Top Research in Axial Spondyloarthritis Presented at ACR Convergence 2022

An illustration of ankylosing spondylitis. Source: Dominique Duval/Science Source

PHILADELPHIA—Approximately 100 research abstracts on axial spondyloarthritis (axSpA) were accepted for presentation at ACR Convergence 2022. It is exciting to see a wealth of research on axSpA being undertaken worldwide. Here, we highlight important points from 10 of these studies.

1. Abstract 0378: Prevalence of Axial Spondyloarthritis (axSpA) in Patients Treated for Chronic Back Pain in Chiropractic Clinics; Deodhar et al.¹

AxSpA is a common inflammatory disease that usually presents with chronic back pain. Because of the frequency of back pain (both acute and chronic) in the general population, patients may initially seek care from providers other than rheumatologists. Chiropractors are often these providers, raising a question about the prevalence of undiagnosed axSpA among patients seen in chiropractic practices.

In their study, Deodhar and colleagues used phone consultations to determine whether patients seen in four chiropractic clinics had findings suggestive of axSpA, thereby providing the basis for referral to a rheumatologist. For this purpose, patients were screened for evidence of axSpA using a referral strategy from the Assessment of SpondyloArthritis Society (ASAS), inquiring about inflammatory back pain, family history, inflammatory bowel disease, psoriasis, response to non-steroidal anti-inflammatory drugs (NSAIDs), history of heel pain, uveitis and peripheral arthritis. With more than one of these features present, the patient was referred to a rheumatologist for a more complete examination involving testing for HLA-B27 and imaging, including magnetic resonance imaging (MRI) of the sacroiliac joints. An academic rheumatologist made the final determination of whether the patient had axSpA.

Results of this study indicated that approximately 12% of patients with chronic back pain in the care of chiropractic clinics had undiagnosed axSpA; for these patients, non-radiographic axSpA was the most common form of disease.

Although the results of this study are important, perhaps they are not surprising: Diagnostic delay of axSpA appears common, screening questions about inflammatory back pain have limitations, and axSpA is only one of many conditions that cause chronic back pain. Certainly, greater implementation of screening by the ASAS strategy would be worthwhile, but identifying patients with axSpA, especially those with non-radiographic disease, will remain challenging.

2. Abstract 0420: What Does It Mean—A Good Response to NSAIDs? A Systematic Comparison of Patients with Axial Spondyloarthritis and Controls with Chronic Back Pain; Baraliakos et al.²

In identifying patients with axSpA, an important screening question relates to the response of symptoms to NSAIDs, with a rapid response suggesting underlying inflammatory arthritis. The definition of a rapid response is unclear, however, as is the specificity of a rapid response for axSpA, as opposed to other common forms of back pain.

In addition to the setting of diagnosis, the rapidity and magnitude of an NSAID response are relevant to decision making about therapy, including the institution of biologic disease-modifying anti-rheumatic drugs (DMARDs). Providing a more precise definition of a rapid NSAID response would, therefore, be valuable for both initial diagnosis and subsequent management.

To assess the velocity and magnitude of NSAID response in axSpA and other conditions, Baraliakos et al. studied 68 patients with axSpA, 107 patients with degenerative back pain and 58 patients with non-specific back pain. Patients had more than 10 years of symptoms, but were either NSAID naive or had an inadequate response to an NSAID at lower than the maximum dose. Inflammatory back symptoms were reported by 75%, 57% and 60% of patients in the three groups, respectively. Importantly, after treatment with the maximum doses of NSAIDs, with an agent previously tolerated at lower doses, the cumulative response rates of the three groups were similar, with an overall response in 27–30% of patients. Among patients with axSpA, the responses to NSAIDs were considered to be better, but not more rapid, than those of the other groups; better responses occurred in men and those earlier in the disease course.

The findings in this study are intriguing and suggest limitations in the current screening questions to identify patients with axSpA among the larger group of patients with chronic low back pain. These limitations also affect decision making. Future improvement in the identification and treatment of patients with axSpA will depend on better screening questions or better biomarkers (or both) to get beyond the ambiguities inherent in the concepts of inflammatory back pain and rapid response to NSAID treatment.

3. Abstract 0406: Are the BASDAI & BASFI Capturing the Full Impact of Disease Activity on Quality of Life in Women with Axial Spondyloarthritis?; Maguire et al.³

Historically, ankylosing spondylitis has been long viewed as a disease of men, just as systemic lupus erythematosus (SLE) has been viewed as a disease of women. Although the ratio of the sexes in SLE has not changed much over the years, the situation with ankylosing spondylitis is quite different. Reflecting new technologies (i.e., MRI) and new disease classifications (i.e., radiographic and non-radiographic axial spondylitis [nr-axSpA]), the prevalence of axSpA determined in women has increased significantly. Certain aspects of axSpA may differ between men and women (e.g., pattern of axial involvement), but the difference in the frequency of disease in men and women is continuously narrowing.

Because some of the metrics used to assess axSpA were developed primarily with male patients, their application to women has not yet been well defined. In this study, Maguire and colleagues assessed the sex-specific relationship between a measure of disease activity (the BASDAI) and patient quality of life (ASQoL) in a large patient population from the Ankylosing Spondylitis Registry of Ireland.

For both measures, women had significantly higher scores than men, although a strong correlation between BASDAI and ASQoL existed in the overall population. The relationship between these two measures differed for men and women, with the correlation between disease activity and quality of life stronger for men than women.

This study is important in highlighting the potentially greater impact of disease activity of axSpA on quality of life in women; it also suggests the need to develop new tools that can better capture disease activity and the patient experience in women, as well as men.

4. Abstract 0427: Recapture Rates with Ixekizumab After Withdrawal of Therapy in Patients with Axial Spondyloarthritis: Results at Week 104 from a Randomized Placebo-Controlled Withdrawal Study; Landewé et al.⁴

In the treatment of any form of inflammatory arthritis with biologics, an important issue concerns the consequences of stopping therapy in patients who have achieved significant benefit. Landewé and colleagues have addressed this issue in a study of patients with ankylosing spondylitis and non-radiographic axial spondyloarthritis who achieved remission with ixekizumab, a monoclonal antibody to the cytokine interleukin (IL) 17A. In a phase 3 extension study, 155 patients participated in a randomized withdrawal-retreatment period (RWP), at a ratio of 2:1, using one of two regimens of ixekizumab (once every two weeks or once every four weeks) or placebo. The frequency of flare and recapture rates were determined. With flare, patients were restarted on open-label ixekizumab.

Of the patients who received placebo, 36% did not flare. Of 28 patients who did flare, four improved without retreatment with open-label ixekizumab. Switching to open-label ixekizumab recaptured low disease activity in 82%. Twelve of 28 patients met criteria for inactive disease. Thus, with flare after discontinuation of therapy, the majority of patients responded well to treatment again and attained either low disease activity or remission.

Studies of this kind are important in determining the use of biologics for a chronic disease, such as axSpA. Perspectives on results of this kind will likely vary and relate to the perceived outcome of drug discontinuation.

If a sizable number of patients do not flare after drug withdrawal and those who flare can improve after resumption of therapy, then a trial off medication could be considered reasonable. On the other hand, a flare has its own problems, including a period of increased patient symptomatology, the potential for occurrence of damage by uncontrolled disease in the interim and the attendant worry that the patient will not respond well again if therapy is resumed.

Future studies are needed to determine the consequences of withdrawal of biologics, not only in axSpA, but also in other forms of inflammatory arthritis treated with these agents.

5. Abstract 0547: Profiling of Systemic Immune Responses in Axial Spondyloarthritis Patients Reveals Strikingly Distinct Cellular & Molecular Mechanisms of Action of IL-17A Inhibitors & TNF-Blockers; Rosine et al.⁵

Inhibitors of tumor necrosis factor (TNFi) and IL-17A (IL-17i) have both received regulatory approval for the treatment of axSpA and produce generally similar levels of effectiveness. This similarity, however, does not imply the mechanisms of action of these biologics are the same. Because clinical trials for agents to treat axSpA have not yet involved extensive study of biomarkers beyond C-reactive protein (CRP), the immune system changes produced by these biologics are, by and large, unknown. Biomarkers would be very valuable for making treatment decisions in personalized medicine approaches, as well as in assessing risks for potential side effects.

In a fascinating study, Rosine and colleagues characterized the in vitro immune responses of peripheral blood cells of axSpA patients before and after therapy with either a TNFi or an IL-17i. Using whole-blood stimulation assays, the researchers determined patterns of gene expression by RNA sequencing; the production of secreted molecules was measured by proximity ligation assays.

The results of this study were striking. Treatment with IL-17i did not lead to significant changes in gene expression when comparing responses before and after 16 weeks of treatment. In contrast, treatment with a TNFi was associated with changes in the expression level of more than 4,000 genes. Assays of secreted cytokines and chemokines following TNFi therapy also showed strong effects on both innate and adaptive arms of the immune system. In contrast, treatment with IL-17i led to only minor changes in responses. For easily measured biomarkers, TNFi, but not IL-17i, affected neutrophil and monocyte numbers.

These studies suggest that TNFi and IL-17A have distinct mechanisms of action and that IL-17A may act on non-immune cells or preferentially affect immune cells at the site of inflammation.

Assessing effects of drugs on cells at the sites of inflammation will be very difficult, however, due to the nature of involvement of axSpA; lesions of the spine are not readily accessed except at the time of surgery or autopsy. Perhaps, studies of patients with peripheral arthritis could provide at least some synovial fluid or synovial tissue for comparative analysis of drug action.

Leading to more questions than answers, this study indicates that peripheral blood may not always yield information about relevant pharmacodynamic actions of an agent and that an effective agent may have seemingly few effects on circulating immune cells.

6. Abstract 1255: Rheumatologists Overcall Sacroiliitis on X-ray & MRI in Axial Spondyloarthritis Patients: Data from the Belgian Inflammatory Arthritis & Spondylitis Cohort (Be-GIANT); De Craemer et al.⁶

This year’s ACR meeting featured many abstracts on the use of MRI to evaluate axSpA, seeking to improve and refine this important imaging modality. This paper asks a simpler, but nevertheless important, question: How reliable are sacroiliac (SI) joint X-ray (X-SIJ) and MRI readings of rheumatologists compared with those of experienced, so-called calibrated, central readers?

To answer this question, De Craemer and colleagues evaluated imaging studies of newly diagnosed, biologic-naive patients with axSpA in a cohort from Belgium. Rheumatologists came from both academic and community hospitals.

The results of this study indicated differences between rheumatologists and central readers in 13% of X-SIJ, most of which were reclassified as not showing radiographic sacroiliitis. Agreement with central readers was higher with academic rheumatologists than community rheumatologists. For MRI images, about 25% were reclassified by central readers, the majority of which were considered negative. Again, agreement between central readers and rheumatologists was greater for academic than community practitioners.

Two important conclusions emerge from this study. The first is that the prevalence of radiographic evidence of sacroiliitis is low with newly diagnosed patients. The second conclusion is that sacroiliitis on MRI may be overcalled by rheumatologists in both academic and community practices. This study indicates the need for additional training in imaging of the SIJ, but it also suggests the opportunity for innovation; because current technology readily allows transmission of imaging data for remote reading, the regular involvement of more experienced readers is possible.

7. Abstract 1155: Epigenome-Wide Integrative Association Study on Spondyloarthritis & Psoriatic Arthritis; Carnero-Montoro et al.⁷

Big data describes a type of scientific study that generates a large number of data points for a single determination. With current technology, big data encompasses genetics, genomics, proteomics and metabolomics, among other omics, and can provide an extremely granular picture of molecular changes associated with disease. Epigenomics is a particularly interesting approach because it can identify changes in the methylation status of genes that impact the level of gene expression and, thereby, the function of cells. A host of factors can lead to changes in the epigenome, with inflammation an important influence.

The study by Carnero-Montoro et al. measured genome-wide DNA methylation at more than 800,000 sites in the peripheral blood of patients with spondyloarthritis (SpA) and psoriatic arthritis (PsA). The changes in methylation were related to both genotyping and clinical data.

Using the Illumina MethylationEPIC Array to study 207 patients with SpA and 56 patients with PsA, the largest number of methylation differences could be explained by HLA-B27 positivity. Other differences related to genes with enriched expression in the large intestine. Overall, the pattern of methylation could distinguish axSpA and PsA and involved genes for Janus kinase 2 (JAK2), which is involved with cytokine stimulation, and the TRIM25 gene, which has a role in antiviral response. Of interest, some methylation differences were associated with disease activity.

Because clinicians are accustomed to simple markers, such as CRP (little data), learning to use results of big data will require education and, hopefully, informatics approaches to reduce the size of big data into more manageable and accessible sizes.

8. Abstract 1162: Improvement of Gut Microbiota Dysbiosis in Patients with Axial Spondyloarthritis After One Year of Biological Treatment; Rios-Rodriguez et al.⁸

The microbiome defines the array of bacterial, fungal and viral organisms that inhabit the human body, existing symbiotically but nevertheless impacting metabolism as well as the immune system. Disturbances in the composition of the microbiome are termed dysbiosis and have been explored in a wide number of inflammatory diseases including SpA. Although dysbiosis has been associated with SpA—as well as other diseases, such as rheumatoid arthritis and systemic lupus erythematosus—the effects of treatment are only now being determined. Defining these changes is important to understand the relationship of dysbiosis to disease pathogenesis and, correspondingly, the manner in which therapy affects the overall immune system, including the contribution of the microbiome.

This study by Rios-Rodriquez and colleagues characterized the microbiome of 99 patients with axSpA and 63 healthy controls, using 16 rRNA gene sequencing to determine the composition of the gut microbiota. The microbiome was then evaluated following treatment primarily with TNF inhibitors (97.9%), with 2.1% receiving an inhibitor of IL-17.

Results of these studies indicated that, at the genus level, patients showed reduction of Lachnospiraceae taxa, such as Blauti, Roseburia and Fusicatenibacter, but enrichment in Collinsella, compared with controls at baseline. With a year of treatment, shifts in the microbiota occurred, with an increase in those species depleted at baseline; Collinsella also showed a slight increase. Shifts in abundant species, such as Prevotella and Bacteroides, were correlated with disease activity.

The changes in the composition of the microbiome of patients are fascinating and point to the complex interactions of the host and its microbial inhabitants. The enrichment of Collinsella in patients with axSpA and the stability of expression over time may point to a more direct role in pathogenesis. What is Collinsella doing in patients with SpA? The answer is unknown, but it could shed important insights on underlying disease mechanisms.

9. Abstract 1673: Preconceptional NSAID Treatment Exposure Is Associated with a Significantly Longer Time to Conception in Women with Spondyloarthritis: Analysis of the Prospective GR2 Cohort; Hamroun et al.⁹

Spondylitis, once considered a disease of men, is now recognized to be a common form of arthritis in women, with disease onset often occurring during the child-bearing years. Reflecting the focus on men in previous studies, only limited data on the impact of SpA on fertility and pregnancy outcomes exist. Information on these issues is very important in shared decision making with patients about therapy both prior to and during pregnancy. Although these issues have been most extensively investigated in systemic lupus erythematosus, extrapolation is uncertain in view of the differences in underlying pathophysiology and medication choices among various inflammatory diseases.

In this study, Hamroun and colleagues investigated factors associated with time to conception in a large, multicenter, national cohort in France. Patients were included in the study for a pregnancy wish or because of an actual pregnancy. The main end point was time to conception, with an analysis of the effects of a large number of variables, including disease activity, medications and age, among other factors. For the 207 patients analyzed, the factors that contributed to a long time to conception were age and use of NSAIDs in the pre-conception period.

These results are interesting and important as the understanding of SpA in women evolves, but they are likely incomplete because study entry required a prior diagnosis of SpA. Data suggest the diagnosis of SpA is often delayed—10 years may not be unusual—so it would be expected that many women with SpA go undiagnosed prior to pregnancy attempts. It is relevant to ask whether sub-fertility in some women may result from underlying SpA or other inflammatory disease.

A related issue concerns the origin of musculoskeletal symptoms associated with pregnancy and the post-partum period; among these, back pain is one of the most common. Are these symptoms manifestations of SpA in some women? With the recognition of SpA in women, the investigation of fertility and pregnancy outcomes are now important items on the research agenda.

10. Abstract 1492: Treatment with Non-Steroidal Anti-inflammatory Drugs Is Associated with Retardation of Radiographic Spinal Progression in Patients with Axial Spondyloarthritis: 10-Year Results from the German Spondyloarthritis Inception Cohort; Torgutalp et al.¹⁰

For many years, NSAIDs were the mainstay in the treatment of axSpA, with evidence suggesting an impact on radiographic progression. With the development of new therapies (especially biologic inhibitors of TNF and IL-17), the role of NSAIDs has become uncertain; concerns about toxicity (e.g., gastrointestinal bleeding, hypertension and renal insufficiency) suggest against prolonged use. Nevertheless, the question of whether NSAIDs can impact the radiographic progression of axSpA remains.

Torgutalp et al. analyzed radiographic disease progression of 243 patients from the German Spondyloarthritis Inception Cohort (GESPIC) over a period of two years. Radiographs were scored according to the modified Stoke Ankylosing Spondylitis Spine Score (mSASSS) and related to intake of NSAIDs, both non-selective and COX-2 selective agents (COX-2i). In this analysis, higher NSAID intake was related to reduced radiographic progression, with most of the effects among patients with radiographic axSpA in contrast to those who were considered to have nr-axSpA. Interestingly, progression was lower in those on COX-2i’s than those on non-selective agents.

The findings are interesting and raise important question about the role of NSAIDs in the management of axSpA. Perhaps these agents are a useful adjunct to current biologic therapy, further blocking inflammation and thereby radiographic progression. The benefits of COX-2i’s are particularly interesting because indomethacin was long considered of particular benefit in axSpA, even though this agent is non-selective and is quite active against COX-1. The toxicity of NSAIDs has diminished enthusiasm for their use; nevertheless, they are effective agents and still worth considering among the growing armamentarium of agents used to treat axSpA.

Dr. Pisetsky

David S. Pisetsky, MD, PhD, is a professor of medicine and immunology at Duke University School of Medicine in 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

1. Deodhar A, Kiwalkar S, Phung Nguyen K, et al. Prevalence of axial spondyloarthritis (axSpA) in patients treated for chronic back pain in chiropractic clinics [abstract]. Arthritis Rheumatol. 2022;74(suppl 9).
2. Baraliakos X, Redeker I, Bergmann E, et al. What does it mean—A good response to NSAIDs? A systematic comparison of patients with axial spondyloarthritis and controls with chronic back pain [abstract]. Arthritis Rheumatol. 2022;74(suppl 9).
3. Maguire S, Wilson F, Gallagher P, et al. Are the BASDAI & BASFI capturing the full impact of disease activity on quality of life in women with axial spondyloarthritis? [abstract]. Arthritis Rheumatol. 2022;74(suppl 9).
4. Landewé R, Poddubnyy D, Rahman P, et al. Recapture rates with ixekizumab after withdrawal of therapy in patients with axial spondyloarthritis: Results at week 104 from a randomized placebo-controlled withdrawal study [abstract]. Arthritis Rheumatol. 2022;74(suppl 9).
5. Rosine N, Koturan S, Guillemot V, et al. Profiling of systemic immune responses in axial spondyloarthritis patients reveals strikingly distinct cellular and molecular mechanisms of action of IL-17A inhibitors and TNF-blockers [abstract]. Arthritis Rheumatol. 2022;74 (suppl 9).
6. De Craemer A, de Hooge M, Renson T, et al. Rheumatologists overcall sacroiliitis on X-ray and MRI in axial spondyloarthritis patients: Data from the Belgian inflammatory arthritis and spondylitis cohort (Be-GIANT) [abstract]. Arthritis Rheumatol. 2022;74(suppl 9).
7. Carnero-Montoro E, Arias-de la Rosa I, Lopez-Medina C, et al. Epigenome-wide integrative association study on spondyloarthritis and psoriatic arthritis [abstract]. Arthritis Rheumatol. 2022;74(suppl 9).
8. Rios-Rodriguez v, Essex M, Rademacher J, et al. Improvement of gut microbiota dysbiosis in patients with axial spondyloarthritis after one year of biological treatment [abstract]. Arthritis Rheumatol. 2022;74(suppl 9).
9. Hamroun S, Couderc M, Flipo R, et al. Preconceptional NSAID treatment exposure is associated with a significantly longer time-to-conception in women with spondyloarthritis: Analysis of the prospective GR2 cohort [abstract]. Arthritis Rheumatol. 2022;74(suppl 9).
10. Torgutalp M, Rios-Rodriguez v, Dilbaryan A, et al.Treatment with non-steroidal anti-inflammatory drugs is associated with retardation of radiographic spinal progression in patients with axial spondyloarthritis: 10-year results from the German spondyloarthritis inception cohort [abstract]. Arthritis Rheumatol. 2022;74(suppl 9).