LONDON – Patients with ankylosing spondylitis who remained on long-term treatment with a nonsteroidal anti-inflammatory drug and a tumor necrosis factor inhibitor had significantly less new spinal-bone formation in a cross-sectional analysis of a multicenter cohort of 527 U.S. patients.
A related analysis of the same cohort also showed significantly less ankylosing spondylitis (AS) progression as measured by radiographic progression among patients who received treatment with a tumor necrosis factor–alpha (TNF) inhibitor for 2.1-3.5 years regardless of their treatment with a nonsteroidal anti-inflammatory drug (NSAID), although this link trended to a stronger effect among the patients taking both, Lianne S. Gensler, MD, reported in a pair of posters at the European Congress of Rheumatology.
Dr. Lianne S. Gensler
These finding suggest “there may be synergy between NSAIDs and TNF inhibitors [for slowing or preventing progression] in a select group of AS patients at high risk for progression,” said Dr. Gensler, a rheumatologist and director of the Ankylosing Spondylitis Clinic at the University of California, San Francisco.
But Dr. Gensler also cautioned that these findings are merely “hypothesis generating” and should not be used as a rationale to place or maintain AS patients on long-term treatment with an NSAID, a TNF inhibitor, or both drugs.
“You treat the disease burden. The message is absolutely not to always put AS patients on both types of drugs. When an AS patient is well controlled on a TNF inhibitor alone, I would not tell them to also take a NSAID,” she said in an interview. “This is only relevant for patients who require treatment with both drug classes because of their clinical status.”
As the list of treatment options for patients with AS grows – it now includes NSAIDs, TNF inhibitors, and the interleukin-17 inhibitor secukinumab (Cosentyx) – the impact of these agents on disease progression as assessed by radiography and new bone formation becomes a new dimension to start to consider in addition to the standard criterion of immediate clinical response, Dr. Gensler explained. AS progression “is another factor to think about as we decide on treatment strategies. There is growing evidence that long-term treatment with a tumor necrosis factor inhibitor and with a NSAID have potential roles in disease modification.” But the evidence is indirect, from cohort studies that make cause and effect assessments difficult because of possible unidentified confounding factors, she noted.
“There has never been a randomized, controlled trial examining the disease-modifying effects of a TNF inhibitor because you can’t keep patients on placebo for a long enough time to see this benefit,” Dr. Gensler said. It takes a long time to see progression in AS patients, she noted.
A prior cohort analysis run by Dr. Gensler and her associates found evidence for an effect of long-term treatment with a TNF inhibitor and reduced AS progression measured using the modified Stoke AS Spine Score (mSASSS), compared with AS patients not on a TNF inhibitor in a propensity-score matched analysis of 334 patients. A link between TNF inhibitor use and a discernible difference in mSASSS only occurred when patients were on TNF inhibitor treatment for at least 3.9 years (Arthritis Rheum. 2013 Oct;65[10]:2645-54). In addition, a separate report at the EULAR congress on 168 AS patients maintained on treatment with secukinumab for 2 years showed evidence for slowed progression of mSASSS scores, compared with historical controls as well as with similar patients who were not on secukinumab treatment for as long a period of time.
The new analysis reported by Dr. Gensler looked at 527 AS patients in the multicenter Prospective Study of Outcomes in AS cohort followed for a median of 3.7 years. Clinicians participating in this cohort saw patients every 6 months, and radiographic assessments by mSASSS and for new bone formation occurred every 2 years. At entry into the registry, about 57% of patients received a TNF inhibitor and about 63% received an NSAID, with a third on an NSAID only, 27% on a TNF inhibitor only, 30% on both drugs, and 10% receiving neither drug.
The analysis showed that among patients followed for 2.1-3.5 years, the fraction of patients on a TNF inhibitor who showed progression of their mSASSS was 77% lower than patients not on a TNF inhibitor, a statistically significant difference, Dr. Gensler reported. The researchers saw no statistically significant difference in mSASSS progression rates between the patients on a TNF inhibitor at baseline and those not on a TNF inhibitor at baseline among patients followed for 2 years and among those followed for more than 3.5 years, although the analysis did show nominally higher levels of response among TNF-inhibitor users followed for more than 3.5 years. Dr. Gensler speculated that one reason for the loss of a statistically significant difference during longer follow-up could be that fewer patients reached these levels of prolonged follow-up, making statistically significant differences harder to see. This analysis also showed a strong trend for less progression among the patients treated with an NSAID, a 51% relative reduction in mSASSS progression, compared with patients not taking an NSAID, but this relationship just missed statistical significance.