This transcript has been edited for clarity.
Stanley Cohen, MD: I am Dr Stanley Cohen and welcome to Medscape's InDiscussion series on psoriatic arthritis. This is episode 1 of the second season. Today we'll be discussing outcome measures, the disease activity measures used in clinical trials as well as practice, and minimal disease activity, what our thoughts are and what our expert's thoughts are on what might be the best target both for practice and clinical trials. I'm really excited today to have with me one of the young stars in rheumatology, Alexis Ogdie-Beatty. She's associate professor of medicine and epidemiology in the Perlman School of Medicine. She's also director of the Center for Clinical Epidemiology and Biostatistics and director of the Penn Psoriatic Arthritis and Spinal Arthritis Program. It's always an honor for me to be in a room with an epi and bio stats guy because that's far above my pay level. So, I'm glad you're here, Alexis; thank you for joining us and welcome to InDiscussion.
Alexis Ogdie-Beatty, MD, MSCE: Thanks so much for having me.
Cohen: Before we start, I'm always curious how young folks get into our specialty. What was your special interest in psoriatic arthritis and spinal arthritis?
Ogdie-Beatty: Well, my introduction to rheumatology came as a third-year medical student. In my internal medicine clerkship, one of the first things that I did was subspecialty clinics. When I was rotating with the rheumatologists over at Washington Hospital Center back then, I saw them taking care of these really complex patients and sorting through things like a detective. And then I had the opportunity to see a young AS (ankylosing spondylitis) patient who was receiving etanercept for the very first time, seeing his remarkable response and return to a more normal life. And with both of those opportunities, it was just like a light bulb went off and that was automatically what I was interested in. As I went through residency, I worked with Dr Schumacher at the University of Pennsylvania and really learned a lot about synovial tissue and then was fascinated by why people develop inflammatory arthritis and then found psoriasis is the ideal place to study why people develop inflammatory arthritis, because we know some of them will go on to develop PSA (psoriatic arthritis). That led to me getting to know PSA and really falling in love with the condition and the patients who have the condition.
Cohen: It's really phenomenal what the targeted therapies that came out in the late nineties did to patient outcomes. They also stimulated a new generation of clinician scientists to enter the field. Otherwise, there wasn't a lot of interest for the decade prior to that. And then you were fortunate to work with one of the giants in the field. Mentoring makes such a big difference. And I know you're doing that as well for others through the ACR [American College of Rheumatology] and the foundation and so forth. So, let's launch into what we wanted to discuss: clinical trial outcome measures and practice outcome measures. We've done a lot of work in psoriatic arthritis in our CRC throughout the years and I am always intrigued that, even today, with studies that we're doing at this moment, the ACR20 response is still the primary outcome. And in the European versions of the study, sometimes the DAS28-CRP is still the outcome measure, and we do look at skin and so forth in a rudimentary way. What are your thoughts on what we should be doing in clinical trials? Obviously, we developed a lot of drugs using that methodology and it worked. I'm just curious what an expert thinks about those outcome measures.
Ogdie-Beatty: That's a great question. Laura Coates and I recently had the opportunity to talk to the FDA about this, and we advised the FDA to consider changing the ACR20 as a primary outcome. There are a couple of reasons why. The ACR20 at 20% response, even in patients with rheumatoid arthritis, 20% response means very little to an individual patient. And so, yes, you can see differences from placebo. But now that we're in a world where we have many targeted therapies and we're trying to think about how we choose between them, that 20% difference from placebo means very little. We want to get better within the current benchmark. And we suggested that we use other outcome measures that allow you to understand what proportion of patients will get to a deeper level of response. We'll talk about those some more. But examples of those are minimal disease activity. We've also talked to the FDA about the DAS28 and how that is really a rheumatoid arthritis measure. In psoriatic arthritis, toes and ankles are involved much more commonly than in rheumatoid arthritis. And that matters when we're measuring the disease. And so that 28-joint count really doesn't capture the activity level of patients with PSA. It's highly correlated but you miss a lot, and then you also are increasing the number of patients in a particular trial who are going to be fitting those particular outcomes. And what that results in is patients with PSA in trials that look like rheumatoid arthritis.
Cohen: What is minimal disease activity? Can you define for our audience what that is and the genesis of that and the benefit of using that as an outcome measure?
Ogdie-Beatty: The minimal disease activity (MDA) counts seven different items. In order to meet MDA, you have to satisfy five of those different items. The items are a swollen joint count of 1 or less; a tender joint count of 1 or less; an enthesitis count of 1 or less; and a PASI (psoriasis area severity index) score of 1 or less or a body surface area, which we really use more commonly in clinical practice, of 3% or less (so three palms, of a patient's palms, on the body); and then a patient global assessment of 2 or less on a scale of 0 to 10; a patient pain assessment of 2 or less on a scale of 0 to 10; and a HAQ [health assessment questionnaire] of 0.5 or less. Essentially, you're incorporating four elements of the physical examination and three elements of the patient's functioning when in pain for an overall assessment of the disease activity. How did this come about? Well, the Outcome Measures in Rheumatology Clinical Trials (OMERACT) circle that we think about in designing clinical trials and what kinds of outcome measures to include back in the previous version from 2006 included those elements. They originally started by just thinking about what is important to patients. Since then, there's been a lot of research on MDA, what it means, and does it track with meaningful outcomes? Does it separate between two active therapies? And the answer to both of those is yes. We have a lot of great data now showing that this is a great outcome to use for trials and also in clinical practice.
Cohen: As an epi and biostats person, if you use the MDA as your primary endpoint at 12 or 16 weeks, would that be a larger number of patients or a smaller number of patients that you would need? Because doing clinical trials, as you know, is extremely difficult now because we have so many wonderful therapy options for patients. I'm curious what sample size you would need to have?
Ogdie-Beatty: That's a great question. So again, it's a binary outcome like ACR20. If we had our druthers, we would love a continuous outcome because then you can use fewer patients and you need smaller differences to be statistically significant. When you talk about any binary outcome, it just takes more patients. But actually, from a sample size perspective, it doesn't necessarily differ a lot because you're looking at proportion difference. Compared to placebo, you may actually see a bigger proportion difference. ACR20 responses in clinical trials of psoriatic arthritis are around 50% to 60% among patients on the treatment arm compared to placebo, [which are] around 25% to 30%. Actually, the placebo rates in our clinical trials are quite high. MDA is usually somewhere around 40% and around 25% or lower, so the difference is relatively similar.
Cohen: I noticed you mentioned that the HAQ is part of the MDA. As far as patient-reported outcomes go in clinical trials, is that sufficient? Do you need SF-36? Do you need any specific psoriatic arthritis patient-reported outcome (PRO) measures that have been developed? Is one really better than the other?
Ogdie-Beatty: Another great question. This also depends on what the use is for that PRO. So here's the back story: the HAQ is not really that great for psoriatic arthritis, and actually we see this in rheumatoid arthritis in clinical practice … it really doesn't change much. The HAQ is very largely driven by the degree of damage and not so much by the degree of activity. So even in a trial, the HAQ doesn't change a whole lot and actually that's where it changes the best. In clinical practice or observational studies, it really doesn't change so it's not a great outcome measure. In fact, the MDA has a modification that allows you to use a PSA-specific disease activity measure called PsAID. And in psoriatic disease, this is becoming a very common outcome measure that you see in clinical trials (the psoriatic arthritis disease impact questionnaire); it has 12 items that talk about all different aspects of the disease, including their skin, their joints, their fatigue, sleep, and pain, among other things. It allows you in clinical practice to scan down and see what's really bothering the patient. And a quick 12-item measure is pretty easy to score as well. Then you have to single out your patients in clinical practice who just have PSA. And that works if you're in a clinic like mine where I'm seeing mostly psoriasis, PSA, and ankylosing spondylitis in a day and not seeing much of anything else. But in a typical clinical practice, the Routine Assessment of Patient Index Data (RAPID3) is still one of the most commonly used in the United States, and I think that's for good reason. It changes really well. It tracks along with how patients are feeling, and it has nice validity within PSA as well. It can be used across diseases, which is one of the advantages for clinical practice in the US.
Cohen: That's interesting about the RAPID3. Let's shift gears and talk about in the clinic. I know you're in more of a specific clinic than most of us. So, a typical patient comes in, has two or three psoriatic plaques, a BSA (body surface area) of 1 or 2%, has two or three swollen joints, and has dactylitis, and you're starting them on therapy. What do you assess at the time of your initial visit? What are you going to have in your chart that you're going to use to monitor disease activity? And which measures do you act on and say this patient is not adequately controlled and needs additional therapy?
Ogdie-Beatty: Great questions. And we have a group called Psoriasis and Psoriatic Arthritis Multicenter Advancement Network, PPACMAN. On that website, we have our templates, so I can tell you what that says and then you can go look at it if you want and put it into your Epic EMR [electronic medical record], if you use Epic. What I do in the first visit is, obviously, a full history, understand the previous treatments, and then also understand what other risk factors there might be in terms of selecting therapy; for example, obesity and smoking, when you're thinking about NSAIDs and cardiovascular risk as well. And then with the exam, I will do a 66/68 tender and swollen joint count and then the enthesitis count. I use a Spondyloarthritis Research Consortium of Canada Enthesitis Index (SPARCC) count but that's 16 point or 18 points actually instead of the six for the Leeds Enthesitis Index (LEI). It's really quick and it takes me less than a minute to do it because I'm doing it along with my joint exam — as I'm examining the shoulder, I touch the humeral head and so on. I do those two things. I do a body surface area. I don't necessarily undress the patient every time. If I know they have established psoriasis, I find the psoriasis and then I might ask them to help me figure out how many palms they might have underneath their clothing. So swollen, tender joint count, enthesitis count, body surface area, and I also look at their nails just to see if they have nail disease, yes or no. But if it's really bad, I'm able to ask them more about whether or not that's bothering them and whether or not we need to consider medication that's going to address that. And finally, I try to get a sense of whether or not they have inflammatory back pain or they might have axial disease, for example, because that will also influence my treatment. As I'm doing all that, the things that I'm collecting are a tender and swollen joint count and the LEI and BSA to put into my MDA. And then I also use a Clinical Disease Activity Index for Psoriatic Arthritis (cDAPSA) because the components are shared, and in order to do that you need the patient-reported outcomes. We actually have RAPID3 incorporated into our Epic, so the patient gets the questionnaire before they come in for the visit. And then I have their pain, I have the patient global assessment, and I have a HAQ, so I can use all three of those components for the MDA, and the pain in the global, I can use for the cDAPSA. So, we have a smart little form with the four elements of the cDAPSA that calculates it right there for me.
Cohen: Have you polled to see what penetration there is in the clinic as far as people actually objectively measuring these features?
Ogdie-Beatty: I think it is very low in general. Just from polling individuals, a lot of people are not doing all the joint counts. And I think there is not enough education about enthesitis counts — a lot of rheumatologists don't necessarily feel comfortable with that, in particular, even fellows. We could do a better job on some of those pieces. But there's also a time crunch and we all know that we're seeing so many patients these days, not to mention all the messages and everything else. We're just trying to get through it. The perception is that the exam takes a long time. In reality, it doesn't take a long time. Once you've done it several times, you can buzz through that exam within 3 minutes. I've timed it and made a video just to confirm that you can do it that quickly.
Cohen: When you're making treatment decisions, what threshold do you use to say this patient's not adequately controlled and I need to make a change or addition?
Ogdie-Beatty: I think this is where the MDA or the cDAPSA can help. if you're in MDA, you probably don't need to make too many changes. And if the patient is in MDA, they are less likely to go on and progress and develop erosions, for example, while they're on therapy. So, it can be pretty comfortable if someone's at an MDA. In cDAPSA, there's a similar threshold. Like the RAPID3, there are cutoffs for remission, low disease activity, moderate disease activity, and high disease activity — very similar to the Clinical Disease Activity Index (CDAI). If you're in remission, again you can feel pretty confident that the patient is doing quite well. I use that information but let's say they're not in one of those two [MDA or the cDAPSA]. Sometimes patients are pretty comfortable and that's where the RAPID3 can be really helpful for getting a sense of your patient-reported outcome. If it's really low and they're feeling great, but you see a few swollen joints like you mentioned, minus the dactylitis, you might say, “How are you doing? Is this bothering you? What's going on?” And maybe make some other adjustments as opposed to switching therapy, for example. The patient that you described has active disease in multiple domains. They wouldn't be in MDA, but they'd be missing multiple criteria. The skin is okay, it sounds like. For some people, if that skin is on the forehead, that 1%, that's still enough reason to want to switch therapy because that can really bother patients.
Cohen: I'm curious. This isn't really so much to do with disease activity measures and outcomes. Do you stratify your patient by domains, by skin or joints, when you're picking a particular therapy? Are you still a fan of or forced to use conventional synthetic disease-modifying antirheumatic drugs (csDMARDs) first? What is your perspective on that?
Ogdie-Beatty: I don't necessarily use csDMARDs first, unless the patient has relatively low disease activity or there's another reason, such as they don't want a biologic. In Pennsylvania, we can generally get therapies that fit the patient early on, a biologic therapy, for example. So, if they have dactylitis and multiple active joints and maybe moderate to severe skin that's really bothering them, they're already a candidate for biologic therapy. I will often go with biologic therapy first in that circumstance.
Cohen: Are there any data to suggest any difference as far as dactylitis or enthesitis between IL-17 and IL-23 or TNF inhibitors or JAKs. Is there anything that discriminates? I know in skin, we believe that the 23s and 17s are better for the skin. I am just curious about the enthesitis or the dactylitis?
Ogdie-Beatty: I would say there are not substantial data. There is a trial comparing ustekinumab to a TNF inhibitor, which found that it was slightly better for enthesitis in that particular trial. But overall, there's not really a lot of great data comparing dactylitis or enthesitis that would be convincing enough to say you must go one pathway or another. I think with dactylitis we want to be a little more aggressive. Being aggressive with the NSAIDs can be really helpful for both of these two pieces [dactylitis and enthesitis] as well. You just want to make sure you're trying to get it under control sooner rather than later.
Cohen: One question I always found intriguing is that when looking at your DAPSA (Disease Activity in Psoriatic Arthritis) and your MDA, there's no physician global or MD global assessment. Why did the psoriatic arthritis community or psoriasis community choose not to include that? Obviously, it's included in all of our disease activity measures in rheumatoid arthritis in general, other than the RAPID3.
Ogdie-Beatty: So that's been a data-driven piece. It fell out when they were creating the DAPSA. It didn't seem to change it at all, and it didn't have as good a responsiveness as the other measures in that tool. So physician global [assessment] was not included there for that reason. But there have been a number of studies [showing] that physician global doesn't necessarily track well with patient global but tracks fairly well with tender and swollen joint counts. So, because it's already captured in those other things and not a great measure in PSA, it wasn't included in the OMERACT Filter either. So, it's not in our OMERACT circle of things that have to be included.
Cohen: What would your message be to the busy community rheumatologist who is following a patient with psoriatic arthritis? What would you strongly suggest that they monitor in their clinic as far as patients with these diseases?
Ogdie-Beatty: What I strongly suggest is making sure you have a PRO, whether that's a RAPID3 or PsA Impact of Disease (PsAID) or Patient Reported Outcomes Measurement Information System (PROMIS) measure. I strongly suggest that so you can track things over time and then I strongly suggest following your joint count. I think if you have the PRO and the joint count, the cDAPSA is really easy. It is basically a CDAI that you're already used to doing for rheumatoid arthritis. If I can give you the minimal thing to do, I think it's following a patient-reported outcome and the tender and swollen joint counts. And I think you'll be surprised at what you find from tracking these data over time.
Cohen: That's incredibly helpful. This has been fascinating and I appreciate you being with us today. And it's, again, been a great help to all of us in the clinic who see these patients. I want to thank you for being with us. And I look forward to another great discussion in episode 2 — this is Dr Stanley Cohen for InDiscussion.
Medscape © 2022 WebMD, LLC
Any views expressed above are the author's own and do not necessarily reflect the views of WebMD or Medscape.
Cite this: Psoriatic Arthritis and the Disease Activity Measures Used in Practice and Clinical Trials - Medscape - Jul 20, 2022.