Most of the minority of axial spondyloarthritis patients who do not respond to their first tumor necrosis factor inhibitor still meet classification criteria for the condition 5-10 years later, but more than half respond to a second TNFi and most also have comorbidities that could affect the spondyloarthritis evaluation, according to findings from a single-center cohort study.
The study’s finding of TNFi primary inefficacy in 27 (12%) of 222 patients with axial spondyloarthritis (axSpA) who were given their first TNFi during 2004-2009 is in line with previous results of about 5%-15% primary inefficacy for a first TNFi in axSpA. However, the French investigators, led by Sandra Kossi of Cochin Hospital in Paris, noted that the difficulty of making an axSpA diagnosis and the presence of certain comorbidities “could interfere either with the activity of SpA (falsely heightened disease activity) or with the response to TNFi (falsely heightened inefficacy).” They sought to report the characteristics of axSpA patients with primary inefficacy after their first TNFi in the 5- to 10-year period after their prescription (Rheumatology [Oxford]. 2017 Jan 9. doi: 10.1093/rheumatology/kew456).
A total of 25 of the patients with primary inefficacy underwent re-evaluation 5-10 years (mean of 6 years) later. The investigators defined primary inefficacy as “treatment interruption 3-4 months after treatment onset, with a rheumatologist assessment in the medical file citing lack of efficacy as primary reason for drug interruption.”
The patients with primary TNFi inefficacy had a mean age of 53 years at the time of follow-up, and about half were female. All the patients still had symptoms and back pain, but symptoms were moderate based on a mean Bath Ankylosing Spondylitis Disease Activity Index score of 42. Overall, nine were taking a TNFi and nine were taking nonsteroidal anti-inflammatory drugs, while others used analgesics or nonpharmacologic measures.
Primary TNFi inefficacy occurred significantly more often occurred among females (48% vs. 27%; P = .04), older aged patients at first TNFi use (45 vs. 39 years; P = .04), patients with higher mean Bath Ankylosing Spondylitis Functional Index scores (68 vs. 42; P = .03), and in those who did not have an abnormally elevated C-reactive protein level (33% vs. 63%; P = .02). Patients with primary TNFi inefficacy had lower rates of HLAB27 positivity (56% vs. 72%) or radiographic sacroiliitis (67% vs. 81%), but the differences were not statistically significant.
At follow-up, 21 (84%) patients met Assessment of Spondyloarthritis International Society classification criteria for axSpA, and 20 (80%) met the axSpA diagnosis according to the rheumatologists’ opinion, with 17 (68%) fulfilling both.
A total of 18 (72%) had at least one of the following comorbidities: widespread pain syndrome, osteoarthritis, or depression. Five patients – all females – had widespread pain syndrome according to the Fibromyalgia Rapid Screening Tool questionnaire. Another 10 patients had osteoarthritis of lower-limb peripheral joints or of the spine, while an additional 8 had self-declared depression, for which 3 were taking antidepressants.
By the time of follow-up, 16 (64%) had switched to another TNFi, including 9 who had received two TNFi drugs and 7 who had received three or more. The second TNFi was considered efficacious in nine patients. The retention rate of the second TNFi at 1 year among those with primary inefficacy was 50%, and overall, nine patients were still prescribed a TNFi at the time of follow-up.
The fact that most of the patients with primary inefficacy to their first TNFi had confirmed axSpA but also had comorbidities that could affect axSpA evaluation “is important because practitioners might consider that primary inefficacy to TNFi leads to reconsidering the diagnosis of SpA (i.e. the notion of a TNFi prescription being used as the diagnostic test). We suggest here that primary inefficacy should not be considered as equivalent to a diagnostic error, and that a second prescription of TNFi may be of use in such patients, although painful comorbidities should certainly be screened for and taken into account.”
The study was funded by the Assistance Publique des Hôpitaux de Paris. The investigators had no conflicts of interest to declare.