Michael Smith, MD (Host): Welcome to the Children's of Alabama Pedscast. I'm your host, Dr. Michael Smith. And joining us today is Dr. Michael Seifert. He's the Medical Director of Pediatric Renal Transplantation at Children's of Alabama and the University of Alabama at Birmingham. Now, today we're going to dive into the fascinating world of spatial transcriptomics and explore the innovative advances that are happening inside the UAB spatial core lab. Dr. Seifert, welcome to the show today.

Michael Seifert, MD: Thanks for having me. It's great to be able to talk about this.

Host: I love it and I'm fascinated by it. I never really understood this before and I started reading some of your stuff and I think I got a better understanding of it now. But I would love to start off, like get your high level definition of what we're talking about today, spatial transcriptomics.

Michael Seifert, MD: Yeah, so I think for my field, whenever we talk about making a clinical diagnosis in a kidney transplant, we're always dealing with spatial context, meaning where in the tissue is the disease happening? Because it's usually not happening everywhere. Some parts of the tissue are going to be normal and some parts are going to be abnormal.

And you sort of take an average of what you see to make a decision clinically. But that's not the approach that we've traditionally taken with research approaches to studying kidney transplant biopsies, meaning that most of the platforms that have been available up until recently, force an investigator to lose that spatial context and you study an average over the whole graft, or transplant and you lose where some of that important data might be coming from.

So for us, we like the spatial platform because it allows us to maintain the tissue's original architecture so we can study disease where it's happening not just from a clinical basis, but also using some more advanced research techniques like looking at the gene expression patterns or transcriptomics, as you said.

Host: And a lot of what I've been reading about this, I think specifically with your work, correct me if I'm wrong, it's focused on, I guess, identifying, these transcription signals of rejection, correctly? That then I guess almost they become like biomarkers of rejection, but you know exactly where that's happening, not just, overall. Right. Is that kind of the concept there?

Michael Seifert, MD: Yeah, very much so. So there's been decades now of data showing that gene expression patterns coming from a transplant are maybe a little bit more sensitive or more informative early on for problems that are coming from a kidney transplant. The problem is, is that we've never exactly known where those signals are coming from.

Are they coming from cells in the kidney that we care about, or are they coming from cells in the kidney that may not be as relevant? So, kind of reinterpreting those studies using spatial transcriptomics allows us to see those same gene expression patterns, but now we can see visually exactly where they're coming from.

Are they coming from the immune cells that are coming into the transplant, which would be really relevant. Or are they coming from the endothelial cells, which line the blood vessels that are kind of like the door to the transplant, or coming from some other cell type that maybe we didn't think was as important until we were able to study the transplants in this way?

Host: So this is really allowing you to get almost down to the actual cell in its transcription versus, a brick area of tissue?

Michael Seifert, MD: The way that we profile it's, referred to it as regions of interest. Meaning we can get all the way down to a custom defined shape of tissue. So literally, you can look at a picture of a kidney biopsy on our instrument screen, and you can take your mouse and draw, a dodecahedron around it, if you wanted it to be that specific, uh, or just do a circle and, it will profile everything in that shape, ignoring everything else around it.

And within that shape you can say, well, I'm just interested in this particular cell type that's inside of that shape, and you could ignore everything else. There are spatial platforms that do allow you to get all the way down to the cellular level, as well. So this field is really growing a lot and it's an exciting time to be a part of it.

Host: So what is, you know, a lot of our listeners are, are clinicians, general practitioners. What does this mean to them in the clinic? What's the clinical application of this or that you hope it eventually will be?

Michael Seifert, MD: I would look at this kind of technology as a discovery tool, meaning that it, it is unlikely that a clinician, like on the other part of my job when I'm not doing research, I'm seeing transplant patients. So I can't foresee a scenario where I would do a biopsy and then using spatial transcriptomics is a way to make a diagnosis.

Because it's a very labor intensive, time intensive, cost intensive technique. But it's a great technique for, I want to understand more about what's going on inside of the transplant when it comes to a particular process or a particular disease. So I could see this as a discovery tool where an investigator might partner with clinicians who are going to allow us access to tissues that we can study in this way and make discoveries that will then inform things that are going to impact clinicians directly, like new diagnostic tests that are based on spatial profiling data, new mechanisms involved in certain types of diseases that we didn't understand before, that's going to help fill out the knowledge base for clinicians that are taking care of these patients.

Host: So there's a big movement in medicine I think, that's helping us to understand that disease is not the same in one child versus another, or even in adults, and you see this in cancers, you know, we're learning that each type of cancer can behave differently. It seems that is this leading down towards the idea of more personalized medicine then, being able to really look at somebody's kidney and find what's really wrong in in that specific area? Is that part of this?

Michael Seifert, MD: I hope so. I think we're, maybe personalized medicine in a different sphere than you would think about traditionally. And what I mean by that is what a spatial transcriptomic platform would allow you to do is to say, I'm going to be able to uncover some of the heterogeneity within a tissue.

So I'll explain what I mean by that. You might look at a kidney transplant as a clinician or a researcher and say, well, I'm interested in what all those vascular lining cells or endothelial cells are doing, assuming that they all behave the same way, but in reality; each of them behaves a little bit differently depending on what part of the kidney they sit in.

So this kind of technique allows you to personalize, from the standpoint of the transplant, I guess, where the signals are coming from. Is this a problem in the blood vessels, in the filters of the kidney? Is it a problem in the blood vessels and the tubules or the urine generating parts of the kidney?

So that kind of personalization and precision we can definitely figure out with this technique. And then I hope that'll allow us to understand the signals that vary more from person to person so we can really apply that more personalized technique.

Host: So rejection pathology is much more nuanced than just the body rejecting it. Right. There's a lot more to it.

Michael Seifert, MD: Yeah, for certain. That has been part of the problem that's led to these studies is that the governing bodies that, if you will, that decide this is rejection, this is not rejection; they're the first to acknowledge that their criteria are

somewhat messy. They're kind of designed to make a complicated and heterogeneous process appear a little bit more homogeneous than it actually is. So I, I hope this technique allows us to get at some of that nuance that you were just referring to and understand that these disease processes have a lot more interesting subtypes than might be obvious at first glance.

Host: Fantastic. Anything specific that you're doing in your lab, you and your team? New techniques that you want to share with the listening audience?

Michael Seifert, MD: Yeah, so I think there's two ways to answer that question. So one is what we're doing as a lab that happens to house this instrument. And the second set would be what we're doing as a core, meaning we're offering the use of this instrument to other investigators that may not have access to technology like this.

So for the things that are happening in my personal lab, we're studying biopsies of kidney transplants that have been stored for years because they were used to make a clinical diagnosis and then just left in an archive in case they might be needed for something down the road. This particular platform allows us to get access to those biopsies and use them and kind of ring out some more useful information than otherwise these biopsies would just kind of be sitting in storage forever. So we're using the platform to make diagnoses on old tissues without having to get new specimens from patients, which is really helpful for the patients. But also we're using this platform to, as I said before, really understand what different cell types in the kidney might be contributing to these diseases.

So we're not just looking at tubular cells that make urine. We're looking at what are the different types of tubular cells and how do they behave? What are the different types of vascular cells in the kidney and what different roles might they play in protecting the kidney from rejection or also kind of permitting rejection to occur?

So that kind of specificity is really fun for us and we're just starting to study that, in our kidney transplant biopsies. From the standpoint of the core, it's really fun because we're offering our services to the whole UAB campus and children's campus of investigators. So just last week we had a study where we were cooperating with people in the ophthalmology department.

And so we're studying diseases of the eye. We're studying cancers, we're studying diseases of the lungs. And so being able to study different organ diseases as well, because we operate as a core has been a lot of fun for us and we would be open to collaborating with any investigator that had a good question that spatial biology was the right path to answer.

Host: It does sound like, with all the work you're doing, learning new and new stuff, you know, even looking at the old tissue and all that, this must impact down the line, at least some of the transplant protocols, right?

Michael Seifert, MD: Yeah, we hope so. A lot of the ways that we manage transplant patients with medicine haven't changed appreciably in the last 20 or 30 years; even though our knowledge base hasn't evolved to the extent that we're able to design or deploy new therapies. And so my hope is that this will allow us to have a deeper understanding of the processes involved in transplants doing well, but also transplants doing poorly. So that'll help us design better management programs, whether that's using our existing medicines in different ways, or designing new medicines that can be more targeted and more effective than what we currently have available.

Host: You wrote an interesting paper. You were interviewed by somebody else and you were talking about these noninvasive biomarkers, what spatial transcriptomics does for you. Tell us a little bit more about that. How are those types of biomarkers advancing the diagnosis and management of transplant rejection?

Michael Seifert, MD: That's a great question. And and I think that topic is in some ways very much linked to what we've been talking about. And I mentioned before that the spatial platforms are really more discovery platforms designed to understand new things about what's happening in a transplant.

Those new understandings can lead to the development of new diagnostic tests; a lot of which are designed to be less invasive than a biopsy would be, so that you can deploy them more often and have a better understanding of what's happening in the kidney without having to go in and get a biopsy and subject the patient to a hospital admission and those kinds of things.

So, a lot of the non-invasive biomarkers that we have available in blood and urine, they've been derived from these kinds of discovery studies that have been done. Not with spatial studies because those are so new, but a lot of the basic science advances that have happened in transplant have led to some of these biomarkers.

And, just one example is we've developed a panel of four proteins that you can measure at the same time in a urine specimen of a patient. And it has a really good indicator of whether a transplant is truly doing well or not better than some of the traditional clinical markers that we have available.

And so we're hoping to develop more tools like that that can be used to better understand when patients need us to be more invasive and investigate to keep their transplant safe, versus those patients that may be doing better than we think, and we don't need to apply as much care to them because they're actually doing better than some of their traditional markers might suggest.

So it's all about developing just better tests to help us better understand how a kidney's doing at a given point in time.

Host: When you look at this, what you're doing in this industry of non-invasive biomarkers, specifically what you're doing with the transcriptomics, what are some of the biggest challenges that you see in the future that you're going to have to overcome?

Michael Seifert, MD: Two issues come to mind. One is just implementation. Some of the biomarkers that we study in our lab have been known to be effective for many years, but being able to develop them on platforms that can be easily studied in a clinical lab, in a hospital, at reasonable cost can be a big challenge.

And so I think that's part of it, is just taking the tests that we know work better and finding ways to make them reproducible, more accessible, lower cost so we can get them to the point of care a lot more easy, uh, easily than we are now. The second thing is there are certain types of diseases that have traditionally been resistant to profiling by non-invasive biomarkers.

And one of those diseases that comes to mind is this idea of borderline rejection, which is a kidney transplant biopsy that's not entirely normal. But it doesn't entirely have enough inflammation to rise to the level of rejection either. And that's actually one of the more common diagnoses that we see when we biopsy a kidney.

And the literature isn't exactly clear what we're supposed to do about a biopsy like that. Do we treat it as though it's true rejection or do we consider it's more normal and don't treat it? Or do we do something in between? My hope is, is that these kinds of biomarkers we've been talking about are going to help us tease out a very heterogeneous and confusing category, like borderline changes. So we'll know more about what to do about an individual case. The problem is, is that most of the biomarkers that are available just aren't that good at distinguishing those milder cases. And so that's really been a big challenge for the field that we need to overcome is the blatant rejections; sure, we can develop biomarkers to help with those, telling somebody's normal. Sure. But, those in-between cases are the ones that really need more effective biomarkers than we have available. And it seems like every time a new one comes out, it performs well in the extreme cases. Well, in the normal cases, but not so well in those in betweens.

Host: This has been fascinating. We could talk so much longer on all this, but, let's end by having you give us kind of the last word. What's the last thing you would like the listening audience to hear from you about this?

Michael Seifert, MD: I think I would just want them to be aware that spatial biology is really emerging as a cutting edge technique for understanding diseases in children in general. You know, we spend a lot of time talking about how my lab and my core is using a specific spatial instrument to study diseases in transplantation.

But I think what's come out of that is an appreciation that so many of the diseases that we study in general, the spatial context is incredibly important. Understanding that diseases are not diffuse; oftentimes they're more focal. And so just being on the lookout for broad scientific discoveries in the spatial biology domain and how those might inform all kinds of different diseases and give us a better understanding of how they're operating than what we currently have.

Host: Fantastic. Thanks for coming on today.

Michael Seifert, MD: Yeah. Thanks for having me. It's been fun talking with you.

Host: That was Dr. Michael Seifert. If you'd like more information, check out the link in the show notes. That concludes this episode of Children's of Alabama Pedscast. If you found this episode helpful, go ahead and share it on your social channels and be sure to check out our entire podcast library for other topics. Thanks for listening. I'm your host, Dr. Michael Smith.