Melanie Cole, MS (Host): Welcome to Better Edge, a Northwestern Medicine podcast for physicians. I'm Melanie Cole, and joining me, we have a thought leader conversation for you today. In this panel, we have Dr. Harris Perlman, he's the Chief of Rheumatology and the Mabel Green Myers Professor of Medicine at Northwestern Medicine, and Dr. Arthur Mandelin, he's an Associate Professor of Medicine in the Division of Rheumatology at northwestern Medicine; and they are here to to highlight how deconstruction of rheumatoid arthritis synovium defines inflammatory subtypes.

Doctors, thank you so much for joining us today. Dr. Perlman, I'd like to start with you. Can you just give us a little overview of the key factors of rheumatoid arthritis synovium, and why understanding its inflammatory subtypes is so important in clinical practice?

Dr Harris Perlman: Thank you. So, I think one of the things you think about with RA synovium, the synovium is the area we're just most affected in patients with rheumatoid arthritis. And in the synovium, we see an increase in cell number, and we see an increase in destructive behavior of these cells. And I think what we think about now in RA compared to years ago is we used to very much focus on looking at peripheral blood as a marker or medium to understand biomarkers of disease, trying to understand the pathogenesis. However, now we've taken a page out of oncology, where we said, "Let's go to the focus of the tissue that's being most affected," and that's the synovium.

So in the synovium, we have two major cell types, fibroblasts and macrophages. And during rheumatoid arthritis, we see a complete change in the synovium, where you've gone from fibroblasts and macrophages to now include T-cells, B-cells, NK cells, gamma delta cells, a huge number of different immune cell types. In addition, what we've also learned is that all these different cell types are heterogeneous. So, there's multiple populations of each individual cell type. And I think that's the biggest change. So, one little population can have a major effect and that could be one of the deleterious population that is being most affected in patients with rheumatoid arthritis. And that's the key part, is that going to actually target tissue, the synovium, to understand how the disease is behaving.

Melanie Cole, MS: Excellent explanation, Dr. Perlman. Thank you so much for that. Thank you, Dr. Mandelin. I'd like you to briefly describe the study to deconstruct the cell states and pathways that characterize this pathogenic heterogeneity that Dr. Perlman was just speaking about, and your roles as investigators in this research. Explain this concept of deconstructing rheumatoid arthritis synovium.

Dr Arthur Mandelin: This was a very big project. It was a multicenter international study, and Dr. Perlman was the basically primary investigator at the Northwestern site. And I'll let him describe when I'm done his role more fully. My role in the study was to actually obtain the sample of the synovium. So as Harris had mentioned, our thought is that people have been looking for decades in the serum for a biomarker that would tell us what a patient responds to what therapy. And our theory now is that the reason that no such biomarker has been found isn't because the people who have been working diligently on it for decades were in any way inadequate in their research. We think the marker just isn't there. We think the marker is in the target tissue, in the synovium.

So, you need to do what the oncologists do. You need to go in there and get a biopsy, get a piece of that tissue, and analyze it. So, this study got synovial tissue from 70 RA patients and nine osteoarthritis patients. And these RA patients were recruited at one of three inflection points in their treatment. Either they were treatment-naïve, never really had their RA treated before, or they were first experiencing their methotrexate inadequate response and getting ready to go to a biologic, or they were first experiencing their tumor necrosis factor, TNF, non-response, and getting ready to switch to a different biologic mechanism of action.

So, a total of 82 samples were collected from those 79 patients. Three of the RA patients were kind enough to let us sample them twice. We got more than 3,800 cells per sample on average across the various international sites where this was done, and that totaled up to nearly a third of a million cells that were analyzed in this work. The cells were analyzed for surface expression of 58 different proteins and also RNA data coding for those proteins. And long story short, you put that into a computer and have it analyzed for the different cell types in those tissues. And what you come up with are what are called CTAPs, CTAP, cell type abundance phenotype. And it turns out that there are six different cell type abundance phenotypes of RA. For example, there's one phenotype where the majority of the cells in the synovium are fibroblasts, and there's one phenotype where the majority of the cells in the synovium are T and B-cells, and so forth and so forth. It turns out then that these different CTAPs are associated with what's seen under the microscope if you look at these samples histologically. And these CTAPs also have different profiles of inflammatory cytokines in the rheumatoid disease. And so all of this, if you're following along in the paper, is figure one in the paper. And basically, what was done then is these different CTAP categories were compared with data from another trial that has been previously done called the R4RA trial. And those patients in that trial had already failed the TNF inhibitor drug, and were either on tocilizumab or rituximab as their next therapy. And they were looked at before therapy was started versus 16 weeks after therapy was started.

And two important things were found. First thing was that these cell abundance phenotypes can change over time when the patient is on treatment. So, the abundance of various different cell types in their synovium can be altered by treatment. And the most exciting thing, I think, from my perspective that was found was that those patients who were responders to their new therapy versus those who were non-responders to their new therapy tended to sort into different CTAP groups. So, for example, I mentioned earlier that one of these CTAP groups is the fibroblast predominant group, and it turns out that the odds ratio for being a responder to either rituximab or tocilizumab, if you are a fibroblast CTAP group member, is only 0.26. So, significantly less likely to respond to either of those therapies if you are fibroblast-predominant. So, that's kind of the important thing about the study.

And my role in the study then, as I said, was to actually obtain this tissue by ultrasound-guided synovial biopsy, where we scan the most affected joint of the patient with ultrasound. We identify the synovium based on its appearance in the ultrasound. And using a spring-loaded device, we can go in there, of course using quite a bit of Novocaine for patient comfort, and this is an extremely well-tolerated procedure. We obtain a number of these samples, and we can then send them in for analysis, and that's kind of where I stop, and Dr. Perlman takes over, because then, a representative from Harris' lab is right there in the biopsy room with me, collecting these tissue samples. And then, they go off for this analysis, and that's a great point for me to pass over to Harris for his role.

Dr Harris Perlman: Yeah. Thank you, Art. Art, you're very modest, because in this patient population, we at Northwestern and our REASON Consortium provided the majority of all the samples to this paper. So, I think that is something which, even though they had 13 other sites, which Art trained these 13 other sites to perform ultrasound-guided synovial biopsies, Art and UAB and Columbia University, were our three main sites of REASON Consortium that we've led. And we are the ones who did the bulk of all the samples for the United States. So, I think, ART, thank you so much. I think that's the key part that people may not recognize, the amount of your contribution, because without that, they wouldn't have a paper. So, you know, I think that's something, which is amazing that you've led across the United States, this concept of building an ultrasound-guided synovial biopsy program.

And I think, you know, Art, you said it very succinctly of really what this article maintains. And I think the concept, if you wanted, of how to break it down to these CTAP populations is really thought about when I was mentioning how you have a heterogeneous population. So when they looked at each of the different cell types, they found 77 different states of cells. So imagine, you know, how do you really try to understand how each different state is affecting the disease?

So, what they actually did is try to combine those different states into six major categories. And I think that was one of the strengths, because beforehand, studies by Konstantinos Papadopoulos and other groups tried to use histology, looking at the histologic features of the synovium to say, "Okay, this patient would be X, Y, and Z." This group decided, "You know what? I don't think that looking at histology may be sufficient. Let's look at the gene expression, protein expression to create these subcategories." And that's what they did here in order to really have a better set subcategory. And they showed it with their initial work in the paper. And the issue was in the beginning, in the patients that we collected, they didn't have followup, so they didn't know which patients responded or not. But they then utilize that other trial to show the potential of these CTAP populations. I think that is a key part, like Art said, is that you have this now potential for precision medicine that we may be able to predict if you have different CTAP populations, you might be one of the patients that respond to an individual biologic therapy.

Melanie Cole, MS: This is an absolutely fascinating conversation, doctors. So Dr. Perlman, then what are the clinical applications of the cell type abundance phenotypes, that CTAP you've been discussing, such as their use in clinical trials or the ability to infer them from different types of data for cellular molecular insights? I'd like you to speak a little about bench to bedside and where you see this going from that.

Dr Harris Perlman: I think you're actually going to have bench to bedside and then even back to bench. So, let's just talk about the bench to bedside aspect. For each patient, which undergoes a different therapy, even for clinical trials now, the goal is to perform that ultrasound-guided synovial biopsy prior to any treatment and then four to six weeks later, because then we can start to predict which patients who will respond at a much earlier aspect. So, most trials wait 12 to 24 weeks after initial treatment. We're trying to say either we can stop the treatment at six weeks if we don't see a change in their genomic behavior, so looking at the different genes that are being expressed and different CTAP populations that are accumulating or being reduced.

I guess the best part would be is if we can now say at time zero, right before they have the therapy, is that if you don't have this CTAP population, then you are not going to respond. And I think that's really the holy grail for all of our trials, is to identify really which patients should be treated with a different biologic therapy. And I think that would be the most amazing aspect of it. And I do think that we have that potential now. We're getting closer. And what I mean by that is once we go from bench to bedside, then we can go back to the bench to further define these populations, further refine these CTAPs and really maybe even find individual cell populations within each CTAP that might be a better biomarker instead of using the six general categories for it.

Melanie Cole, MS: I love that you said bench to bedside, back to bench, because that is really so important as you're discussing this great research. And Dr. Mandelin, I'd like you to speak to other providers about the role of CTAP and the classification process as you determine a patient and patient selection, which Dr. Perlman has made clear, is so important and its potential clinical applications.

Dr Arthur Mandelin: As Harris mentioned earlier, I think the important part about this research is that it goes beyond simple histology. Granted, it does show that the CTAP classification does match well with what is seen under the microscope in terms of histology. But by going deeper and looking for 58 different proteins in this case, and who knows, as we go bench to bedside to bench to bedside around and around the cycle of healing, maybe you want to call it, maybe we'll get even deeper and into even more proteins or other things in the future that we might look for in these histological samples. Maybe by looking beyond just what things look like under the microscope and getting under the hood, so to speak, into the nitty-gritty of exactly how these cells are working, what cell surface markers they're expressing, even down to the level of what RNA expression is happening in these individual cell types, maybe we can gain much more granular information, much deeper knowledge than simple histology could give us. In a perfect world, if there was such a thing as a methotrexate responder histology and a methotrexate non-responder histology, there you've already used up two of your six groups. There's many more than just six drugs on the market, so you're going to have to eventually have much, much more detailed information than just what you see under the microscope. And two different samples of RA synovium that look the same under the microscope and perhaps even that may stain the same with certain cell surface markers might be different down at the RNA level. So hopefully, that's what we're moving toward in the future, is to get just a more and more refined understanding of who has what type of RA so that we can gain the ability then to treat more specifically.

Melanie Cole, MS: I'd like to give you each a chance for some final thoughts here, because this is so important for other providers to hear. And Dr. Mandelin, I'm going to stick with you for a second here. How does the research that you and your colleagues have been doing contribute to that overall understanding of rheumatoid arthritis as you're speaking to other providers, whether they are in the specialty of rheumatology or primary care and seeing these patients on a regular basis? Tell us about what you want them to take away from this research as they are working with their patients and the development of future research plans that you'd like to share.

Dr Arthur Mandelin: I think the important thing is to continue to encourage your patients to participate in this type of research, because we know that 20-40% of patients, even in clinical trials, which is an artificially good environment being in a clinical trial, 20-40% of those patients never reach superiority versus placebo, the ACR 20 level of response. And because of that, in the real world, outside of clinical trials, just here in the United States alone, we waste roughly $2.5 billion a year on drugs that end up not working. And we're treating RA by guesswork. "Well, I think this might work better," "Well, I suppose maybe I'll try this drug." And it would just be really good to be able to do a much more targeted, much more personalized type of therapy where we biopsy a patient and learn what they actually respond to. Obviously, this research doesn't get us there yet, but it's a step down that pathway. And so, I guess what I would say to other physicians, both rheumatologists and non rheumatologists, is keep encouraging your patients to participate in research and to help us to look what we can do in the future.

Melanie Cole, MS: That's an excellent message, Dr. Mandelin. So, Dr. Perlman, as we look at personalized medicine approaches and the treatment of rheumatoid arthritis, which is really what you're looking at here, tell us how these findings support the possibility of physicians customizing that drug regimen for their patients based on their specific type of disease, which is what is so interesting in this study. And I'd like you to speak about future directions of this research, if you would.

Dr Harris Perlman: I'm going to actually focus on what we've been doing at Northwestern. When I became chief about eight years ago, Art and I sat down and really thought about how we can look at the future. And the first thing we did is, of course, build this ultrasound-guided synovial biopsy program. We built the infrastructure, the ability to recruit patients. We built the infrastructure to train other rheumatologists across the country and how to use this technique. And what we've done here is we've been working very diligently in working with our community, the rheumatology community, the patients with rheumatoid arthritis, and have been performing ultrasound-guided synovial biopsies on patients prior to any therapy. And then, we do a second biopsy four to six weeks later. And right now, we have an NIH grant to support this work, and we have another NIH grant to support even another aspect of this work, so we have been really fortunate to receive funding for this. And right now, we're finishing off a paper that can speak to identified one or two populations that correlate with response to therapy. And again, this kind of supports what this work has been doing, but bringing it that granularity that Art has been speaking at, is, you know, getting a little better understanding of which of these subpopulations are important. And I think once you start to do that, then we can kind of remove the concept of trial and error that Art was speaking about that when we treat a patient right now, it's guesswork. It's trial and error. "Let's try this biologic, this biologic." There isn't any rhyme or reason for it. And I think, you know, we're very fortunate to have many different therapeutics on the market for patients with rheumatoid arthritis. We want to stop the trial and error for the patient's sake, financially for the patient's sake and every aspect for the patients.

And I think that we are moving towards that way of identifying the important cells and their genetic behavior of these cells, of what genes are expressing and giving a better understanding of what they do during disease and maybe what they're doing during response to therapy. Do they change? And if they change, and again, we want to see that those populations change in their behavior, that might indicate that this is the cells that we are looking at that are going to be the ones that could determine if a patient responds or not.

Melanie Cole, MS: Thank you both so much for joining us today and sharing your studies and incredible expertise and what you're doing that's so exciting at Northwestern Medicine. To refer your patient or for more information, please visit our website at breakthroughsforphysicians.nm.org/rheumatology to get connected with one of our providers. That concludes this episode of Better Edge, a Northwestern Medicine podcast for physicians. I'm Melanie Cole.