Redrawing the Dividing Line in Nonradiographic axSpA
As a general rule, rheumatologists are comfortable with ambiguous diagnoses, given that so many of the conditions they manage are marked by a wide spectrum of presentations and manifestations. A prime example is axial spondyloarthritis, which is divided into two categories: radiographic axial spondyloarthritis — also called ankylosing spondylitis — and nonradiographic axial spondyloarthritis. Discussion surrounding the nonradiographic form of this disease has been known to prompt the almost preposterous fundamental question: Is it even a thing?
“It absolutely is a real thing,” Philip J. Mease, MD, of the Swedish Medical Center and the University of Washington, told Healio Rheumatology. “But, hopefully, in 10 years, we will no longer have this arbitrary categorization of radiographic and nonradiographic axSpA. It is all one condition, all just part of a spectrum of disease.”
There are a few central issues pertaining to nonradiographic axSpA that must be considered, and from which all other issues stem. The first is the 2009 Assessment of SpondyloArthritis International Society (ASAS) classification criteria that split the two conditions and more or less defined the nonradiographic patient population.
ASAS took on the task of attempting to create a dividing line where a dividing line does not exist, given that these patients largely fall somewhere on a spectrum rather than in clearly defined categories. Patients with radiographic and nonradiographic disease often have similar clinical presentation, pain parameters and reductions in both function and quality of life.
The key difference, as the names would suggest, pertains to radiology results. For ankylosing spondylitis patients, there is definite sacroiliitis on X-ray, and for nonradiographic axSpA patients, there is no sacroiliitis on X-ray, although there may be evidence of inflammation on MRI.
But even this is debatable, according to Atul Deodhar, MD, professor of medicine in the division of arthritis and rheumatic diseases at the Oregon Health & Science University School of Medicine. “The degree of sacroiliitis on X-ray often depends on who is reading the X-ray,” he said. “I might see grade 2 bilateral sacroiliitis, and so I would say it is ankylosing spondylitis. But another person might see grade 2 on one side and grade 1 on the other and define it as nonradiographic axSpA. It is often in the eye of the beholder.”
This disparity bleeds into the second key issue. “Essentially, the FDA was not convinced that they liked the ASAS criteria,” Eric M. Ruderman, MD, associate chief of clinical affairs in the division of rheumatology at the Northwestern University Feinberg School of Medicine, said in an interview. “They required proof that nonradiographic and radiographic patients had comparable disease.”
While treatment regimens for nonradiographic and radiographic patients are largely the same, the FDA’s decision has had a ripple effect on prescribing behaviors and technicalities in making a diagnosis.
A third key issue is that because nonradiographic disease can present in myriad ways, rheumatology is often the last stop for many of these patients. They have back pain, so they see primary care, an osteopath or physiatrist, and are often misdiagnosed — to the point that they may undergo unhelpful, unnecessary surgeries. By the time they find their way to rheumatology, years have passed, or longer. Rheumatologists hope to fill knowledge gaps and get these patients diagnosed sooner.
To increase clinician awareness of the disease, it will be useful to also address other critical questions about this patient population, including those surrounding sex distribution among patients under the ankylosing spondylitis umbrella; the prevalence of human leukocyte antigen (HLA)-B27 and C-reactive protein (CRP) status; and uncertainties about what proportion of patients progress from nonradiographic to radiographic disease.
Nonradiographic axial spondyloarthritis is most certainly a thing, but it is a complicated thing.
Understanding the Patient Journey
To understand exactly what this thing is, it is important to determine who these patients are and what they experience.
While most rheumatologists are clear on the enthesitis, uveitis, family history information and other components of the ASAS criteria, experts like Vibeke Strand, MD, adjunct clinical professor in the division of immunology/rheumatology at Stanford University School of Medicine, know that nonradiographic axSpA can generally be diagnosed in minutes, and with a few simple questions.
“If I have a patient who is younger than 45 years who has inflammatory back pain, and that back pain often comes on in the middle of the night, and is only relieved by getting up and walking around, then, even before I do evaluations, I am thinking that this person has spondyloarthropathy,” Strand said. “This is the classic symptom for nonradiographic axSpA.”
Each expert who spoke with Healio Rheumatology provided a nearly word-for-word description of such a patient. “They will have a normal X-ray and an abnormal MRI showing inflammation of the sacroiliac joints, or bone marrow edema in joints adjacent to the sacroiliac,” Mease said.
There can be variations, of course, he noted. “They can be HLA-B27 negative, but if they are negative, then they need a history of uveitis or a family history of spondylitis. But these patients look largely similar.”
If most patients with nonradiographic axSpA have the defining hallmark symptom of back pain that wakes them in the middle of the night, it begs the next important clinical question: Why are so many of them misdiagnosed?
Running the Diagnosis Gauntlet
“Delayed diagnosis of axial spondyloarthritis has always been a problem,” Marina N. Magrey, MD, a rheumatologist at MetroHealth Medical Center, said in an interview. “The average patient takes a journey of 7 to 14 years from first symptoms to diagnosis.”
The primary reason for this, as discussed previously, is that they may see a podiatrist for a complaint in the plantar fascia, or they may see a chiropractor for their back pain, or they may simply think that their inflammatory back pain is actually mechanical back pain due to some real or perceived injury.
The difference between inflammatory and mechanical back pain is critical to understanding nonradiographic axSpA, according to Strand. “Inflammatory back pain gets better with activity,” she said. “Mechanical back pain gets better with rest. If a patient feels better when they move around, again, we can be fairly certain that it is inflammatory.”
While Strand acknowledges that non-rheumatologists are often the physicians missing the diagnosis in these patients, she does not let her specialty off the hook. “Performing a Schober’s test or a sacroiliac stretch test can tell you if the patient has sacroiliitis, even if it is early in the progression of the disease, or if there is no evidence of it on X-ray,” she said. “These tests used to be de rigueur for rheumatologists, but many of the fellows I encounter are not familiar with them. We need to do a better job of educating the next generation about these diagnostic tests.”
The other issue in diagnosing nonradiographic patients, of course, is that there are no changes on standard imaging, according to Magrey. “As physicians, we heavily rely on imaging,” she said. “If we do not see anything, it can put us in a predicament that often delays diagnosis.”
The solution seems to be MRI. But even this can come at a cost, according to Deodhar. “MRI scans can show us evidence of inflammation and damage in the sacroiliac joints even before it can be seen on X-ray, because MRI scans are so sensitive,” he said. “But this sensitivity created the problem of overdiagnosis. Again, you have to be careful about how you interpret all imaging analyses.”
For Magrey, there is one final component that can be problematic in this regard. “HLA-B27 has been used as a biomarker, but about 7% of [white people] in the U.S. can have HLA-B27 positivity without having disease,” she said. “That exemplifies another problem we have, which is that the biomarkers are not perfect.”
Unfortunately, the problems facing these patients — and facing the rheumatologists who treat them — are often larger than a faulty biomarker, and stem from regulatory decisions made over the last decade.
In 2013, the FDA was not willing to consider nonradiographic patients as part of the overall spectrum of disease, according to Mease. “They recognized that there were patients, but they felt more studies were needed for patients who did not fulfill the classification criteria for radiographic axSpA,” he said.
“We have spent a lot of time trying to convince the FDA that nonradiographic axSpA was the same disease, just a different set of manifestations on the spectrum of spondyloarthritis,” Strand added.
To that end, Deodhar and Strand wrote an editorial offering commentary on the FDA’s decision. “The main reason to distinguish between these ends of the spectrum of axSpA was that tumor necrosis factor (TNF) inhibitors approved for [ankylosing spondylitis] could obtain additional labelling for [nonradiographic] axSpA and be used to treat all patients manifesting clinical features of axSpA,” they wrote. “These two terms are distinguished by the degree of ‘radiographic sacroiliitis’ assessed by conventional radiography, according to the 1984 [modified New York] criteria for [ankylosing spondylitis]. Since this differentiation has been shown to be not very reliable, we argue that the terms [nonradiographic] axSpA and [ankylosing spondylitis] should only be used for classification of patients with axSpA and not as separate diagnoses.”
The reason for this is simple: Essentially the same drugs are effective in nonradiographic and radiographic patients. However, if a patient is diagnosed with nonradiographic disease, they may run into coverage or reimbursement issues if they are treated with an ankylosing spondylitis drug that is not specifically indicated for the nonradiographic population.
The good news is that 2019 was a landmark year for all patients under the axSpA umbrella. Certolizumab pegol (Cimzia, UCB) gained the first FDA approval specifically for patients with nonradiographic axSpA.
Also in 2019, Ward and colleagues updated the ACR/Spondylitis Association of America/Spondyloarthritis Research and Treatment Network recommendations for the treatment of ankylosing spondylitis and nonradiographic AxSpA. Importantly, they suggest that treatments for the two conditions are “similar,” with TNF inhibitors recommended as first-line therapy. For second-line therapy, IL-17 inhibitors secukinumab (Cosentyx, Novartis) or ixekizumab (Taltz, Lilly) are recommended over a second TNF inhibitor or tofacitinib.
Methotrexate with a TNF inhibitor is contraindicated, as is a strict treat-to-target strategy or discontinuation or tapering of biologic therapy in patients showing stable disease.
Regarding imaging, spine or pelvic MRI may “aid assessment,” according to the authors, but they recommend against routine monitoring of radiographic changes with serial spine X-rays.
“Some of these guidelines are very conditional, rather than hard and fast rules for treatment,” Magrey said. “If I could highlight one important message, it is that methotrexate does not have any role in patients with axial disease, neither does oral prednisone.”
While Deodhar is encouraged by the growing number of treatment options, he said that the reality of treating a patient with axSpA does not necessarily hinge on FDA approval. He noted that etanercept (Enbrel, Amgen), infliximab (Remicade, Janssen), adalimumab (Humira, Abbvie) and golimumab (Simponi, Janssen) all have FDA approval only for ankylosing spondylitis but are often used to treat nonradiographic axSpA. “You cannot practice rheumatology by using only what the FDA approves,” he said. “I am not bound by the FDA, I am bound by my medical license. All of these drugs can have efficacy in nonradiographic axSpA patients, so I use them.”
While clinicians have a largely reliable armamentarium of drugs to treat this population, experts like Ruderman see further complications on the horizon as a result of decisions made not only by the FDA, but by European regulators as well.
The FDA approvals of secukinumab and ixekizumab came on the heels of what Ruderman described as a “bizarre” series of phase 3 studies. In the EU, just 16 weeks of treatment is required to show efficacy of an axSpA drug, whereas 52 weeks is necessary in the U.S. “These studies had a 16-week endpoint for Europe and a 52-week endpoint for the U.S.,” Ruderman said.
One of the trials Ruderman is referring to is the PREVENT trial of secukinumab, which was presented by Deodhar at ACR 2019. The study included patients treated with both a loading dose and a non-loading dose of secukinumab, along with a placebo arm. Results at 16 weeks showed response rates for both the loading (40.8%) and non-loading (40%) groups bested placebo (28%) in terms of the primary endpoint of ASAS40 response. Comparable results persisted through 52 weeks.
For ixekizumab, Deodhar and colleagues published findings from the COAST-X trial, which showed response rates of 35% and 40% for ixekizumab, compared with 19% for placebo, at 16 weeks. For 52-week results, response rates were above 30% for two study drug arms and 13% for placebo.
The 52-week studies presented a number of complications for the research community, according to Strand. “The FDA thought that nonradiographic patients being treated with placebo would spontaneously go into remission, but they did not,” she said. “The majority of patients could not stand it; they had pain and worsening of disease, and discontinued protocol treatment for rescue.”
Ruderman believes this is going to present ethical conundrums moving forward. “Why do we need a 52-week trial if we can prove that these drugs work at 16 weeks?” he said. “If I was AbbVie or Gilead looking to get approval for a JAK inhibitor, I would think twice about spending the money and putting patients through a 52-week study. I just do not know if we can ethically put patients on placebo for that long.”
Most experts believe that both ASAS and the FDA acted in good faith. But Ruderman voiced the frustration many of these clinicians have had with the distinctions. “My beef with the FDA criteria is that it got lost in the semantics,” he said. “Many people are confused by what constitutes nonradiographic and radiographic axSpA. We have to just keep reminding ourselves that these are all patients with one disease who exist on this spectrum.”
Rate of Progression
If there is another key question for clinicians to consider about patients who exist on a disease spectrum, it pertains to the proportion of patients who will progress from the milder to the more severe form of the disease. But data are beginning to show that confirmed nonradiographic axSpA does not necessarily portend ankylosing spondylitis.
In their recent study in PLoS One, Hebeisen and colleagues studied 88 nonradiographic patients and 418 radiographic patients to determine whether there is an association between spinal radiographic progression and structural damage at the sacroiliac level. Results showed significantly lower rates of progression at 2 years among nonradiographic patients compared with radiographic (OR = 0.33; 95% CI, 0.13-0.83).
Similarly, Protopopov and colleagues conducted a comprehensive review of nonradiographic axSpA patients to determine rates and predictors of radiographic progression in the sacroiliac joints and in the spine. Their results showed that most studies reported that between 10% and 40% of patients progress over a period of 2 to 10 years.
“Not everybody progresses,” Magrey said. “On average, we think that about 10% progress over 2 years and 25% will progress over 10 years. But the rate of progression is variable.”
Magrey added that clinicians are beginning to paint a picture of who is and is not likely to progress. “Patients who have some damage and a high severity score are likely to progress, along with patients who have elevated markers like ESR and CRP,” she said. “Smokers are also more likely to progress to ankylosing spondylitis.”
Other findings from Protopopov and colleagues published in Expert Review of Clinical Immunology indicated that not enough evidence was available to determine whether any treatment modality was associated with progression.
“Radiographic progression in the spine is in general low in [nonradiographic] axSpA,” they wrote. “Thus, long-term studies are required to investigate the natural course of the progression and possible treatment effects.”
Despite all of these uncertainties, the clinical and research communities have arrived at one fairly certain conclusion about who is likely to progress: Men.
Battle of the Sexes
“Males have higher risk of progression, by a 2:1 or 3:1 ratio,” Deodhar said. “However, we do not know why.”
For a long time, ankylosing spondylitis was viewed as a male-dominated condition, with a 2:1 incidence ratio of men to women. But when nonradiographic disease is factored into the equation, it balances out, as roughly an equal number of women as men are impacted with nonradiographic axSpA.
What experts do not yet know is why this phenomenon occurs. “Women tend to have more peripheral disease and more generalized body pain,” Magrey said. “Incidentally, this is why women get misdiagnosed with fibromyalgia or chronic pain syndrome. But we do not yet know why men tend to progress and women do not.”
Strand added that researchers have also not determined if it is simply the case that women progress more slowly than men, or that women generally have more benign disease. Magrey agreed. “In terms of the burdens of everyday functioning, quality of life, and pain parameters, men and women with nonradiographic disease are equally impaired,” she said.
All of that said, the fact that epidemiologists have answered at least some of the questions surrounding sex distribution in these patients is good news, according to Deodhar. “We have a better idea of where these patients are hiding,” he said.
The FDA approval of certolizumab pegol for nonradiographic disease is another big step, and experts hope that the message about the 52-week endpoint is finally hitting home.
But if there is an issue that requires ongoing attention, it is the frequency of misdiagnosis by other health care professionals, and the resulting complications.
Strand believes that rheumatologists are doing better in filling knowledge gaps, but she stressed urgency in mitigating misdiagnoses. “These patients have suffered a long time, and have often had to recover from unnecessary surgeries, and that is sad,” she said. “We need to do better at recognizing the symptoms and physical findings of inflammatory back pain in nonradiographic axSpA patients.” – by Rob Volansky
- Deodhar A, et al. Ann Rheum Dis. 2016;doi:10.1136/annrheumdis-2015-208852.
- Deodhar A, et al. Lancet. 2020;doi: 10.1016/S0140-6736(19)32971-X.
- Hebeisen M, et al. PLoS One. 2020;doi:10.1371/journal.pone.0230268.
- PREVENT trial: https://clinicaltrials.gov/ct2/show/NCT02696031. ClinicalTrials.gov Identifier: NCT02696031.
- Protopopov M, et al. Expert Rev Clin Immunol. 2018; doi:10.1080/1744666X.2018.1477591.
- For more information:
- Atul Deodhar, MD, can be reached at 3270 SW Pavilion Loop, Portland, OR 97239; email: firstname.lastname@example.org.
- Marina N. Magrey, MD, can be reached at 2 500 MetroHealth Dr., Cleveland, OH 44109; email: email@example.com.
- Philip J. Mease, MD, can be reached at 601 Broadway, Seattle, WA 98122; email: firstname.lastname@example.org.
- Eric M. Ruderman, MD, can be reached at NMH/Galter Room 14-100, 675 N Saint Clair Chicago, Ill., 60611; email: email@example.com.
- Vibeke Strand, MD, can be reached at 306 Ramona Rd., Portola Valley, CA 94028; email: firstname.lastname@example.org.
Disclosures: Deodhar reports consulting for and being on the advisory boards of AbbVie, Amgen, Boehringer Ingelheim, Celgene, Eli Lilly, Galapagos, GlaxoSmithKline, Janssen, Novartis, Pfizer and UCB; and receiving research grants from AbbVie, Eli Lilly, GlaxoSmithKline, Novartis, Pfizer and UCB. Magrey reports consulting for Eli Lilly, Jansen, Novartis and Pfizer, and conducting research with AbbVie and UCB. Mease reports grants, consulting fees and/or speaking fees from AbbVie, Amgen, Boehringer Ingelheim, Bristol Myers Squibb, Celgene, Eli Lilly, Galapagos, Genentech, Gilead, Janssen, Novartis, Pfizer, SUN and UCB. Ruderman reports consultant relationships with Pfizer. Strand reports consulting fees from AbbVie, Amgen, AstraZeneca, Bristol-Myers Squibb, Celgene, Genentech, GlaxoSmithKline, Janssen, Eli Lilly, Novartis, Pfizer, Regeneron, Sanofi and UCB, as well as advisory board involvement with AbbVie, Amgen, AstraZeneca, Bristol-Myers Squibb, Celgene, Genentech, GlaxoSmithKline, Janssen, Eli Lilly, Novartis, Pfizer, Regeneron, Sanofi and UCB.