Update on Axial Spondyloarthritis

Abstract

Understanding of the spondyloarthritis diseases has changed significantly in the last 15 years. It is now clear that there are patients with and without radiographic changes and the terminology has changed to reflect that: radiographic axial spondyloarthritis and non-radiographic axial spondyloarthritis. In addition, the importance of the presence of inflammatory back pain with spondyloarthritis in making the diagnosis is now well established. It is also clear that women are much more likely to develop axial spondyloarthritis than previously thought. Finally, there are treatments now available to treat axial spondyloarthritis and more hopefully to be approved in the next year.

Introduction

Better and more thorough information about the incidence, prevalence and natural course of the disease ankylosing spondylitis (AS) is now known and has changed our understanding of this disease. We now know that women are much more likely to develop this disease compared to what was initially understood. The development of MRIs has also led to the recognition of earlier disease in patients showing normal plain x-rays but abnormal bone marrow edema in the sacroiliac joints. With this there is now interest in renaming ankylosing spondylitis to reflect these changes – to radiographic axial spondyloarthritis (rAxSpA) for patients with plain radiographic changes in the sacroiliac joints and with those patients with normal plain x-rays now being designated as non-radiographic axial spondyloarthritis (nrAxSpA). And, although not universally accepted it is generally thought that patients with non-radiographic axial spondyloarthritis are part of the spectrum of disease with AS (i.e., rAxSpA).

History

Ankylosing spondylitis was first described from skeletal remains back in the late 1600s by Dr. Connor, an Irish physician.¹ In the 1890s at about the same time, three physicians, Dr. Bechterew (Russian), Dr. Marie (French), and Dr. Strumpell (German), attributed the signs of what we call AS to inflammation and considered it to be a rheumatologic condition^2–4 and in some medical literature, one will see that AS has also been called Bechterew’s disease or Marie-Strumpell disease. The term ankylosing spondylitis was not coined though until 1904 by Dr. Frankel.⁵ In the last 10 to 15 years with the use of MRI showing bone marrow edema (BME) in SI joints and with other typical features of AS in patients who do not have plain x-ray evidence of AS, it has been thought that there is a spectrum of disease ranging from nrAxSpA to frank rAxSpA. The term AS has been thought to be not as descriptive with the understanding that there are patients with normal plain radiographs of the sacroiliac region, but similar clinical features and evidence of bone marrow edema on MRI in the SI region. These patients have been called non-radiographic AxSpA, and with this, there has been a trend toward renaming AS to rAxSpA.

Table 2.

AxSpA Features

Family history of ankylosing spondylitis Heel pain Uveitis Psoriasis Inflammatory bowel disease Peripheral synovitis Response to NSAIDs Elevated CRP/ESR

Incidence and Prevalence

Newer population studies have found that the incidence prevalence is about 1.5 to 2 million patients.^6,7 To put that in perspective, rheumatoid arthritis is thought to occur in about 1.3 million⁸ and systemic lupus erythematosus in about 0.3 million.⁹ However, rheumatology offices are as yet not seeing these numbers of patients. This may be in part due to patients not recognizing their symptoms as well these patients not being referred to a rheumatologist or not being correctly diagnosed even after being seen in rheumatologic practices. There is good data to show that it is about 10 years from onset of symptoms to patients being diagnosed with AxSpA. Initially it was thought that AxSpA was really only seen in men with a ratio initially thought 10:1 male to female. It is now clear that the ratio is more likely 2:1 men to women for rAxSpA and 2:1 women to men with nrAxSpA.¹⁰

About 20% of U.S. adults have chronic low back pain, but only 5%⁷ will have Inflammatory Back Pain (IBP). Of those with IBP, somewhere between 14% to 25% will develop rAxSpA.¹¹ A recent retrospective study of about 150 patients with IBP were followed over 15 years. In this real-world study, about 25% of these patients developed rAxSpA and 50% had resolution of their symptoms. The other 25% developed either nrAxSpA or other diseases. Of patients with either form of AxSpA, 60% are a little more likely to have nrAxSpA form while the other 40% will have rAxSpA. For those who are initially diagnosed with nrAxSpA, about 60% to 90% will not develop further radiographic changes but about 10–40% will develop frank SI changes and become rAxSpA. Those initially diagnosed with rAxSpA will more likely continue to progress radiographically (60% to 70%) with about 30% to 40% not showing any further progression.⁶

Clinical Characteristics

A classic and key symptom of AxSpA is inflammatory back pain (IBP). This differs from mechanical back pain¹¹ in the following ways: first the onset of back pain is before age of 40 and not associated with a history of trauma. In addition, pain at night or early morning is also common. Unlike mechanical back pain, IBP is worse with rest and improves with exercise. NSAIDs usually seem to help IBP pain. IBP patients often have complaints of alternating buttock pain (Table 1).

Table 1.

Inflammatory Back Pain versus Mechanical Back Pain

Inflammatory Back Pain	Mechanical Back Pain
Onset prior to age 40 No history of trauma Pain better with exercise Pain worse with rest Pain better with NSAIDs Alternating buttock pain	Onset at any age Often history of trauma Pain usually worse with exercise Pain usually not better with NSAIDs Pain usually improved with rest

What are called spondylitis (SpA) features (Figure 2) help to identify patients with AxSpA.^12–14 These SpA features do occur at the same incidence in both rAxSpA and nrAxSpA.¹⁴ One of the most common SpA features is enthesitis. This is inflammation in the tendon areas near the joint and often occur in the heel and Achilles’ tendons but also can occur in the knees, shoulders, hips and chest wall. Enthesitis is very commonly seen in AxSpA and is seen in about 30% to 50% of patients. Dactylitis which refers to a swelling of the whole digit (either toe or finger) is less common in AxSpA. In AxSpA, it occurs at about 6% while in peripheral spondyloarthropathies, such as psoriatic arthritis, it is seen about 30% to 50%.¹¹ Other SpA features include uveitis, which occurs in about 10% of patients with AxSpA as well as peripheral arthritis which is often asymmetric and involves joints such as the knees and hands in about 30% of patients.¹² Family history of AxSpA or a history of early onset of arthritis in family member occurs about 50%. Inflammatory bowel disease is also a SpA feature and occurs in only 5% to 10%. However, there is evidence of microscopic colitis (noted by random colon biopsies) in about 30% to 60%. And finally, psoriasis may be present in as many as 10% of AxSpA patients.¹²

Pathophysiology

There appears to be both environmental as well as genetic factors involved in the development of AxSpA.^15,16 Environmental factors include smoking as well as infections (possibly mycoplasma) and the gut microbiome. There are clearly genetic factors involved, and although the most common gene is the HLA B27 gene, there are at least 50 other genes associated with AxSpA including ERAP1. It is important to stress that though having the HLA B27 gene does increase the risk of developing AxSpA, its presence is NOT diagnostic for AxSpA since it is seen in patients who never develop AxSpA and is sometimes absent in patients with AxSpA. Of those with the HLA B27 gene, only about 6% will develop IBP. However, if one identical twin has AxSpA with the HLA B27 gene, the other twin has about 65% chance of developing AxSpA. First-degree relatives of AxSpA patients with the HLA B27 gene have about 20% chance of also developing AxSpA. In addition, the incidence of B27 varies with ethnicity: about 6% to 7% in Whites, 4% in Mexican Americans and about 1% to 3% in Blacks. As for indigenous populations in the U.S., HLA B27 is more prevalent with an incidence of 9% in Hopi, 11% in Navajo, and about 35% in Haida.

Biomechanical trauma is likely one of the environmental factors involved in the initiation of AxSpA and seems to be due to release of the cytokine interleukin (IL) 23 in enthesial areas which appears to lead to the develop of chronic inflammation. However, IL23 does not appear to be as important in inhibiting disease compared to the cytokines IL17 and Tumor Necrosis Factor inhibitors (TNFi). Different than other autoimmune diseases, we now know that innate as well has adaptive immunity plays a role in the pathogenesis of AxSpA. The immune cells that seem especially important in ongoing chronic inflammation in AxSpA include the Th17, cd8 negative, and Natural Killer (NK) cells. What is still unclear in the pathology of AxSpA is the role HLA B27 plays. There are several hypotheses, including HLA B27 contributing to misfolding of proteins, HLA B27 acting as arthritogenic peptide itself, HLA B27 impacting the intestinal microbiome so there are more antigenic bacteria, and B2 microglobin which is part of the HLA B27 heterotrimer separating and depositing in synovial tissues causing destruction.^17–19

Diagnosis

Patients with axial spondyloarthropathies often do not get appropriately diagnosed for average of 10 years in men and even longer in women.²⁰ There are likely several reasons for the delayed diagnosis. It may be in part due to the difficulty in recognizing the disease since no clear-cut diagnostic markers have been identified. Although 80% to 90% of patients with rAxSpA and 70% of nr AxSpA are positive for HLA B27, about 6% of the normal White population also have the HLA B 27 gene. As previously mentioned, of those who are HLA B 27 positive, only about 6% to 10% develop inflammatory back pain, and approximately 10% to 30% of AxSpA patients will be negative for the HLA B27 gene. Standard measures of inflammation (ESR and CRP) are only elevated in 50% to 60% of patients.

The key feature in recognizing AxSpA is inflammatory back pain. As already mentioned, the classic features of inflammatory back pain that differentiates from mechanical back pain are as follows: onset of back pain without trauma and prior to age 45; back pain improves with exercise and worsens with rest, alternating buttock pain and improvement with NSAIDs (Figure 1). It is important to ask about the features of IBP to separate those patients with IBP and thus possible AxSpA from those with mechanical back pain and therefore not likely candidates to have AxSpA.

Recognizing the typical AxSpA features can aid in recognizing patients with AxSpA. The typical SpA features include^12,13 peripheral synovitis often asymmetric, heel pain (or plantar fasciitis), uveitis, psoriasis, family history, and inflammatory bowel disease. Response to NSAIDs also is seen in patients with inflammatory back pain. An assessment of pre- and post-test probability in patients with IBP showed the presence of family history alone increases pre-test probability from 6% to 51%. The presence of heel pain and response to NSAIDs also increase probability of diagnosis.²¹ The presence of positive MRI and positive HLA B27 increases post-test probability in patients with IBP to 91%. An algorithm for diagnosis has been published¹⁶ which emphasizes the importance of the SpA features in helping to narrow the diagnosis to AxSpA versus not AxSpA. If four or more SpA Features are present, then there is a high likelihood of AxSpA. If 0 to 1 are present, then AxSpA is unlikely. When two or three SpA features are present, then it is reasonable to order MRI of the pelvis and HLA B27 gene test to help direct toward or away from the diagnosis of AxSpA.

Gender Differences

Women have long been thought not to have AxSpA, but it is now clear that the original estimate of 10:1 ratio was incorrect.²² It is now recognized that females do get AxSpA—both rAxSpA and nrAxSpA, and the ratio men to women is about 2:1 in rAxSpA and 2:1 women to men in nrAxSpA. Diagnosis of AxSpA is generally delayed in both men and women, but usually there is a much longer delay in diagnosis of women with AxSpA. This is in part due to the failure of recognizing IBP in women (and men), but also the prevalence of misdiagnosis of women with other conditions—usually with fibromyalgia or depression. Women with AxSpA are more likely to have normal inflammatory markers and less likely to have response to TNF inhibitors. They are also more likely to have enthesitis and increased disease severity, peripheral joint involvement, and to present with neck and upper back pain.^23–25

Differential Diagnosis

Diseases that may be confused with AxSpA include peripheral spondyloarthritic diseases such as psoriatic arthritis and diffuse idiopathic skeletal hyperosteosis. There are other health issues that can cause abnormal bone marrow edema (BME) in sacroiliac joints. Studies have shown BME in athletes, particularly runners and hockey players, and in women postpartum. And of course, there have been many patients, particularly women, who were diagnosed with fibromyalgia but actually had AxSpA. One famous example of misdiagnosis of AS were two Egyptian pharaohs—Ramses 11 and Amenhotep. For a number of years, archeologists thought that these pharaohs had AS based on thoracic spine x-rays of their mummies. However, further x-ray data showed no evidence of SI changes and now it is thought that these pharaohs have either DISH or psoriatic arthritis with axial involvement.

Comorbities

Like other inflammatory diseases, patients with AxSpA have increased risk of developing other conditions. AxSpA patients are four times more likely to have cardiovascular disease—including hypertension and coronary disease.²⁶ These patients are at very high risk of osteoporosis—about 25% will have osteoporosis and 40% with osteopenia within 10 years of disease. As a result, they are at high risk of vertebral fractures and approximately 10% will experience this type of fracture.²⁷ Fibromyalgia is seen in about 20% of patients (males and females) with AxSpA.²⁸ Depression is also not uncommon in both men and women with AxSpA.²⁹

Treatment

Guidelines for treatment of AxSpA have been developed by organizations including ACR as well as SPARTAN.³⁰ NSAIDs are used first line and have been shown to be helpful for pain as well as possibly associated with slowing radiographic progression.³⁰ Conventional immunosuppressives such as sulfasalazine and methotrexate are NOT helpful with axial disease but might be useful with the peripheral arthritis.³⁰ Systemic steroids are also not indicated, but local injections may be beneficial as long as not done in areas where there is high risk for rupture of the tendon (such as the Achilles tendon). TNFi and IL17 inhibitors are now approved for both radiographic and non-radiographic AxSpA.³⁰ Results have shown that approximately 60% of patients on TNF inhibitors get at least 20% improvement in their symptoms of AxSpA and 40% of these patients with AxSpA obtain 40% improvement. Cohort studies have shown less radiographic progression in patients on TNFi than those only on NSAIDs; however no formal study has been done to confirm this.

Studies with IL17 drugs have shown similar results. In terms of data, the trials have shown resolution or improvement of MRI bone marrow edema. Not yet approved but expected to be so in the next 12 months is the use of JAK inhibitors in radiographic AxSpA which in clinical trials have some similar response rates to IL17 drugs. The guidelines recommend that if a patient does not respond to one agent—primary failure—to switch to another agent with a different mechanism of action. If a patient is having a secondary failure—initially responding, then losing response over time—it is not unreasonable to stay with agents with the same mechanism of action. If there is no response, it may be reasonable to reassess by bone scan or MRI to see if there is still active inflammation present.

Conclusion

The understanding and approach to recognizing and diagnosis of the AxSpA diseases has dramatically changed in the last 15 years. Understanding the presence of BME in MRIs has helped to allow recognition of patients with AxSpA without plain x-ray findings. The pathophysiology of AxSpA is now better understood which has led to the development of agents effective in treating these patients although there are still patients with limited response to agents presently available. Unfortunately, we are missing or misdiagnosing patients, and finding better diagnostic tools is important. In closing, more and better agents are still needed for those AxSpA patients with limited or no response to the agents we presently have available.