Key data on pain and prognosis: Clinical insights and implications

touchIMMUNOLOGY coverage of data presented at EULAR 2024:

Summarizing his top five EULAR 24 abstracts on pain in rheumatoid arthritis (RA), Prof. Peter Taylor (University of Oxford, Oxford, UK) shares insights on their clinical implications. 

Beginning with the Canadian Early Arthritis Cohort (CATCH) we explore non-articular pain in patients with early RA and discuss how the findings could support a link between disease activity and non-articular pain, as well as confirm the need to intervene early. Staying with early RA, we then discuss the two-year PRAIRI trial data and what the findings demonstrated in terms of the impact of rituximab on the quality of life of patients with pre-RA. 

Next, we compare treatments, starting with the ANSWER longitudinal cohort study (NCT02069964), which examined persistent non-inflammatory pain relief between biologic and targeted synthetic therapies. Following this, we review and examine the RA-BE-REAL study results which investigated the impact of advanced therapies with differing modes or actions on patient reported pain. We then conclude with the post-hoc analysis of the phase 3 SELECT-COMPARE (NCT02629159) study, exploring the potential reasons for the difference in pain relief previously observed between JAK inhibitors and anti-TNF treatments.

Associated abstracts:

Meng C, et al. Characterizing Prevalent Non-Articular Pain at Early RA Diagnosis and Evolution Over the First Year of RA Treatment: Results from the Canadian Early Arthritis Cohort Study. EULAR 2024. Abstract OP0064

Frazzei G, et al. The effects of rituximab on quality of life, pain and mobility scores in the preclinical phase of rheumatoid arthritis: 2 year data from the PRAIRI study. EULAR 2024. Abstract OP0053

Yamashita M, et al. Comparison of b/tsDMARDs about non-inflammatory pain in patients with rheumatoid arthritis – ANSWER longitudinal cohort study. EULAR 2024. Abstract OP0041

Taylor P, et al. Association between patient-reported pain and achieving remission or low disease activity at 3 months in patients with rheumatoid arthritis from the RA-BE-REAL study. EULAR 2024. Abstract OP0086

Taylor P, et al. Direct and Indirect Effects of Upadacitinib or Adalimumab on Pain in Rheumatoid Arthritis: Results from a Randomized Phase 3 Study. EULAR 2024. Abstract OP0072

Disclosures: Prof. Peter Taylor discloses acting as a consultant for AbbVie, Acelryn, AnaptysBio, Biogen, Fresenius, Galapagos, Gilead, Immunovant, Janssen, Lilly, Moonlake, Nordic Pharma, Pfizer, Sanofi, and UCB; and receiving grant/research support from Galapagos (AlfaSigma).

This content has been developed independently by Touch Medical Media for touchIMMUNOLOGY and is not affiliated with EULAR. Unapproved products or unapproved uses of approved products may be discussed by the faculty; these situations may reflect the approval status in one or more jurisdictions. No endorsement of unapproved products or unapproved uses is either made or implied by mention of these products or uses by Touch Medical Media or any sponsor. Views expressed are the speaker’s own and do not necessarily reflect the views of Touch Medical Media.

Transcript:

Hello. My name’s Peter Taylor, Professor of Musculoskeletal Sciences at the University of Oxford. And I’m going to talk to our viewers today a little bit about some of the pain in rheumatoid arthritis abstracts that were presented at the EULAR meeting in Vienna, in June 2024.

So first of all, we’ll start off by talking about early phase rheumatoid arthritis.

And for those of us who see patients, we’ll be very aware that pain is often the presenting feature of rheumatoid arthritis.

But even if we adequately treat the inflammatory components of disease in terms of suppressing swollen joints and diminishing acute phase response and normalizing it, we know that many patients continue to complain of pain, fatigue and other patient-reported outcomes.

So the first abstract we’re going to look at is in fact from the CATCH cohort in Canada. It’s called ‘Characterizing prevalent non-articular pain at early rheumatoid arthritis diagnosis and evolution over the first year of rheumatoid arthritis treatment: Results from the Canadian Early Arthritis Cohort Study’.

Now in this very interesting study, the investigators looked at non-articular pain. So that’s to say pain that the patients complain about but away from the joints themselves.

And to do this, they divided the body of of a human homunculus as it were into five areas. So the upper part of the torso, the lower part of the torso, and then divided them into left and right, so they were quadrants. So as it were, upper left chest, upper right chest, and then, lower right abdomen, lower left abdomen, and also the axial region, so that is to say the spine.

And what they found was that in fact, non-articular pain is really very common in early rheumatoid arthritis throughout the first year of presentation and diagnosis, over half the patients experiencing pain of this nature.

Now the common patterns of regional pain involved both the axial regions, both upper quadrants, and both lower quadrants.

And there was a significantly higher frequency of occurrence of tender and or swollen joints within most areas of non-arthritic pain over time. And this gives some preliminary data really to suggest that rheumatoid arthritis disease activity may contribute to this non-articular pain. In other words, they may be linked in some way.

More studies are obviously needed to look at this, and we also need to intervene early in the disease course to prevent and treat this non-articular pain. And there have been many other published studies in recent years, the last two or three years, that strongly support that that’s the case. The earlier we intervene in rheumatoid arthritis, the better the pain outcomes tend to be. Although, as we’ll see as we progress with other abstracts, problems can remain.

So the next abstract that I’d like to talk about also involves very-early phase rheumatoid arthritis if you like, or even a step before that, pre-rheumatoid arthritis.

And this abstract was entitled ‘The effects of rituximab on quality of life, pain and mobility scores in the pre-clinical phase of rheumatoid arthritis’, and it was the two year data from the PRAIRI study. Now just a quick reminder for any viewers who are not familiar with this study, this was a study in people who were seropositive with arthralgia but not yet meeting classification criteria for rheumatoid arthritis, and they received a single cycle of rituximab.

And the consequence of this single cycle of rituximab is that it delayed the conversion of classification criteria rheumatoid arthritis as compared to patients receiving placebo.

However, it didn’t seem to have any positive impact on quality of life as measured by various patient-reported outcomes.

So in these at-risk individuals at-risk of developing rheumatoid arthritis, rituximab didn’t significantly alter either the patient-reported outcomes or the perceived disease severity at the time of arthritis development.

And that included pain very interestingly. So this is in contrast to various other studies we’ve seen, whereas if you treat-to-target early, it does seem to, lessen the degree of pain experience later on.

And although, rituximab, didn’t improve any of the patient-reported outcomes, it didn’t have a negative effect on quality of life, which is good news. And it does underscore that rituximab is well tolerated in early-phase rheumatoid arthritis or even in pre-clinical disease. So again, we need further studies to understand what the best intervention is to prevent both articular pain and non-articular pain, and also to understand whether there is any difference between the targeted therapies that we have. And we’ll come on to consider that issue next.

So the next study that I’d like to talk about addresses this very point, and it’s a nice study from a registry, investigation in Japan, which is called the ANSWER longitudinal cohort study. And the title of this abstract was ‘Comparison of biologic or targeted, synthetic DMARDs about non-inflammatory pain in patients with rheumatoid arthritis: The ANSWER longitudinal cohort study’.

Now in this particular study, our colleagues in Japan, took patients who had had an inadequate response to their first conventional synthetic DMARDs, and they were going to start an advanced therapy, either, a biologic therapy or a targeted synthetic therapy.

And in order to ask the question, do targeted synthetic therapies actually give a better relief of pain than biologic drugs, they defined something as non-inflammatory pain as follows. If the patient had an unacceptable degree of pain, which was quantified as a score of 40 or more out of a 100 millimeter visual analogue scale, and the patients also had a raised CRP, they were considered to be having inflammatory pain. But if they had a CRP that was less than 10 and only one or fewer swollen joints, then they were considered to have potentially non-inflammatory pain. Now, of course, this is an arbitrary definition, but that’s how they set up the study.

They had a large cohort of patients, and they followed them on their advanced therapy for 12 months.

And what they found in this particular study is that the persistence of non-inflammatory pain, in fact, was not any different with biologic therapies vs. targeted synthetic therapies. And in fact, this is in contrast to some other studies that we will come to talk about.

But maybe this difference from some other studies is also due to the way in which the definition of non-inflammatory pain was made and that arbitrary cutoff of 40 millimeters out of a 100 visual analogue scale for the description of of significant pain or unacceptable pain.

They also noticed interestingly that persistent non-inflammatory pain was slightly more marked in patients receiving abatacept than other drugs. I think we have to be careful about drawing any clear cut conclusions about it. But it does raise the issue that drugs that are similarly effective with respect to improving composite scores of disease activity may not necessarily have the same benefits with respect to experienced pain.

So we now come on to to the next study that addressed a very similar issue. And this was, in fact, a study that I presented, and it was based on another prospective observational study, the so called RA-BE-REAL cohort.

And the title of this abstract was ‘Association between patient-reported pain and achieving remission and or low disease activity at 3 months in patients with rheumatoid arthritis from the RA-BE-REAL study.

Now in this study, patients who were commencing any advanced therapy for the first time, but by that it was distinct from the ANSWER study because they could have had other advanced therapies before. But let’s say the patient was starting baricitinib, they could have been on one or two or three previous advanced therapies, But as long as this was their first exposure to baricitinib, they were enrolled.

And the patients were divided into three groups, those starting baricitinib, those starting an anti-TNF, which of course would be a biologic anti-TNF, or those starting any other mechanism of action advanced therapy, which could include another JAK inhibitor.

Now what we did in this study was to look at those patients in the total cohort who achieved low disease activity or remission at three months. And in fact, more than half of the patients receiving any one of the advanced therapies, baricitinib, anti-TNFs, or other mechanisms of action drugs, actually achieved a pain improvement in a pain threshold of at least 50 percent from baseline. So that’s pretty impressive.

Now these data are completely in keeping with what we just heard from the ANSWER study. In other words, all three different categories of drugs are giving a similar benefit in terms of pain relief based on that magnitude of improvement, a 50 percent improvement from baseline. But when we cut the data in different ways, and this was a post-hoc analysis of the RA-BE-REAL study, we begin to see some different results emerging.

Now another way of looking at this data is to look at the existence of what we call residual pain. And for the purpose of this study, residual pain was defined as those patients having 20 millimeters or more out of a 100 millimeter visual analogue scale of pain, and that was considered to be residual pain.

And in fact, of the patients who achieved remission or low disease activity at three months, less than half the patients, receiving any one of the categories of drugs experienced, residual pain.

But among those who did not achieve remission, or low disease activity at three months, the large majority of patients still continue to have residual pain, over 90 percent of the patients. That was interesting and, again, points to a similarity in outcomes between the treatments and also probably points to the fact that a lot of the pain that the patients experience is in fact inflammatory pain in origin.

But we mustn’t forget that there are other origins of pain as well, including mechanical pain, neuropathic pain, and osteoblastic pain.

Now in fact, a large majority or a large reduction in pain when we look at the absolute pain reduction from baseline. So again, this is looking at the data in a different way. So we’re not now looking at a percentage change. We’re not now looking at residual pain, but we’re looking at the absolute reduction in pain. And in fact, there was a large reduction in pain for all the different mechanism of action drugs, but it was in fact largest for baricitinib compared with anti-TNFs and other mechanism of action drugs. And that’s despite the fact that the baricitinib patients were on average, having longer disease activity and have been exposed to a greater number of prior, mechanism of action drugs.

So what these analysis seem to suggest is that there is an association between the achievement of pain improvement and the achievement of remission or low disease activity. I don’t think there are any surprises there. And among those patients who achieve remission or low disease activity at three months, the patients treated with baricitinib, besides having longer disease duration and a higher number of prior biologic or targeted synthetic DMARDs, they had a numerically higher mean reduction from baseline in pain visual analogue score compared to the patients receiving anti-TNF.

And in fact, almost half the patients who achieved remission or low-disease activity and were treated with baricitinib achieved a threshold of pain improvement of at least 70 percent improvement, which is is a very impressive amount.

And I think what these studies, when we look at these abstracts collectively, are telling us that there are not necessarily apparent differences between different mechanisms of action when you look at the data, in certain ways. But when you look at the data cuts in other ways, looking at those patients who get the highest magnitude relief, then some differences begin to emerge.

So to come to the final abstract which pursues this line of thought a little bit further was another one that I presented, and it was entitled ‘Direct and indirect effects of upadacitinib or adalimumab on pain in rheumatoid arthritis: Results from a randomized phase 3 study’, and that study was the SELECT-COMPARE study. So this was a post-hoc analysis.

And one of the striking findings in the SELECT-COMPARE study, as we’ve previously seen and reported in the RA-BEAM study and in the FINCH trials with filgotinib is that the magnitude and speed of improvement in pain appears to be greater with the JAK inhibitors than is the case with a biologic anti-TNF comparator.

In the case of the SELECT-COMPARE study as in the RA-BEAM study, the difference between the JAK inhibitor and the anti-TNF became statistically significant as an early time point, and that differential was maintained with statistical significance throughout the first 52 weeks of drug exposure.

So in this post-hoc analysis, we decided, well, could we ask the question to what extent is this difference, apparent difference in absolute pain reduction due to differences in the anti inflammatory potency of anti TNF, which targets a single cytokine versus the JAK inhibitor, which of course is a multi-cytokine inhibitor.

And in order to do that, we undertook a mathematical analysis called a mediation analysis.

And we took the available surrogates of inflammation, which comprise swollen joint counts, ESR and CRP, and we compared the active drug, which would was either upadacitinib or adalimumab to placebo.

And we look to see to what extent could the reduction in pain be accounted for by the reduction in those surrogates of inflammation.

And that reduction we call indirect improvement in pain.

And the quantity that could not be accounted for on that basis, we called the direct improvement.

And here’s the findings. So in this particular study, we looked at two weeks, 12weeks, and 26 weeks of drug exposure.

And by the time we get out 12 weeks of drug exposure, we find that in fact the magnitude of pain improvement that can be accounted for by inflammation relief is very similar for adalimumab vs. placebo, and upadacitinib vs. placebo, suggesting that in fact, the contribution of inflammatory pain reduction is similar for both agents.

But on the other hand, from 12 weeks onwards, the difference in the direct component of pain relief, in other words, the component that could not be accounted for on the basis of the surrogate swollen joint counts, ESR and CRP, was very much greater with upadacitinib than it was with adalimumab, almost twice as much.

And these findings are of interest and generate some hypotheses.

It’s important to stress that these are post-hoc analyses, but the hypotheses they generate are as follows. It’s possible that JAK inhibition has some effect on non-inflammatory pain. That’s one hypothesis.

It’s also possible that the differential between the two mechanisms of action could be accounted for in part by some other aspects of inflammation, which we weren’t able to measure and we didn’t have quantified in the clinical trial, and therefore, we couldn’t include in the mediation modeling analysis.

And those two hypotheses are not mutually exclusive. Both could be correct. But once again, it needs further investigation to tease all this out. So what this particular study suggests, and other studies that we’ve done looking at filgotinib and baricitinib, is that in fact there is a difference in pain relief in some patient populations and in those with a high baseline pain when you compare the JAK inhibitor to, a biologic anti-TNF. But again, those differences may not be so apparent in other patient populations with lower baseline pain and so on. And so these are giving pointers as to how we might translate these findings into our clinical practice.

Interviewer/Editor: Gina Furnival

Cite: Taylor P. Key pain in RA abstracts: Clinical insights and implications. touchIMMUNOLOGY, July 25 2024.