The Top Research in PsA Presented at ACR Convergence 2023
Dr. David Pisetsky, the founding editor of The Rheumatologist, weighs in with his picks for the top research in psoriatic arthritis presented at ACR Convergence 2023.
SAN DIEGO—Psoriatic arthritis (PsA) is a complex clinical condition that can present with variable joint findings and variable skin findings (e.g., psoriasis). PsA was once classified as a rheumatoid variant but, nevertheless, differs from rheumatoid arthritis (RA) in genetics, involvement of the entheses and the presence of axial disease. Differences also include the response to biological agents, as indicated by regulatory approvals. Whereas tumor necrosis factor (TNF) inhibitors and abatacept have been approved for both RA and PsA, anti-interleukin (IL) 17, anti-IL-12/23 and anti-IL-23 antibodies have been approved for PsA and psoriasis.
These differences highlight the need for the development of imaging modalities, laboratory tests and other biomarkers that are explored and validated specifically for PsA and, therefore, can advance the goal of personalized or precision medicine. This is a large challenge because PsA itself is so heterogeneous.
The abstracts for ACR Convergence 2023 illustrate the challenge well. Although advanced imaging (e.g., ultrasound, magnetic resonance imaging [MRI]) can reveal joint involvement in great detail, the interpretation of findings is complicated because of the uncertain relationship of clinical findings (e.g., pain) to the structures affected (e.g., synovium, entheses, bone). Big data approaches (e.g., transcriptomics, mass spectometry) carry enormous promise, but the mass of data that these analytic tools can generate will require new informatics platforms, including machine learning and artificial intelligence (AI) to make sense of the data and provide actionable information.
1. The Burden of Oligoarticular PsA
Abstract 0966: Olopoenia et al.1
The clinical findings of PsA are highly heterogenous, contributing to challenges in diagnosis and treatment. The extent of skin disease varies widely, and patients display distinct patterns of involvement of joints and entheses to allow designation of disease subtypes. These subtypes include polyarthritis (poly), oligoarthritis (oligo), classic arthritis (with distal joint and nail involvement), arthritis mutilans and spondyloarthritis.
The polyarticular form of PsA tends to receive the most attention because, in clinical trials, the number of inflamed joints determines patient eligibility; enrollment, therefore, favors patients with polyarticular disease. Further, metrics similar to those used to evaluate RA, the prototypic form of polyarticular inflammatory arthritis, can be readily adapted to the study of polyarticular PsA.
To better understand the oligo PsA subtype, Olopoenia et al. used cross-sectional data from the 2019 National Health and Wellness Survey to determine features of patients with oliogoarthritis—defined as four or fewer joints involved—compared with those of patients with polyarticular disease—more than four joints—based on self reports; the study also involved comparisons with the general population (i.e., no psoriasis or PsA).
As the data indicate, patients with olioarticular arthritis differ from the general population in age, sex, smoking and use of alcohol among other features. Interestingly 44.4% of patients with oligo PsA were diagnosed by the primary care physician. A larger proportion of those with oligoarthritis were not treated than those with poly PsA (23.3% vs. 16.7%) and a smaller proportion was diagnosed by rheumatologists (23% vs. 40.7%). Importantly, despite the limited number of joints involved, the oligo PsA patients had lower health-related quality of life (QoL), as well as lower mental and physical functioning, work productivity and activity engagement compared with the general population.
These findings are interesting because they suggest the impact of arthritis reflects not just the number of joints involved, but the intensity and severity of joint involvement, as well as its impact on patient functioning.
Given the burden of comorbidities and the impact on QoL of the oligo population, it is surprising that so many patients with this form of PsA in this study were either not seen by a rheumatologist or received treatment. The gap of care of this patient population points directly to the need for more education on the heterogeneity of PsA for primary care providers as well as better referral for specialty care.
2. Ultrasound for PsA
Abstract 0746: Gutierrez Manjarrez et al.2
Ultrasonography is an increasingly popular imaging modality to evaluate arthritis, assess disease activity and develop biomarkers that may predict treatment response. For PsA, US is complicated because of the various findings associated with joint involvement. These findings include the following features: synovitis, peritenonitis, tenosynovitis, new bone formation (NBF), bone erosion and enthesitis from inflammatory or structural lesions.
Gutierrez Manjarrez and colleagues used ultrasound to evaluate 107 PsA patients to determine correlations with drug persistence, Disease Activity in PsA (DAPSA) change and DAPSA low disease activity (LDA) at three to six months after treatment. Among the relationships investigated, ultrasound synovitis and peritonitis showed the highest correlation with physician global assessment and swollen joint counts. The correlations with other ultrasound features were low.
In terms of treatment response, the investigators did not find any association of ultrasound features and DAPSA-LDA, although a greater reduction in DAPSA showed an association with high baseline synovitis, peritenonitis and enthesitis structural scores.
As the authors conclude, imaging and clinical features of PsA can be discordant, although synovitis, peritendinitis and tenosynovitis may correlate more strongly with disease activity and treatment response than other imaging features.
These findings suggest only limited utility of ultrasound to evaluate the course of PsA, but they are nevertheless important in highlighting the complexity of joint lesions in PsA and the need to better understand the relationship of different clinical manifestations to underlying disease mechanisms, on one hand, and treatment response on the other.
3. What Drives Pain?
Abstract 1388: Gazel et al.3
The evaluation of disease activity in patients with PsA is challenging because a major site of involvement, the entheses, is difficult to assess on clinical exam; current approaches lack sensitivity and specificity. As a new approach to assess better enthesitis, Gazel and colleagues combined ultrasound with functional MRI (fMRI) of the brain in a proof-of-concept study to determine whether advanced imaging can illuminate the underlying pathogenesis and drivers of pain in PsA.
The study involved 12 patients who were divided into two groups: five patients in Group 1 had Achilles enthesitis on exam and a positive ultrasound (hypechogenicity and Doppler signals), and seven patients in Group 2 had negative ultrasound, with or without Achilles tenderness. Patients had fMRI imaging with rest and after induction of pain by increasing pressure with a blood pressure cuff.
The results of the study indicate that those patients with positive findings on ultrasound imaging had more neural activity than ultrasound-negative patients when processing pain. Further, the ultrasound-positive patients showed more neural activity in brain regions related to movement, body representation and pain. Together, these findings suggest that those patients with enthesitis on ultrasound process pain differently than patients without enthesis on ultrasound imaging, even with pain.
This study is important because it suggests new ways to better assess drivers of pain in PsA patients, delineate pathogenesis and categorize treatment responses. Given the large number of PsA patients who currently have persistent pain despite current agents, new biomarkers are needed to determine differences in the effects of therapy on the arthritis vs. enthesitis. In addition, new biomarkers would be valuable in determining whether persistent pain represents nociplastic pain or central sensitization as opposed to inadequately treated enthesitis.
4. Spinal Inflammation
Abstract 0503: Raychaudhuri et al.4
Axial joint involvement is a prominent feature of PsA and can coexist with the various forms of peripheral joint involvement. The frequency of axial disease in PsA has been unclear because of the uncertain relationship between imaging findings and symptomatology. As another approach to elucidate spinal inflammation in PsA, Raychaudhuri and colleagues used total body (TB) PET/CT with an 18F-FDG radiotracer. This approach assesses glucose metabolism and can therefore measure inflammation.
This study involved 25 PsA patients, of whom eight had symptoms of inflammatory low back pain, and determined inflammation in both joints and entheses. As the data indicate, PET/CT imaging detected axial abnormalities in 21 of the 25 patients (84 %) involving various joints alone or in combination. Twenty percent of patients had evidence of sacroiliac joint involvement, either unilateral or bilateral but asymmetric. Involvement of the entheses was also high, at 80%.
These findings are, perhaps, surprising because relatively few of the patients had inflammatory back symptoms; one of the patients with inflammatory low back pain, however, had a negative MRI but a PET scan positive for sacroiliitis. For the spine, imaging is inherently difficult because of the complex anatomy of this structure, as well as the frequency of findings by conventional imaging in the general population. Nevertheless, the burden of spine disease by PET scans in this study is impressive, although the role of this type of imaging in routine patient care is not yet clear.
Future studies will be important in assessing the relationship between axial and peripheral disease (both arthritis and enthesitis) as well as determining any differences in treatment response of the arthritis and the enthesitis.
5. ‘Fibromyalgia-ness’
Abstract 0483: Macfarlane et al.5
Like other forms of inflammatory arthritis, PsA is a painful condition and, like other forms of inflammatory arthritis, PsA is associated with “fibromyalgia-ness.” Indeed, studies suggest one in five patients with PsA have what can be termed “co-morbid fibromyalgia.”
Because the frequency of fibromyalgia in PsA is so much higher than that of the general population, this pattern of symptomatology seems inherent or intrinsic to disease rather than a co-morbidity. The frequency of 20% is similar to that in other inflammatory diseases and suggests that inflammation can lead to central sensitization, the hallmark of fibromyalgia. Fibromyalgia has also been called nociplastic pain.
To assess the frequency of fibromyalgia in PsA, Macfarlane and colleagues analyzed data from the British Society for Rheumatology Psoriatic Arthritis Register (BSR-PsA). The studies included 148 patients with PsA who met criteria for the fibromyalgia polysymptomatic distress score or the fibromyalgia 2016 criteria and explored the relationship between fibromyalgia and a series of demographic variables, as well as clinical and patient-reported outcomes.
The findings indicated that patients with fibromyalgia (29% by the criteria used) were less likely to respond to either a biologic DMARD or targeted synthetic DMARDs, irrespective of other predictors of outcome.
These findings are important for practice and suggest the value of assessing fibromyalgia in PsA patients to interpret better treatment response; the presence of fibromyalgia could also prompt initiation of adjunctive therapies directed specifically directed at fibromyalgia.
In the trial setting, the findings suggest the efficacy of new agents may be difficult to determine; it is not unlikely that patients enrolled in trials would have a greater burden of fibromyalgia that has contributed to their apparent lack of response to prior agents. Such patients could be categorized as having difficult to treat PsA although the difficulty may reflect concurrent fibromyalgia rather than ongoing inflammation.
6. Opportunity Window for Treatment
Abstract 1641: Snoeck Henkemans et al.6
The window of opportunity is an interesting concept that has been explored most often in the context of RA; this concept suggests that, early in the course of RA, there may be a time in which therapy is uniquely effective in modifying subsequent disease course. It is difficult to determine whether such a window of opportunity actually exists because, in the real world, treatment can be significantly delayed. These delays may occur related to patients seeking medical attention, providers recognizing the signs of early synovitis, and providers referring patients to rheumatologists. Given the gradual onset of symptoms in many patients, the window may close (to the extent that a window exists) before treatment starts.
To apply the paradigm of the window of opportunity to PsA, Snoeck Henkemans and a research team from the Netherlands investigated newly diagnosed PsA patients from the Dutch Southwest Early PsA cohort (DEPAR), dividing the population into three groups on basis of the total delay from onset of symptoms until diagnosis and treatment (<12 weeks, 12–52 weeks and >52 weeks); follow-up was three years.
As the data indicate, delays from both patients and providers were long and totaled 42 weeks; delays occurred more commonly in women than men and were associated with lower joint counts, as well as lower values of C-reactive protein and erythrocyte sedimentation rate. Patients with delay also more often had enthesitis.
Among treated patients, those with >52 weeks’ delay were less likely to achieve minimal disease activity or a remission than the other groups. Delay was also associated with higher values for HAQ-DI over the three-year follow-up. This study supports findings on the effects of early treatment on RA and highlights the need for reducing treatment delays.
The situation with PsA differs from that of RA because the presence of skin disease should alert providers (including dermatologists) to the possibility of PsA when patients report arthralgias and other symptoms. Correspondingly, providers can educate patients with psoriasis about the possibility of developing arthritis so that they can seek medical attention sooner. Given the age of patients with PsA and likelihood of coexistent osteoarthritis in this patient demographic, developing and implementing strategies for early diagnosis of PsA may be challenging, but can be part of an overall program of rheumatologists to educate providers on the full range of rheumatic diseases.
7. PsA vs. OA
Abstract 1792: Ganatra et al.7
I have often said that differentiating PSA from osteoarthritis (OA) in a patient with psoriasis and joint pain is one the biggest diagnostic challenges in rheumatology because of the paucity of biomarkers. Given the overlaps in the demographics of patients with PsA and OA, making this distinction is not uncommon in the clinic.
To fill in the gap in biomarkers, Ganatra and colleagues analyzed microRNAs (mRNAs) in the synovial fluid of knees from 12 patients with PsA and 12 with OA; additional samples from 35 PsA and 37 OA patients were used for validation. miRNA are small non-coding RNA molecules that serve as regulatory elements to modulate gene function in immune and other cells; thus, miRNA expression can be used as a marker for inflammation.
As shown by miRNA sequencing in this study, 51 miRNAs showed differences in expression comparing PsA and OA samples. While PsA and OA samples both showed increases in mitogen activated protein kinase (MAPK) and WNT pathways, two miRNAs (miR027b-30 and miR-223-3p) showed significant dysregulation in PsA compared to OA. The results were confirmed in a validation cohort.
Although these are interesting findings, early stage biomarker studies are often exploratory in nature, more likely generating data for understanding mechanisms than developing actionable laboratory tests. For a laboratory test, the information gained by miRNA testing needs to be beyond that of joint fluid cell counts, for example. For miRNA, data are needed on the patients with psoriasis and OA but not PsA to determine whether any of the changes observed with miRNA in the joint fluid reflect generalized immune system findings that could occur with just psoriasis alone.
8. Upadacitinib vs. Adalimumab
Abstract 2251: Taylor et al.8
Pain is dominant symptom of PsA and can be heterogeneous in origin, reflecting inflammation, damage and central sensitization (nociplastic pain). Although many different therapies have now been approved for the treatment of PsA, both biologic and non-synthetic DMARDs, the mode of the action of these agents is not fully understood. Important questions concern whether modulation of pain is a direct analgesic action or is an indirect action resulting from a reduction in inflammation.
This study presents results of a clinical trial involving 1,281 patients and comparing the effects of upadacitinib, a selective and reversible Janus kinase inhibitor, with adalimumab, a tumor necrosis factor (TNF) inhibitor, and placebo. Direct effects on pain were assessed from patient global assessment of pain and tender joint count; in contrast, indirect effects were assessed from indicators of inflammation, such as itch, total enthesitis, Leeds enthesitis index and CRP.
The results indicate that direct and indirect effects of upadacitinib and adalimumab were both greater than that those of placebo and that upadacitinib causes greater pain reduction than does adalimumab in terms of total effects well as direct and indirect effects.
These findings are important because they indicate that the inclusion of a limited number of clinical outcome measures for both direct and indirect effects can illuminate the treatment response and help dissect the mode of action of DMARDs (disease modifying anti-rheumatic drugs). It will be of great interest to apply this approach to other agents used to treat PsA and determine whether direct and indirect effects always track together or whether some agents act predominantly or exclusively by either direct or indirect activity.
Importantly, this approach extends the usual focus on tender joint count to encompass signs and symptoms related to the skin (i.e., itch) and the entheses. In this regard, elucidating the effects of anti-inflammatory and immunosuppressive agents on nociplastic pain will be an intriguing direction of research to, hopefully, identify treatment approaches for this kind of pain.
9. MACE & VTE Risks
Abstract 0739: Merjanah et al.9
PsA, like other forms of inflammatory arthritis, is associated with an increased cardiac risk compared with that of the general population; this increase in risk could reflect the effects of inflammation, as well as the demographics of the population affected (e.g., features of the metabolic syndrome). Although atherosclerosis has an important contribution from inflammation, the effects of anti-inflammatory agents and DMARDs are complicated; agents may show similar benefits on inflammatory arthritis, but nevertheless differ in the effects on major adverse cardiovascular events (MACE) and venous thromboembolism (VTE).
In two nested case-control studies from a large MarketScan database, Merjanah and colleagues conducted analyses to determine any differences in the frequency of MACE and VTE in patients with either axial spondyloarthritis (axSpA) or PsA receiving a Janus kinase inhibitor or a TNF inhibitor or neither.
As the data indicate, the frequency of comorbidities in the population was high and included prior cardiovascular disease, obesity, diabetes and hypertension. Non-users of TNF inhibitors and Janus kinase inhibitors had greater odds of MACE relative to users of TNF inhibitors; further, in a comparison of Janus kinase inhibitors and TNF inhibitors, the odds of MACE were not increased. The odds for VTE among Janus kinase inhibitor users were increased, but this difference was not statistically significant.
These data are important in decision making on the treatment of PsA, recognizing the many comorbidities that put this population at risk. Further, as the landscape of agents to treat PsA changes, the effects of biologics directed at other cytokines (e.g., IL-17, IL-12/23, IL-23) need investigation. The same will be true of Janus kinase inhibitors with different patterns of Janus kinase enzyme inhibition.
10. Predicting Response
Abstract 2235: Zi et al.10
An important goal in the treatment of PsA, as is the case of other forms of inflammatory arthritis, is personalized medicine, also called precision medicine. In this treatment approach, biomarkers can hopefully allow the prediction of treatment response so that therapy can be prescribed more effectively and efficiently. These biomarkers can range in type although analysis of gene expression is currently popular and is an example of so-called big data approaches.
Better prediction of treatment response in PsA is important because between 30 and 40% of patients are primary non-responders to current agents; further, subsequent responses can diminish as successive agents are prescribed. Achieving treatment success with the first agent prescribed would be a significant advance.
In this study, Zi and colleagues obtained transcriptomic data at baseline to identify predictive markers for response to either TNF inhibitors or IL-17 inhibitors (IL-17A); for these studies, the investigators focused on CD8 positive cells in peripheral blood, analyzing these cells from 49 PsA patients starting either class of agent. Using this approach, the investigators identified differentially expressed genes for TNF and IL-17A inhibitor response, with further analyses indicating the pathways involved.
Pointing to genomic heterogeneity among PsA patients, these studies represent the first steps on the road to personalized medicine. To be clinically useful, these patterns of gene expression will likely need simplification, with the so-called interferon signature in SLE an example of reducing complex patterns of gene expression into a single score. (Of course, the interferon signature is not that simple and signatures would be better descriptor than a single score. Nevertheless, the existence of the signature pointed the way to a new therapy.)
11. Deep Cellular Immune Profiling
Abstract 1689: Eder et al.11
At present, the field of laboratory testing is expanding dramatically, with the application of various platforms to obtain new types of biomarkers that can be useful for screening, diagnosis, classification, prognosis and theragnosis (prediction of treatment responses). Along with the big data studies to characterize patterns of cell expression in disease, mass spectrometry by CyTOF can provide a detailed picture of immune cell phenotype and function in patients with PsA and other forms of inflammatory arthritis.
Eder et al. analyzed peripheral blood immune cells in PsA to assess the relationship with clinical and imaging findings, as well as the response to treatment after three months for 34 patients. The investigators focused on 16 cell populations among CD3+ positive cells and identified three clusters, each of which was associated with different imaging findings by ultrasound assessment of synovitis, enthesitis and peritenon inflammation. Further, among the cell phenotypes observed, CD4+ memory and Th1 cells levels were correlated with more synovitis and reduced responses to advanced therapies while γδ cells and CD8+ cells were associated with milder disease and a better treatment responses.
At this point in time, novel biomarkers (e.g., transcriptomics, CyTOF) are in early phase of development, valuable for elucidating mechanisms and suggesting opportunities for refining and expanding the array of biomarkers that can be used clinically.
In Sum
I was excited to read the ACR Convergence 2023 abstracts. While still pondering their full implications, I already look forward to seeing the abstracts of 2024 to learn about the progress in personalized medicine and the application of modern science to improve clinical outcomes.
David S. Pisetsky, MD, PhD, is a professor of medicine and immunology at Duke University School of Medicine, Durham, N.C., and a staff rheumatologist at the Durham VA Medical Center. He also served as the first physician editor of The Rheumatologist.
References
- Olopoenia A, Martin A, Orroth K, et al. The burden of oligoarticular psoriatic arthritis in the United States [abstract]. Arthritis Rheumatol. 2023;75(suppl 9).
- Gutierrez Manjarrez J, Thib S, Cook R, Eder L. The association between sonographic imaging phenotype and response to treatment in patients with psoriatic arthritis [abstract]. Arthritis Rheumatol. 2023;75(suppl 9).
- Gazel U, Noges K, Ayan B, et al. Pain mechanisms in psoriatic arthritis: differentiating inflammation related pain in enthesitis using ultrasound, in comparison to functional MRI [abstract]. Arthritis Rheumatol. 2023;75(suppl 9).
- Raychaudhuri S, Abdelhafez Y, Mazza D, et al. Spinal inflammation a dominant pathology in psoriatic arthritis: Characterization and Quantification by In-Vivo 18F-FDG Total-Body PET/CT Imaging [abstract]. Arthritis Rheumatol. 2023;75(suppl 9).
- Macfarlane G, Rotariu O, Lembke S, et al. The influence of ‘fibromyalgia-ness’ on treatment response amongst patients with psoriatic arthritis (PsA). Results from the British Society for Rheumatology Psoriatic Arthritis Register (BSR‐PsA) [abstract]. Arthritis Rheumatol. 2023;75(suppl 9).
- Snoeck Henkemans S, de Jong P, et al. The window of opportunity in psoriatic arthritis: Similar to rheumatoid arthritis? [abstract]. Arthritis Rheumatol. 2023;75(suppl 9).
- Ganatra D, Samman A, Nasri D, et al. Identifying synovial fluid micro-RNA signature that distinguishes psoriatic arthritis from osteoarthritis [abstract]. Arthritis Rheumatol. 2023;75(suppl 9).
- Taylor P, Walsh D, Takeuchi T, et al. Direct and indirect effects of upadacitinib or adalimumab on pain in psoriatic arthritis: Results from a randomized phase 3 study [abstract]. Arthritis Rheumatol. 2023;75(suppl 9).
- Merjanah S, Driscoll D, Peloquin C, et al. Major adverse cardiovascular event and venous thromboembolism risk comparing advanced therapies among individuals with axial spondylarthritis and psoriatic arthritis [abstract]. Arthritis Rheumatol. 2023;75(suppl 9).
- Li Q, O’Rielly D, Jenkins K, et al. Identifying differentially expressed genes to predict TNF-alpha and IL-17A inhibitor response in psoriatic arthritis [abstract]. Arthritis Rheumatol. 2023;75(suppl 9).
- Eder L, Li X, Thib S, et al. Deep cellular immune profiling in psoriatic arthritis correlates with imaging phenotypes and response to targeted advanced therapy [abstract]. Arthritis Rheumatol. 2023;75(suppl 9).
