in the in the phase one that um one that we should be able to reduce the dose considerably and then two that there was a big benefit to it. One was that you know it’s going to reduce our cost first of all of you know per application of of the agent. Um but two also going to significantly improve patient comfort um safety tolerability so I think lead to an overall more successful study. But to do that we really needed to substantiate that those claims uh for the IND submission. So that’s what we were really looking to study is how low of a of a dose could we get to and still have a discernable detectable signal. Um and I you think we certainly have gotten there and have what we need to substantiate that and include that in the IND submission. Um but on top of that I think the other thing for me personally and maybe um but also for all the radiologists out there was uh really improving the image quality that we had improving um the you know sensitivity and readability of the sequences that we were reusing. Um that was one of the first things when I came into the company was one I was you know incredibly excited about what they were doing from an MR standpoint but saw a lot of room for improvement by using a lot of these advanced sequences that Dr. Haacke and his team and others around have have developed and perfected. So that was something else we’ve explored in the study um and has given us a lot of potential for future directions and ways that we may be able to apply that. And then third was really looking at the application of quantitative imaging techniques into our protocol. Um and that’s something that I think is going to lead to some really interesting uh interesting results in the future is rather than just looking at a at an image on the screen, now we’re able to actually measure what is there on a pixel by pixel basis. Um we’re able to evaluate these images much differently. Um track changes over time, but also provide a lot more certainty into what we’re seeing. Um and there’s a lot of different ways that we can apply that going forward, looking at um potentially reducing the need for a a predose scan, looking at the development of uh various AI and automatic detection tools. So um that may not be something that we use right away but is setting the stage for a lot of exciting things that we plan to address in the future.

That’s fantastic. And just over to you Mark. So what were some of the key findings of the study and are you able to reduce the dose as was intended?

Yes I think we have successfully got this down to as I mentioned about one microgram per milliliter. This method we use it’s called a gradient echo three-dimensional scan and the advantage of that is that we can run this with very high resolution on the order of a cubic millimeter in terms of voxel size. So that’s quite small and that uh means that if you have some inhomogeneous distribution in the lymph nodes of the cancer for example that you have the ability to really be able to study this fine structure because this may not be some uniform distribution inside inside the lymph node. So the the methodology that we’re using is quite flexible. We can uh change the coverage we have. We can potentially, you know, scan the entire uh breast for example or the upper area where the lymph nodes are probably in just a few minutes with that type of resolution. And so this offers, as Ward said, if we run the scan twice, we’re able to develop a lot of quantitative information at the level of one cubic millimeter. And at that point the iron changes the tissue properties in a way that something called an R2 star map can be used to actually automatically detect these changes relative to surrounding tissue. So the technology is there, the speed is there, the comfort for the patient is there and the advantage of the high field scanners is the signal to noise and contrast to noise are available to make this detection really possible.

It’s all sounding very positive. I might just open this one up to the to the room a bit. Uh so just tell us what are some of the implications of introducing quantitative MRI into the protocol and what does this enable?

I’ll go and give you maybe my you know my analysis or the assessment of this and uh hopefully Dr. Haacke can correct me if we’re anywhere I may be out of line, but um I think there’s a couple really interesting directions we can go. First is eliminating that that predose scan. Right now we scan the patient then administer the Magsense injection and then at some interval later which is something we’re going to try to optimize more in the phase 2 study. We would then do a second scan and the radiologists are kind of looking for the difference between pre and post dose which is a you know fairly standard way to um look at you know contrast MRI. Um however by using the quantitative imaging and being able to sort of capture that unique quantitative signature we may be able to completely eliminate the need for that pre-dose scan that’s something that we’ll be looking at in future directions as well in future work and that is hugely important one for just the the patient you don’t have to have two MRI sessions especially if it’s over a long time period between the pre and post dose you know maybe that even could be overnight in some cases. You eliminate that hassle for the patient themselves. It also makes it much easier from a workflow perspective for the the imaging center um because they’re only having to run one scan. You’re not having to worry about timing. Um so I think that’s going to be a a really important direction for us as we look at clinical adoption. Um not to say that otherwise that it’s you know overly burdensome especially compared to some other diagnostic techniques but you know everything we do to make this faster and easier for the patient and for the clinicians is going to be really important. Um, beyond that, I think there are some other really interesting things that we may be able to do as we look at automatic detection and potentially like algorithmic and AI based detection. Um, radiologists are incredibly busy. They’re reading a tremendous number of scans each day. And anything we can do to make that that job easier and faster and more accurate for them is going to be a huge help. I think it also though beyond making it faster I think makes it more accurate and expands access beyond traditional academic medical centers. You know we had our in our phase one we had our radiologists a reading panel look at these images. We had them draw little pictures of it and see what they were saying and then we had you know good concordance across the board. Everybody’s seeing the same thing. They saw the same ratio of signal change. Um but those are with people who are you know academic radiologist. They’re specialized in this area. There are a lot of patients who are beyond the reaches of those types of systems and anything that we can do to make to put this in the hands of more people to a broader audience and open up the access I think is going to be incredibly important. And then outside of that, I think what we’ll be able to look at um in terms of analyzing the progression of the disease or staging the disease, which really is our our primary focus, can be assisted by that algorithmic approach of looking at different patterns of of the uptake into the lymph nodes seeing how the progression the disease is spreading. Those are all things that we can start to monitor with with quantitative imaging and with AI. So I think that’s incredibly exciting and by implementing these sequences now ahead of the phase two we’ll come out of that with a wealth of data that we can then start to train those models on. There’s some other things I think just related to I mean I could go on around the the current limitations of of AI and radiology. It seems like everybody is so hot to trot about this subject and thinks it’s going to solve everything. But there’s just still some fundamental limitations of what you can do. Whether it’s, you know, starting to build big enough data sets, you know, you still have a lot of u subjective kind of human influence in the rating of the data sets and building these data sets. Quantitative imaging really starts to eliminate a lot of that. So, it’s pretty exciting. I think where we will be able to go in the future.

And what were what were your thoughts there Mark?

Well I agree with what Ward has said. I think the interesting thing about the QMRI and this particular measure called R2 star is that one can potentially especially in terms of following up treatment response one can do a direct comparison rather than just a simple visual comparison. Oh it’s there, it’s not there. One can look at this and see if this R2 star parameter has changed. And if that parameter is very high, it usually means there’s a lot of cancer present. If that parameter is starts to reduce, we can begin to say how effective that treatment has been. So I agree with Ward. And the other thing is that as we you know learn about that part of the body, Ward that you know and map out the R2 star values throughout all the normal tissue there. Then we can have a direct map of what is considered normal and what is considered abnormal. And so that QMRI image can easily be fed through AI and AI can maybe do even more than just say oh yes there there’s a indication of cancer cells there. it can say, you know, here’s what’s happened from the initial scan uh to the second scan and and there appears to be some positive response to the treatment. Uh so, you know, the AI can really make the analysis easier and also, I think AI can help in the sense that it’s not just the time involved for the radiologist to look at this. If we’re talking about really small structures, it’s possible the radiologists could miss those as they’re looking hundreds of images where they’re tired late at night. You know, you have a few little bright spots in there. You just they just you might just think your eyes are tired with those bright spots. So, so in this particular case, AI, I think, has the potential to know where to look, what to look for, and how to assess it. And, and you know, that’s really training it, you know, like an a like a radiologist who never gets tired.

That’s a really important point there, Mark, about looking at these smaller structures, and this is tied to some of the research work that we’ve done in the past is that we’ve seen that by using this targeted imaging agent, we’ve been able to detect, you know, affected nodes that are or tumors that are, you know, up to 10x smaller than you would with traditional CT or MR alone. And I think that just leads to that point is that you can not only we’ve tagged them and find them, but now with AI, we can light them up and make it easy and quick to find and hopefully reduce any potential errors there may be.

Yeah, I think it’s really interesting. Obviously there’s a lot of biotechs you know that are trying to bring maybe new drugs or new technology to market whereas this is more just growing and expanding on technology that’s already there but you know can always get better and more accurate. And it really sounds like it’s setting a setting a bit of a foundation for the future of kind of MRI and AI and automated detection which as you guys have mentioned these are humans looking at these images, they’re, you know, going to make mistakes. So, obviously, it’ll it’ll eliminate some uh human error, I would imagine.

And eliminating the need for things like, you know, confirmatory biopsies. I mean, right now, conventional MR, you still, even with the amazing resolution that you have with it, you still need to go in and confirm with the biopsy. And that’s what we’re hoping to solve with this molecular MRI is being able to provide that sort of precision um and diagnostic information to to hopefully avoid a lot of these unnecessary um and you know painful uncomfortable biopsies that patients have.

Absolutely. Uh maybe just lastly Ward for any investors that might be tuning in here from a from a business sense and maybe you can just give them a little bit of an update on you know how the IND application you know is progressing and maybe where to you know next for for IBX and Imagion.

Yeah so um from a IND standpoint um everything is on track as we have planned and promised to the market we are on track to submit uh you know by the end of the year. This is an important step towards that goal providing a lot of the good kind of confirmatory information about some of the things that we’re doing on the imaging side. Outside of that we’ve completed the manufacturing of our clinical lot of material. We’re now in the analytical testing stage of that. We expect that to be released from the stability testing fairly soon and um after that we we should be fairly ready to submit. We’re completing our you know finalizing our study protocol right now looking at site engagement. So we’re pretty well on track from the standpoint of getting to our phase two.

That’s great news. It’s good to hear. Well, thanks guys for joining us today. Thank you Mark for joining us and yeah I hope these study results come back positive and we get the application passed.

Absolutely. Thank you very much.

Appreciate it.