Rheumatoid Arthritis Topic Panel: EULAR 2026 Save

Hello everyone, welcome to RheumNow. This is our post-EULAR 2026 panel discussion on rheumatoid arthritis. I'm joined today by three of our faculty who were at the meeting in London with me. We're all back home rested and have considered what we thought were some of the best abstracts in rheumatoid arthritis that we want to discuss for you today.

So I'm going to begin with introductions. I'm Jack Cush from Dallas, Texas. Janet — Janet Pope from the other London, Ontario, Canada. John — Jonathan Kay from UMass Chan Medical School in Worcester, Massachusetts. And Bella, hi. This is Bella Mehta from New York. Very good.

Okay. So let's begin with Janet again. I think RA was very well represented. There was a lot of RA content at this meeting. It seems like, you know, the buzz always comes from CAR-T and PsA and whatever, but if you look at volume, you know, I think the number was over 1,700 — almost 1,800 — RA abstracts. So we're going to hear eight of the best. Janet, you're first.

Great. So we always wonder about what's better than something else. So I picked a head-to-head trial. I'm not going to give you the results of the head-to-head because I'm not sure I know them, but it's still a good trial. So this was called the BOACH trial, and it was OP0204. And basically the question here was: what's better in rheumatoid arthritis, csDMARD-IR, randomizing a patient to a JAK — in this case it was filgotinib — or randomizing them to a TNF, and I think it could have been any TNF. It was open label, a pragmatic trial done in the real world in a clinic setting. So that's what the question was.

Now they also decided, rightly so, that they would collect the group of people who were eligible who didn't want to be randomized and look at their choices and look at their satisfaction and see how they did. So it's a parallel universe — the randomized people. They actually had a sample size calculation and achieved it, and interestingly about 100 people were willing to be randomized versus 101 not, or something like that. It was about 100 per group. So of those randomized, about half got a TNF and half got filgotinib. Of those not randomized, the choices were kind of similar between TNF and filgotinib.

So you go, okay, well what are they going to really answer here? So what they told us were, I think, really a few important things: the patients choosing did better with satisfaction. Well, I think that might be expected because they made a choice. They did better though with retention when they chose the treatment, and they actually did a deeper dose or change in disease activity than the people randomized. And I thought — I do clinical trials all the time — I thought people like to be randomized because they go, "Doctor, if you don't know what's best, let's just go for it," you know, if we're in equipoise. So I thought it was kind of interesting that the satisfaction didn't surprise me.

So they had a standardized thing. It was like a little video of a nurse saying you could take this drug — a TNF — or this drug — a JAK. In this case it was filgotinib; the JAK would have been the one people would have had access to by usual care. Everybody was able to make a decision within 24 hours. Of those who refused — like did not want to be randomized — but I think it has implications for future trials because we don't do this parallel arm of people refusing to be randomized and how they do. So I thought that was really cool.

But here's the bottom line: they didn't tell us how they did within the trial that was powered adequately to say what does better — TNF or filgotinib — on rate of response, retention, satisfaction. So maybe they'll present that at the ACR. I don't know.

Yeah. Open label. I think the results were going to be something just like SUNSTAR, something we were not going to present, but I'll say it here now. OP0205. That was the same trial, 200-something patients who had failed a biologic or a TNF, and then they were randomized to either receive open label tocilizumab or abatacept, and they looked at a 52-week outcome. And in the end they all did well. They came in with CDАIs of 25 and they dropped their CDAIs down to 10. And they all did well with no significant difference between them. Although it was designed to be a superiority trial for tocilizumab, it wasn't superior — except unless you did some funny analyses, or if you looked at tocilizumab monotherapy, which was better than abatacept monotherapy; when you added methotrexate it didn't matter which drug you got.

So it was the same kind of thing, and I think it goes to the point of EULAR guidelines: one and done — once you fail a targeted synthetic or biologic, you change classes. And that's what happened in your trial. It's what happened in SUNSTAR. But the dangerous part is what you said: give patients a choice and they do better. Why don't we see that in clinical trials? Because there's no choice in a double-blind randomized placebo-controlled
trial. You got it right. Uh but but the patients were I mean they also knew about the black box warning, right? Which came in in the middle of the trial. So that might have changed the choices, right? Um there's a lot of factors at play, but they said the patients were quite the reasons that they chose were all over the map. My friend like sometimes I wanted it, my friend was on this and liked it or I heard something bad about this one, so I want the other one. So there was no consistent reason why they chose. Oh, I like an oral. Oh, I was on a needle, so I want another needle. Like, the choices were as varied as the patients in the I refuse to be randomized group.

You know, in a clinical trial, they got to show A is better than B. And your comparator can be placebo or an active comparator. It can't be patient choice because that's too outside the box and will screw up your results or could very well screw up the results. And that's why you need studies like this to make this point. But you'll never see it in a drug registration trial.

John, what do you think? I think it's an interesting trial. I don't quite know what to make of it. Uh, in terms of the outcome, I think they said that patients tended to prefer the JAK inhibitor because they like oral medications rather than needles. Uh, but there's a lot of expectation bias in that trial. Uh it's such a completely different approach to trials that I really don't know how to interpret it. If I was going to start all over in rheumatology, um I would rename my clinic the High Expectations Clinic and we would be doing out a lot of placebos and we would do just as well as everybody else in Dallas. Placebo is a very powerful thing.

Let's move on. John, what did you have that you liked? So I there a whole bunch of things I liked. I thought I'm going to talk about OP209 which was a negative trial that was presented by Iain McInnes. This was a safety and efficacy trial uh looking at balinatunfib. Uh it's really a tongue twister. This is an oral TNF receptor one signal inhibitor uh that was studied in patients with moderate to severe rheumatoid arthritis. This study which was negative was interesting to me for two points.

Uh one is that it's an oral molecule, a small molecule that binds to the TNF-R1, the TNF receptor type one, and blocks binding of TNF to that receptor. So it's using a small molecule approach to block TNF signaling uh and not by blocking signal transduction. As you know uh the JAK inhibitors don't block TNF signaling because TNF signaling is not mediated through Janus kinases. But anyhow uh that's one reason why it was interesting.

The other was that it was a negative trial because the placebo response rate was so high. It was conducted all over the world uh including South America where there's a very high placebo response rate and also uh Eastern Europe. Uh but the study was a phase 2B 12-week multicenter randomized double-blind placebo-controlled dose-ranging study looking at — I'll try it again — balinatunfib uh 200 milligrams twice daily, 200 milligrams once daily, 100 milligrams once daily, 50 milligrams once daily, or placebo. And it was 2 to 2 to 1 to 1 to 2 randomization so twice as many patients were on the 200 milligram dose regimens and placebo as were on the 100 milligram and 50 milligram dose regimens.

The primary endpoint was ACR20 response at week 12. Uh key secondary endpoints included ACR50 at week 12, change from baseline to week 12, and DAS28-CRP. Basically uh the placebo response rate was so high that none of the clinical endpoints differed from placebo. They did look at the decrease in blood IL-6 as a pharmacodynamic marker and this decreased with each of the doses suggesting that there was biological activity to the molecule, but in terms of the clinical trial it was a failed clinical trial. Interesting in that the molecule has a unique approach uh to blocking TNF signaling and also um the molecule deserves to be studied again if it can be studied without such a high placebo response rate.

I had to look it up and you're right it's hard — come on, Jack, you can do it — balinatunfib — um we're gonna — it's balinatunfib — we're going to spend the rest of this podcast getting the name right.

What's interesting about this is um that Iain also presented another oral cytokine inhibitor in um icotrakindra which is the IL-23 oral agent from J&J for psoriasis that's being tested in psoriatic arthritis. So there's an oral IL-23 drug that does look good.

The other interesting thing about this — and I love that it's an oral drug — but my I got my start in rheumatology, I went to a meeting, very first investigator meeting I went to was like in the Swiss Alps. Had to take a train and whatnot. It was on lenercept and lenercept was the etanercept equivalent that targeted p55 and it failed for a number of reasons. Not for efficacy but a number of reasons. But that was the first TNF drug I ever saw but it
never made it to market and it had a lot of Achilles heels. Maybe one of the issues here is that P-55 may not be a really good target and they might have to find that out the hard way. That's true.

Uh in terms of the high placebo response rate, probably about 15 years ago at EULAR I presented an oral abstract with about a p38 MAP kinase trial which had enrolled about 200 patients. 100 patients were from Western Europe and North America — there was one from North America, 99 from Western Europe. The other 100 were from either South America or Eastern Europe. The comparison was between North America/Western Europe versus Eastern Europe/South America, and there was a dose response in the Western Europe, North America group, 100 patients, but no dose response because of a very high placebo response rate in the South America, Eastern Europe group. So the company never wanted to go and publish it because they didn't want to offend potential clinical trial sites from those areas. But it was essentially a controlled trial looking at two different study site regions in the same trial, and it showed that there was a dose response in the North America, Western Europe sites but not in South America, Eastern Europe. So the fact that this study was conducted at places where the placebo response rate was so high, it was somewhat doomed to fail.

A sort of a question there — like how do they measure the placebo response? Is it patient reported outcomes or is it like objective swollen and tender joints, because that makes a huge difference? Oh, it's ACR — I'm sorry, it's ACR20. It's all of the outcome measures; the placebo response rate was high. But the problem is — the question is — are the patients that are entered into the study truly on a steady dose of methotrexate, or might they have been starting their methotrexate when they enrolled? The charitable interpretation is that they became more compliant when they were in the clinical trial. [clears throat] I'll leave the other explanations up to people's imagination.

Okay, let's move on to my selection. RA bridge, RA branch. Um, I think this might have been the most important study in RA because it's an incredibly large study. It's a controlled study. It was FDA mandated. You may not be aware of this. This is LB00009 — that's three zeros and a nine — presented on the last day by Peter Taylor, and this was a post-regulatory commitment by Lilly to study the risk of venous thromboembolic events, VTEs, in patients receiving baricitinib. Baricitinib was approved by the FDA only at the lower dose because they were worried about problems with the higher dose, and it was the first JAK inhibitor — actually the only JAK inhibitor — that when it launched had a risk of VTE right from the start.

So, like with tofacitinib, which had a regulatory commitment to do the ORAL Surveillance study, they had this commitment to do a large study. And what they did was pretty much the same as ORAL Surveillance. They had entry criteria for high-risk individuals, and that was going to be people who had either a prior history of VTE, over the age of 60, a BMI greater than 30, or a BMI 25 to 30 with some other criterion. But just think about that — that's cardiovascular/VTE risk and obesity and age, not that dissimilar from ORAL Surveillance. And the people that went into the study were people who were either DMARD/biologic naive, or they could have actually received a prior biologic or targeted synthetic. Two-thirds were biologic or targeted synthetic naive, and they were randomized — over 3,500 patients were randomized — to receive either baricitinib 2 milligrams a day, 4 milligrams a day, or a TNF inhibitor. So there are three arms, and they followed them until they got a certain number of VTEs that they could adjudicate on.

And it turns out they didn't need to get to the end of the study. The study was about three and a half years of exposure. And in the end, baricitinib at all doses — two and four — had significantly more venous thromboembolic events, DVTs or PEs, than did the TNF inhibitors. But unlike the ORAL Surveillance study, the JAK inhibitor did not increase the risk of MACE events — major adverse cardiovascular events — mortality, arterial thrombosis, or opportunistic infections. There was a significant but small risk of serious infectious events with baricitinib over a TNF inhibitor. So that's the take-home. It's different than ORAL Surveillance. It confirms — I think firmly stamps — the risk of VTE with these drugs. The question is, does it muddy up the concerns about cancer and cardiovascular risk? And you could say that, well, they're two different studies — ORAL Surveillance and this study — with two different entry criteria, but are they really that different?

So I want commentary, but let me just quickly go around and ask each of you: is this going to change your thinking about oral
surveillance? Let me start with Bella. Yes or no? And why? Uh, a little bit, but I I think that this most importantly tells me that patient selection for JAKs is going to be the key. Uh there was another poster uh 0145 POSOS0145 which basically said obese patients don't do as well on JAKs. Um and there was this was like pooled trials from three three different trials. So uh basically it's just risk profiling uh putting in the right patient for the right drug. Janet. Um I already think that JAKs are pretty safe particularly in younger people, non-smokers, um etc., etc. So I think it didn't help that like it's not a nail in the coffin for baricitinib because of course in North America we can only get two milligrams anyway and and US you can only get it after TNF. So I don't think it it doesn't help but I don't think it changes my mind on stuff. It's just it's kind of too bad. It's too bad that it was because the non non-inferiority was almost reached on VTE which is a primary outcome. It's the point estimate was where it should have been. The confidence interval went a bit beyond that. And doesn't that sound like ORAL SURVEILLANCE? the primary outcome was the same idea that it was numerically higher on the MACE in in ORAL SURVEILLANCE but the standard deviation was there but it was a bit too far out on on the bad guy end so that's the problem with these trials um figuring out the confidence interval around it but numerically more infections bit more cancer more VTE no increase in MACE so it does depend on patient population. But um well again the only thing that's not non-inferior here is the VTE VTE right and the the non-inferior margin the upper limit was 1.8 um and um and so that was the case for the primary endpoint a few of them did go higher than 1.8 MACE didn't mortality didn't at did but the you know was really low event. So yeah, right. Um and opportunistic was a really wide confidence interval. So again, um I I I think this makes me feel better about what I do, which is I'm fairly liberal on this. Don't give it to really old people with heart disease and smokers. You know, it's like then that's like 8% of my population who might get a JAK. Anyway, John, what do you think of this data? And I think these data are confirmatory of ORAL SURVEILLANCE other than for the VTE. My take on ORAL SURVEILLANCE is that TNF inhibitors lower the risk of MACE and lower the risk of cancer and that uh in that case uh tofacitinib didn't lower it as much as the TNF inhibitors. Uh so here you know I think it's also similar that the risks that are lowered by TNF inhibitors were not lowered more by baricitinib. Baricitinib is a different molecule from tofacitinib. Tofacitinib less selective. Baricitinib is a JAK1/2 inhibitor. Upadacitinib which isn't even being talked about here is a JAK1 selective agent and then uh the Chinese drug is even a more selective uh JAK1 inhibitor and filgotinib JAK1. So yeah I I prescribe JAK inhibitors. Uh I take a JAK inhibitor. I'm not as worried about uh JAK inhibition except in people who are who have lots of cardiovascular risk factors but all the registry data seems to have gone against the conclusion of ORAL SURVEILLANCE. So in the quote unquote real world uh we're not seeing what is observed in these postmarketing uh requirement studies. You can't get a drug approved and you can't get a label changed on registries and real world data. It could be part of the development plan, but it can't be the pivotal trial.

Um, I I would I want to just clarify something that John said, which I agree with 100%, which is there's very very good data that TNF inhibitors are very effective at lowering cardiovascular endpoint outcomes. Um, and um and and uh I don't know about cancer outcomes. We're going to talk about that next, but it's unclear that they lower TNF inhibitors lower the risk of VTEs. I think VTEs are a separate phenomenon. And I think this study proved that really well by showing uh I would argue the ORAL SURVEILLANCE data. It wasn't that the JAK inhibitor was worse than the TNF. It's just that the TNF inhibitor was better than the JAK. And that could be a contrary argument that we go round and round about um that you can't disentangle from the design. But here, this was designed to look at the VTE and I think it shows it quite quite strongly. Um, we could go on on this forever. Let's not.

Um, Bella, what's your your uh big one for the day? So, uh, talking about uh the right patients doing well on the right drugs. Uh, this is um abstract uh OP0112. Um this is looking at real world data the RISE registry which is the ACR registry uh of around 60,000 patients with rheumatic uh conditions um and uh they looked at two GLP-1 receptor agonists semaglutide and tirzepatide uh and both of them are rapidly adopted in the past few years as we all know um one thing that struck me was that around 21% of rheumatoid arthritis patients were exposed to this and then around 13% of osteoarthritis patients were
exposed to it at some point. Especially amongst a population which has a high baseline obesity — the mean BMI was around 36 of those who were exposed here. And both agents caused or helped with clinically meaningful weight loss, with tirzepatide achieving more reductions, which is known because of the mechanism of action. And even non-diabetic patients were losing more weight than diabetic patients. Again, weight loss seemed to plateau at 8–12 months and regained a little bit by 18 months. Again, that happens when either there are breaks in the way they're taking it and/or some sort of resistance. But baseline disease activity burden was higher in these patients who were obese diabetic. Patient reported outcomes were similar, but they also say that maybe there is improvement of patient reported outcomes as well as disease activity measures in patients who were on GLP-1 medication.

So I think that this just shows that — again they didn't show all the data on pain, function, and all of that — I'm sure we'll see it at ACR because this is like they're sort of deep diving into this analysis, Jeff Curtis and the group. But what it shows is that we'll be prescribing these drugs along with all these DMARDs very very soon, because in the right patient population they might be game-changing.

Yeah, we discussed this a few days ago in our panel on PsA — that a high number of our patients are going to be on this. You said 21 and 14% of that obesity group, right? Yes. Of the obesity group. Okay. Yeah. In the US, the general population, 8% of adults are on GLP-1 drugs. So I don't know how that equates to the obese RA and OA population, but it's still a high number. And it says very clearly that you the rheumatologist need to know about these drugs, how they're to be used, whether they can be used in conjunction — and they can — with everything that you're doing, and whether they will be adjunctive, and I believe that they will be.

So I think this is all encouraging, but the companies have got to step forward like Lilly stepped forward in doing the TOGETHER PSO and the TOGETHER PSA trial. A well-designed controlled trial with an active control that tells you what the impact of these drugs are and how safe they are. Without that, it's all anecdotalism, and we've been talking about GLP-1s now for two years. So it's about time we get larger data.

And this is good — I mean the RISE registry is big, but I don't know how they're going to — could they do a target trial emulation if they have that many patients on drugs? Could they not do a target emulation trial from the dataset to make some kind of inferential message? Again, it is not what it was designed to do. It's a little bit like looking at large-volume registry data — it's really hypothesis generating. But until we see the designed trials, I don't know how else to proceed on this. And whether — I know I'm prescribing GLP-1 drugs to my patients when I can get away with it. The vast majority of rheumatologists are not, but some of us are. It seems like it's more in the PsA world than it is in the RA world. John, what do you think?

Well, to your comment about the PsA world, that's because more PsA patients are obese than rheumatoid arthritis patients, probably, because of the association between obesity and psoriatic arthritis. These drugs are amazing and they have pleiotropic effects. They're anti-inflammatory. They lower cardiovascular risk. They lower the risk of obesity-associated cancers. They lower blood sugar. They lower blood pressure. They lower serum lipids. They change diet, they change the brain. So I'm not at all surprised that patient reported outcomes are improved because of the cognitive effect of these drugs. Also the patients are feeling better, they're more active.

The biggest problem is that if the drug doesn't have an indication that the insurance company likes, they're not going to cover it. So patients who are being prescribed GLP-1s who don't have diabetes — it's very hard to get it covered. Even FDA-approved indications like obstructive sleep apnea — insurance companies are saying sorry, your hemoglobin A1C isn't high enough, you don't have diabetes, we're not going to cover it. The cost of all but the lowest two doses — 2.5 and 5 milligrams of tirzepatide on Lilly Direct — are $449 a month. So for the vast majority of patients that price is probably out of reach.

I've recommended GLP-1 agonists to many of my patients and those that have taken it feel great and they're really pleased with the results. Problem is once they're no longer covered and they have to stop it, they gain the weight back. So there really has to be a societal reconsideration of these medications. They seem to be so good, with the only real
side effects being constipation and muscle loss, which can be addressed with frequent liquids, stool softeners, and weightlifting exercise and supplementary protein. So you know this is a medication that is important to society. I don't want to go off on a tangent and talk about single-payer systems versus our system, but if we had a single-payer system where the cost of treating the complications of all of these problems was the responsibility of the same payer who was paying for the GLP-1 agonists, it wouldn't be a problem. That's why other countries are paying for them. But in our country, the average time on a given insurance plan is only a couple of years, which is why insurance companies make the business decision to not cover preventive therapies because they're not going to have to pay for the treatment of the complications.

Janet, how's this being handled in Canada? So, it's highly variable because although we have global health insurance, we have multipayers. So the drugs are approved — of course for weight loss they're reimbursed in some private plans, not others; they're not reimbursed in the provincial plans, and the provincial formulary is for the poorer people, the older people, and so people in the middle might or might not have coverage. So it's a big deal. But when we talk about the obesity epidemic even in rheumatoid arthritis, now in the BRASS cohort in the US and our Canadian early arthritis CATCH cohort, one-third of the patients are overweight, one-third obese.

So here's my question, which isn't really the question that you're asking me, but my question is: what about the patients that are normal weight without diabetes? Are they going to get the benefit, or is this all weight loss, because they're highly effective in weight loss? That's like the million-dollar question. It's a heavy question to answer and nobody wants to do the studies in normal weight because that's not going to be an indication — I don't think — as an adjunctive to DMARD or to biological drugs in our patients with RA or PsA.

Go first. Go ahead. Yeah, I think Janet it depends on — you know there are studies going on and they're trying to find — there are receptors in the synovium even in non-obese patients. They're everywhere. They're ubiquitous, right? The receptors are in synovium. Somebody presented on that too, on a biopsy study or something. But I think the receptors are all over — like these small molecule peptides, or I'll call them gut hormones, are way beyond the gut, as Jonathan had just said.

Yeah, but again I talked to some colleagues from other countries and it's $25 a month for semaglutide there, right? And this is out of pocket, not covered. So there are a lot of cheaper options coming into the market. So things might change. $25 or $30 a month is much much more affordable. And I think that if — I know these indications are not there right now but in a few years they will start coming in. I mean, if I have rheumatoid arthritis patients whose disease activity is completely — you know, they're off meds because they're just better. So in certain cases it does work phenomenally. I had a patient with peripheral spinal arthritis who didn't respond adequately to methotrexate alone, didn't respond to TNF inhibitors, was on an IL-17 antibody — I pushed that to the maximum dose with improvement but still incomplete response. Even though she was normal weight, I put her on tirzepatide 2.5 milligrams weekly and that pushed her over to responding to the anti-IL-17. So now that's an N of one anecdotal experience. But I think that the biology behind that makes sense. Certainly with receptors on synovium, but also the anti-inflammatory effect, which is probably more than just binding to receptors in synovium. The problem is going to be that payers aren't going to be willing to pay for it.

Yeah, I agree. And also as you get more and more of the double combo — I think there's even one that is proposed, or maybe even out, a triple combo where it's GLP-1, GIP, and I forget what — you know, the innovation will cost more money too.

Um, how about we change over to maybe our rapid round, and maybe I'll start. So I was going to do OP0206. I think Jonathan you might have actually presented this in one of the RheumNow things I'm covering — you were on last week. So this is the randomized controlled trial of split-dose methotrexate 25 milligrams a week split dose versus 25 milligrams subcutaneous once a week. I really kind of like this study because it answers a question. So we already know that oral methotrexate at the higher doses isn't as well absorbed. We also know — even though I forget to prescribe it this way — that split-dose methotrexate divided over 12 to 24 hours is better than the full dose at once, even though I forget to prescribe it that way. And there's been this thinking that once
you're above 15 milligrams of methotrexate that subq will do better than oral. So this was a randomized control trial six centers in India um I think 205 252 patients and so it was to be a non-inferiority where they said split dose methotrexate oral 15 milligrams AM 10 milligrams PM once a week should be about the same as 25 milligrams subq once a week and um you could add a DMARD at 16 weeks if things weren't going well so it was a very good pragmatic study. So I guess the drum roll of the results is that 25 milligram subq methotrexate was not non-inferior. It was actually superior to 15 milligrams AM 10 milligrams PM of oral methotrexate. So 25 subq was better on the primary outcome. So it wasn't inferior. It wasn't non-inferior, it was superior, but also some of the key secondary outcomes.

And you know what, in Canada there's a lot of subq methotrexate and I think it will change my practice because I often do oral methotrexate because I often start on combo csDMARDs and I think it's a bit overwhelming. So I liked it. I thought it was something that's a good take-home message for me to maybe change my ways.

I agree and I did a RheumNow video while I was in London about this. There was a plenary presentation at the ACR meeting uh last year where they pointed out that split dose oral was better than not split dose oral. So with that plenary that certainly changed my practice and above 15 milligrams and certainly above 20 milligrams I go to split dose. Uh but this was really eye-opening because split dose I thought was the answer. Patients would rather take oral methotrexate than give themselves a jab. Uh and for a while uh subq methotrexate was unavailable. So I switched most people back to oral. Uh but it points out that at doses of more than 15 milligrams weekly the receptors in the intestine or in the stomach uh for the methotrexate are saturated.

Uh so just today I had a patient on 20 milligrams of oral methotrexate once daily who was doing better but not completely better and rather than pushing her up to 25 I had her split her dose. I had her take 10 milligrams in the morning and 10 milligrams in the evening based on this and the ACR plenary. So this certainly is one of those abstracts that changes practice.

I don't believe these results. Okay. I believe them. I don't because number one, there's so much data that shows that oral split dose is as good as parenteral as far as actually — oh, there's a lot of very good PK studies that show on limited populations. And here the issue is did they — how do we know that they or that the oral tablets were taken?

Well, that's what I was going to say, that I think that subq you either took it or you didn't. My fear is when I have a patient on split dose they're not going to — I don't think we want them taking it on Monday and Thursday and so if they forget Monday evening and they go oh Dr. Pope said don't take it far later — that I wonder what the compliance is, not like the the forgetful non-adherence for the second dose.

But I think in general the idea makes sense. The pharmacokinetics even of split dose, absorption or not, I mean it's highly variable, that's the problem. It's not always obvious in a patient, not when you're talking at doses um of less than — when you're at 15 that's when the wheels are off the tracks as far as absorption and if you're waiting for 20 and 25 to split your dose you you missed the opportunity at 15. Anybody's at 15 should be on three pills Thursday, you know, with at least you know 10 hours between, because at seven and a half and 10 the absorption is very very good. And again the PK studies, area under the curve studies, are really very very clear about this.

And so this open label in India where you're doing the 15 and the 10 I think — anyway, that's by the way that's the same design as the SMART study that John alluded to that was a plenary at ACR that showed that oral split dose was equal to parenteral. No, that was equal — it was superior to um oral once a week. I don't — they didn't do a subq. You're right. Right. It was — but that's why I think you know methotrexate, old drug, sort of sometimes new ideas. Um I change my practice and then I don't — like I flip-flop a bit. So I get where people are coming from.

All right, we got to move on, John. So I'm going to talk about another uh study where you might criticize where it was done. Uh this is OP208. This is a study that was done in China uh randomized double blind placebo control phase three clinical trial that evaluated efficacy and safety of Zmporitinib which is a JAK1 oral small molecule uh drug. You know why do we need another oral JAK? Well, we have upadacitinib which is JAK1 selective in Europe and the rest of the world they also have filgotinib. So this would be the third one but according to the presenter this is more selective, very highly selective for JAK1 with very little off-target. Uh it was a one-to-one randomization 12 milligrams twice
daily of the zempitanib versus placebo in patients on stable conventional synthetic DMARD background therapy and the primary endpoint was ACR20 at week 24 — and I think it was at week 24, don't want to — but anyhow, by week 12 there was clear separation with zempitanib, much better response, rapid response compared to placebo. ACR20, ACR50, ACR70, DAS low disease activity or remission, CDAI low disease activity or remission, DAS28 remission. And the safety profile was comparable to placebo. So very positive study suggesting that this very highly selective JAK1 inhibitor is effective. I don't think it's going to take off in the United States because people are so used to the JAK1 selective agent that we have, and the commercial market in the United States is much more complicated than that. But it's nice to see that there's a trial where active drug separates nicely from placebo.

Yeah. I mean it's exciting and it's a lot like the other — I think it's the same molecule coming out of China, another JAK1 inhibitor that showed it was superior, looking at ACR20 and 50 and 70, but really really high ACR20s. If they want real respect they're going to have to do a global study. They're going to have to do this outside of China involving a lot of other populations. But right now when it's done just as this is — why tell me about it if it's not going to be a tool that's going to be available worldwide? I don't know. I think it's always skeptical. Yeah. Question of reproducibility always. There seems to be virtually no placebo effect, pretty safe, no dropouts or little dropouts in China compared to rest of the world. So you're right, we've got to see something in other populations to see if it's reproducible. But there you go. Who knows?

So I want to present a great abstract — poster 0083 — coming from Askling and colleagues in Sweden. And the prelude to this is many of you believe that there is a risk of lymphoma with TNF inhibitors. I don't believe that. It's in the package insert, but you weren't at the FDA hearing where they looked at the first three TNF inhibitors and showed that the risk of lymphoma — mainly non-Hodgkin's lymphoma, B-cell lymphomas — had an SIR or relative risk of 2.5 to 6.3. And they showed that that SIR, that increased risk with the first three TNF inhibitors, was exactly the risk of RA patients developing lymphoma in the pre-biologic era. Hence, you've got enough inflammation to cause cancer — then it's an RA-related risk — and patients who are sick enough, inflamed enough to get on a JAK inhibitor or TNF, it doesn't matter, they're going to have some signal there about lymphoma.

And then to back that up, at the same time there's a famous study by Baecklund et al. showing that the risk of lymphoma in RA goes up with disease activity, disease severity by functional class, and also by inflammatory load — and I'm talking like an odds ratio of 71 with the highest levels of inflammation. So the question therefore is: if you then control disease activity with effective therapies, especially TNF inhibitors, won't you lower the risk of lymphoma over time?

And leave it to Askling and their colleagues to do marvelous work using the national Swedish registries. They had almost 100,000 RA patients matched by birth year and sex to one-to-five general population controls looking at lymphoma risk, and they showed that the RA lymphoma risk was very high around 2000 — higher than the population. My fingers are apart. And then you look over time, the RA population stayed above the general population, and it kind of goes up and varies year to year, but the RA population was always more than the general population. And they did the hazard ratio, which was around 2 to 3 — kind of the same as, you know, relative risk, odds ratio, relative SIR. And then they showed in different years, starting in the mid-2000s, the rate of biologic use, and it was going up and up and up. They prove the point that greater biologic use didn't lower the lymphoma risk, but it didn't increase it either.

So it doesn't give me the data I really expected, which was more effective therapy lowers inflammation, lowers risk. That's not happening. There's other factors in play. But I'm going to play the advocate and say the exact opposite occurred here by what they showed — they didn't prove in any other way, in any new way, that biologics and TNF inhibitors increase the risk. The risk stayed the same even though the rates of advanced biologic use went up.

So this is, I think, a really interesting study. I still think you have to talk about lymphoma risk in your RA patients. And if you want to talk about it when they go on TNF, go ahead, knock yourself out. But I'm not changing my therapy and what I want to use if there's a story of lymphoma involved. I let the patient make up their mind.
and give them the information. So you make a really important point there Jack that uh lymphoma is a marker reflecting past disease activity and biologic use is initiated in patients with high disease activity despite previous therapy. Ava Beckan's paper which I think was arthritis and rheumatism around 2000 looked at the Swedish RA register and looked at lymphoma and they found they divided them into deciles of disease activity and it was that highest decile the 10th decile where the rate of lymphoma rose tremendously I think you're right 70-fold increase and so the patients that are going to be on TNF inhibitors are the ones who have had high disease activity before they go on the TNF inhibitor which is probably why you're seeing more lymphomas perhaps in that group. Um but it doesn't mean that it's a causal effect and in fact the control of disease activity seems to lower lymphoma risk. So if you put those patients on TNF inhibitors early you're probably lowering the risk of lymphoma. But uh the point that you bring up about uh lymphoma risk being associated with active inflammation and duration of active inflammation is most important. So I want to be clear with the audience that it doesn't affect my thinking because I I have thought about this much much more than most people. Um but you may be guided by the package insert. That's fine and you could have that discussion about the risk of lymphoma with a TNF inhibitor or any drug. The good news for you is you got lots of options and that's that's that's really the other good news. But it's it's based on this study and what's gone on before this is a very murky issue.

Janet, you were going to say I was just going to say one we devil's advocate. It's probably also the HLA-DR4 of some of our patients and whatever triggered their RA. So we'll just blame it on a virus. But because there there are data of two things that um non-Hodgkin's lymphoma and MACE events have an elevated relative risk highest in RA from a year before RA to a year after. But of course the cumulative effect goes up over time. So I think it's also whatever the insult in your body has possibly allowed you to get both a chronic uh autoimmune inflammatory disease like RA and has also allowed your B cells to go a little bit or T cells to go a little bit unchecked. So yes, cumulative um high high disease activity for decades has higher lymphoma, but getting RA there's some kind of polygenic environmental thing that might increase your lymphoma even in our well-controlled patients. So I think we're always going to have an increased risk in our RA population compared to the general population even when we get a lot of people into sustained remission. So [snorts] just a wee devil's advocate way of looking at it as well.

That's fine. Um, Bella, give us your your last quickie. Uh, so this is, uh, OP011 and, uh, you know, I I especially thought of this because we're talking about all these drugs, um, and how they help with patients, but fatigue is a big one. Uh, so um, you know, uh, this was a multicenter real world study with RA patients. Uh and fatigue was highly prevalent affecting around 43% of the patients in this uh 200 or 300 patient group um and this was associated with high disease activity as well as comorbid depression. They did remove or exclude patients who had chronic fatigue syndrome. Importantly uh and that's what my my takeaway was that there was no significant differences in fatigue prevalence across uh csDMARDs, biologics uh or small molecules. So suggesting that treatment class may not meaningfully impact fatigue levels and maybe that's a whole separate pathway and a whole big problem that we are not completely addressing. Um and again again patients who had higher fatigue uh were patients who were on high glucocorticoids too and again maybe that is also a marker of disease activity. Uh but overall uh it just shows that we're treating uh sometimes inflamed joints, swollen joints but you know fatigue which is multifactorial and um can be game-changing for patients still has an unmet need uh in terms of even medications or even non-pharmacological multidisciplinary management strategies.

Yeah, I mean fatigue can be a real um Achilles heel in patient management. Anyone who believes that they're treating fatigue when they say I'm treating RA and I'm using aggressive therapies has missed the boat because there's four other reasons why that patient's going to have fatigue. Um uh and they can be fixed with lifestyle and better sleep and counseling and you know uh there's so much you have to deal with here. I think it's great when a company wants to say they have good pro fatigue data on their um new drug in an RA population. But I'll remind you that uh two weeks before ACR we had a report uh from a large lupus cohort um and it showed that fatigue in lupus had nothing to do with activity. It had to do with um uh damage and depression and depression trumped damage
you know big time. So it's sort of like the chronicity and severity of an overall disease and the maybe like the moral distress that it causes on the patient that starts this other cascade going. Um, as you say there are many pathways into it, but you know in lupus where you'd figure it's got to be related to you know their SLEDAI scores and their complement levels — and not at all the case. And I think that's true in all our diseases. But anyway, John, how do you advise people about how they should address fatigue in their clinic?

Get a lot of sleep. And I tell them be compliant with your anti-rheumatic drug therapy. Uh, but exercise oftentimes is helpful for fatigue. So I encourage aerobic exercise. I know you just got off of riding 10 miles on your bike. And that's terrific for fatigue. And then the usual sleep hygiene — you know, only go to sleep in a bed. Don't take naps in the afternoon. Don't drink coffee before you go to bed. Put that iPhone away before you go to sleep. And don't wake up in the middle of the night and check your email. You know, those common things. But controlling disease is the best I can do as a rheumatologist. But that's not enough.

And patients need to maintain appropriate sleep hygiene. In obese patients, obstructive sleep apnea is often ongoing, and patients who feel unrefreshed — what people call fibromyalgia — oftentimes is the manifestation of a disturbed sleep pattern. GLP-1s are tremendously effective in helping patients who are overweight lose weight and improve their sleep apnea. It doesn't necessarily lead to resolution, but I have probably done more referrals to sleep specialists than any other specialist with my patients. And patients who come in with chronic fatigue, chronic pain — the combination of the two — oftentimes are diagnosed with obstructive sleep apnea, or if they're lighter, restless leg syndrome. My sleep specialist to whom I refer said that I have about a 99% accuracy rate in picking up obstructive sleep apnea based on those symptoms.

Yeah. All right. This is really helpful. I want to thank our panel for a great discussion on a lot of really important presentations from EULAR London 2026. Tune in for more of these panel discussions on RheumNow. Thanks everyone.