Skip to main content

EULAR 2026 Daily Podcast Day 4b

Jun 07, 2026 12:06 am
Are we failing in the treatment of RA-ILD? Using AI to Predict Progression to RA Treat to Target Works in Elderly Onset RA
Transcription
You're listening to a RoomNow podcast coming to you from London in EULAR 2026. Enjoy.

Hi. It's doctor Janet Pope. I'm reporting from London, England, EULAR 2020 '6 for RheumNow. And I hope you've been following us. There's a lot of exciting reports coming out.

I want to talk about treatment of RA with ILD. Are we failing? And this is based on guidelines that have come out over the last two years, but also on a hot topic, which is how to treat for HOT by Doctor. Phillip Dayud. And it was on 06/05/2026.

So I think the first question is, are we failing in the treatment of RA ILD? So why did I frame it that way? We've come a long way where ILD is sometimes being screened for patients. We can ask or take their lungs in the lower bases. We can do PFTs when they go on an advanced therapy, they're getting chest imaging and periodically in high risk patients, maybe we'll even be doing CT scans.

So I think there's an awareness. And what is different about rheumatoid arthritis ILD is that more people than say scleroderma associated ILD, more people with rheumatoid arthritis have a UIP or a usual interstitial pneumonia pattern. More men with RA relative to women will have ILD and they're older and often long standing disease. So older men smoking is a risk, more UIP pattern. They kind of look like IPF patients.

So for a long time, number one, we might have been failing because immune suppression was not always used with the drugs that treated progressive pulmonary fibrosis, such as Nintandanib, although MMF can be used in RA ILD, but it doesn't really treat the rheumatoid arthritis joints very well. Number two, the use of anti fibrotic is probably not optimal. Some of it might've been from a lack of treatment, tolerability, some of it might've been access, and some of it frankly might've been not case finding, lack of awareness from the rheumatologist or the pulmonologist. So the anti fibrotic classes of drugs currently are the PDE4B inhibitor, Neurandolamast, which has been approved in The US, fast tracked and will be approved in many other countries in the near future we think. And then Nintendenib.

Nintendenib is a tyrosine kinase inhibitor. There were some abstracts and some of the review that we heard from the professor who gave this lecture suggests that treating obviously RA effectively is helpful and using concomitant immune suppression. What I did learn, and I thought that this was maybe a failure of my understanding, is that RA disease activity might not change the pulmonary functions, but it will change the overall mortality. So we need to get RA disease activity under control. I always understood that area under the curve of more rheumatoid arthritis active disease increased the chance of both getting ILD and RA and worsening ILD.

And I think a bit of a shadow was put on my understanding of that. I don't think the answer is totally clear. The final thing is I think we have to look at the things that can prolong our patient's survival. Giving oxygen if they're hypoxic and testing for that. Using shared care so that the patients are vaccinated so that they get smoking cessation, so that they get in an exercise program and using a common sense algorithm.

So we know what would be the cut point within this individual where I will trigger a change in therapy for their RA because they have ILD and for the ILD itself. Who should get on a pulmonary fibrosis drug or are we actually putting them on too late? Now one area where we're not failing, and this will be my final point that was a take home message, was once you treat the RA ILD, you often can level the worsening PFTs for at least over the next one to two years. And there was an abstract that suggested that as well in rheumatoid arthritis. So I don't think we're failing, but I do think there's a care gap and we can do better.

Please again follow us at RheumNow. It's Janet Pope reporting as at Janet Berdaupe. Thank you.

Hello. I'm Jonathan Kaye reporting from EULAR twenty twenty six in London, England. This is the third day of EULAR twenty twenty six, and there were lots of interesting posters and some very interesting oral presentations. I was going through the rheumatoid arthritis poster session, and I came upon a poster that was being exhibited by Eleanor Bolton, a PhD student from University of Leeds in Yorkshire in England. She's working with Paul Emery and Culver Manckea and looking at the prearthritis group.

And she used artificial intelligence to identify two distinct preclinical phenotypes among patients with anti CCP antibodies, but no clinical synovitis. She looked at this prospective cohort of 451 individuals who were recruited from Yorkshire. They were adults over the age of 18. All of them had anticyclic citrullinated peptide antibodies and musculoskeletal symptoms, but no clinical synovitis. And she fed this cohort information about this cohort into machine learning, and she came up with two clusters.

There was a low risk cluster and a high risk cluster, and they differed in that there was a higher sedimentation rate in the high risk cluster than the low risk cluster, higher tender joint count, higher rheumatoid factor level, higher patient global health assessment by visual analog scale. And then when they looked with ultrasound for synovitis, there was more synovitis among those in the high risk group than in the low risk groups. They had more morning stiffness. Anti CCP antibodies were higher, tighter. They were older, slightly older, 53 years compared to 49 years, and they had more tenosynovitis.

The tenosynovitis difference wasn't significant, but all of the other differences were highly significant between the two groups. So she clustered these into two groups and then looked at a Kaplan Meier survival curve looking for survival without arthritis and found that in the high risk group, they developed arthritis more and more rapidly or sooner than those in the low risk group. So this was a very interesting use of machine learning, artificial intelligence, to feed in this cohort that had been collected of four fifty one patients with musculoskeletal symptoms but no clinical synovitis and presence of anti CCP antibodies, putting them at risk for possibly developing rheumatoid arthritis. And then when she looked at these two groups and plotted Kaplan Meier's survival without arthritis, it turned out that the high risk patients developed arthritis sooner, and more of them developed arthritis than did the low risk group. So she validated our clinical suspicion that patients with higher levels of acute phase reactants, patients with higher levels of rheumatoid factor or anti CCP antibodies, patients with worse global health assessment, patients with more tender joints and more tenosynovitis are going to be more likely to develop arthritis from the pre arthritis phase.

So I congratulated her on this work, which she's done as the beginning of her first year of her PhD. And I asked her what she she plans to do to take this study further, and she plans to look at this cohort longitudinally and prospectively, which will be very interesting. So her retrospective analysis found a separation between these two groups. But if she looks prospectively at patients who meet these criteria to see whether there's still a difference in the rate of progression from a preclinical phase to a clinical inflammatory arthritis phase, that will be very interesting. So we've heard more about artificial intelligence at this meeting, but I found this a very intriguing and useful and informative and practical application of artificial intelligence to also study this preclinical phenotype of rheumatoid arthritis, which is really the most exciting area to me of trying to identify who's at risk and then looking at ways to prevent progression to rheumatoid arthritis.

So look forward to seeing more work from this group and other groups about progression from preclinical arthritis to arthritis, and especially more studies of prevention. For more from EULAR twenty twenty six in London, go to roomnow.com. I look forward to seeing you again soon. I'm Jonathan Kaye.

This is Jieha Li from Michigan reporting for RheumNow in London for EULAR twenty twenty six. I'm going to be discussing what is a adjacent to my passion, which is aging and rheumatology. This is poster number 1326, looking at to show that treat to target works in elderly onset RA, and if anything, that they need less to get there. So older adults with RA are often undertreated. And in clinical practice, there's a tendency to kind of be hesitant about prescribing for concerns about tolerability risk because older adults have more comorbidities and there's some assumption regarding safety and efficacy.

Although there's been a number of studies to show that our drugs, even the biologic and targeted synthetics are pretty well tolerated in older adults. So I was really excited to see this study to really get at the question of does treat to target work? So by ways of a little bit of a background, elderly onset RA is defined as RA diagnosed on or after the age of 65. And I want to put in a little bit of a plug here because the American College of Geriatrics tries to avoid words that catastrophize aging. So their preferred term is late onset rather than elderly onset.

So it's really interesting to see there's some difference regarding terminology. I think this is one way in which having a unified term approach is how we can also improve care of our aging population. Because the population is aging. In Europe, they're saying by 2050, at least one third are going to be over the 65. I believe in The US that number is by 2030 we're going to be one in four are going to be older adults.

So this is going to be the norm of our study. I think Rise actually shows that forty percent of all clinical patients are over the 65. So this is our reality and our challenge. So it's really important to understand and get data on these because older adults often are not included in trials that actually inform my guidelines. So with that, there's also some clinical differences between those with late onset forgive me if I say late instead of elderly late onset versus younger onset, in that these patients tend to have more abrupt systemic presentation, tend to be more seronegative.

There tend to be less of a gender difference. So instead of the typical three to one or four to one, it tends to be two to one. So it's important not to under diagnose our older male patients. And also there has been some studies to show that have higher levels of IL-six and lower levels of TNF-one. So clinically, physiologically, they tend to be a little bit different.

And also at the time of diagnosis, they're more like to have comorbidities and polypharmacy adding to the challenge of treating these patients. But the question, like I said, for these authors, and they're from The Netherlands, was can older adults receive achieved disease remission or low disease activity with the treat to target approach? To really challenge the assumptions, as I said, that some physicians have because there was a study on ageism, where interestingly most rheumatologists said they were not ageist, but those who have tendency were less likely to prescribe our DMARs and more lean towards symptomatic management with glucocorticoids and NSAIDs. So with that, let me set up the study for you. So this data is from what they call a T REACH trial.

It's a multicenter stratified single blinded randomized control trial that was conducted in The Netherlands. And it was enrolling patients who met classification criteria for RA. And they were either categorized as elderly onset or young onset, depending on the age at which they were diagnosed with RA. And out of I believe they had a total of just over three hundred. So ninety eight patients had a mean age of 73 at the time of diagnosis meeting the elderly age onset criteria and the rest were younger onset and their age ranged between 18 to 65.

And that means roughly about twenty five percent were late onset in The US in the Medicare population. That number is about thirty percent. Again, I think that's reflective of The US being slightly older in general than the European populations. For all of these three twenty seven patients where the overall mean age was 48, they were followed doing a fixed protocol of a treat to target approach of treatment. Their DAS trajectories, their DMARD use, glucocorticoid use, biologic drug survival, and DMARD free remission along with adverse events were assessed over a two year period of time, which I think is very nicely set up and impressive.

And then they looked at stratified analysis based on autoantibody status at the end. So taking each of the outcomes, so for disease activity trajectories, they were actually similar between the two groups. The mean DAS difference between elderly onset and young onset was only 0.03, which is very clinically negligible. And both groups actually achieved comparable rates of low disease activity over the follow-up period of two years. How did they do that?

Did the treatment intensity differ? I think this is where the importance comes and the authors note that appropriately as well. So for elderly onset RE patients, they need a significantly less intensive therapy to achieve the same low disease activity with an odds ratio, what they put as like zero point three. So like sixty percent less intensity compared to the young patients. And they were less likely to require biologics.

And I think the value here is that this was perspective data collection whereas in US when we look at Medicare it's often a cross sectional retrospective. So we find that they get less biologics and we don't know if that's because they didn't require it or that it just wasn't given to them. So I think it's reassuring and informative to see that in this type of study they really did not. At the same time, just as a benchmark number, The US about thirteen percent to fifteen percent of older adults are on biologics. And in this study for them, I have the number.

Give me one moment. I think that number actually was higher in terms of the biologics. So yes, their numbers were actually up into prescribing rates of like twenty five to thirty percent. So they required less however yet their overall prescribing rates were much higher than in The US. Sorry, continuing on with that.

So again, in terms of the treatment intensity, they were left, but in terms of drug survival, in terms of discontinuation, they were very similar in that older adults or elderly onset RA patients were well tolerant of biologic therapy as well. Terms of adverse events. Again, this is important. The safety is something that we very much care about. So elderly onset patients had higher rates of bone marrow depression or osteoporosis, understandably fifteen percent versus seventy percent.

Elevated creatinine, it could be from methotrexate intolerance or underlying CKD, that number being eight percent to two percent compared to the younger onset. Younger onset patients did report more mood related disorders than anything. In terms of the continuity of treatment, though, the DMARC switches or the dose adjustments due to the adverse events, it did not differ overall between those two groups, indicating again that the overall drug tolerated barely is comparable. Although in terms of absolute experiences, older adults may report more. So what does this mean clinically for us?

I think this finding makes a clear case that treat to target is not only feasible in elderly onset RA, that it may work better than we expect with less treatment exposure, especially in this group they did some stratified analysis that's seronegative patients. But again, just keep in mind, I think that seronegativity is just coming from the nature of what elderly onset RA patients usually present with. And it's reassuring to know that the adverse effects are really not different between these two. So the practical message I hope to share with you and for me as well is that we are progressively caring for an older generation of patients. So let's not withhold treat to target strategy from these patients based on age alone.

There's a saying in the geriatric role that chronologic age does not equate to physiologic age. So there are other means and assessments that we can bring in. There was actually a number of abstracts at EULAR about frailty sarcopenia being a risk stratification model or even an outcome model for older adults. Oh, and yes, there was actually I was very pleased to see this. There was actually a whole entire session about aging.

And I think what was most powerful is the patient voice. So there was a patient who had been diagnosed, who was 17, who came and shared what the patients want. And what they want is to be able to independently age at home, to be able to have conversations about treatments and understand that within context of their competing comorbidities and devise a plan that fits in with their goals and priorities. So I think this information and this study gives us some reassurance that our usual treat to target approach helps so that we can have better shared making. In terms of just I do want to point out just a few limitations in that it is relatively small.

Again, like for prospective cohort, I think three thirty is good. But again, the numbers are somewhat small for certain subgroup comparisons, especially with the smaller disease or drug specific measurements as well. And this is a trial setting. So like they had intense monitoring and hands on. So in real world where adherence is different for our patients, there may be some variance in terms of the findings as well.

With that, this was Jeehaw Lee reporting for RheumNow in London for EULAR twenty twenty six. Thank you for listening.

ADD THE FIRST COMMENT

If you are a health practitioner, you may to comment.

Due to the nature of these comment forums, only health practitioners are allowed to comment at this time.

×