Are Older TNF Inhibitors More Effective at Treating Psoriatic Arthritis Than Newer Drugs?

A meta-study of randomized controlled trials indicates that older TNF-inhibitors such as etanercept (Embrel) or infliximab (Remicade) treat psoriatic arthritis more effectively than 3 newer TNF inhibitors: apremilast (Otezla), ustekinumab (Stelara) and certolizumab (Cimizia).

A research team from the Mayo Clinic analyzed data from 12 trials that measured efficacy of medications that inhibit tumor necrosis factor (TNF) in psoriatic arthritis patients who responded inadequately to (or were unable to tolerate the effects of) disease-modifying antirheumatic drugs (DMARDs) or nonsteroidal anti-inflammatory drugs (NSAIDs).

The trials did not compare different medications against each other. Instead, they each compared a single medication against placebo. The researchers were nevertheless able to compare the efficacy of various medications indirectly, because the endpoint of every trial was the same: a standard known as an American College of Rheumatology 20 (ACR20) that represents a 20% improvement from baseline in the number of tender and swollen joints and at least 3 other important disease measurements.

After tabulating results for 1,989 study group patients and 1,175 placebo group patients, the researchers calculated the patients who took the older medications were more than twice as likely to achieve ACR20 than users of any of the newer medication. Indeed, the likelihood of ACR20 outcomes for the older medications was 3.36 times that of apremilast 20 mg (95% confidence interval [CI], 2.10-5.38), 2.42 times that of apremilast 30 mg (95% CI, 1.55-3.77), 2.38 times that of ustekinumab 45 mg (95% CI, 1.68-3.35), 2.08 times that of ustekinumab 90 mg (95% CI, 1.48-2.93) and 2.20 times that of certolizumab (95% CI, 1.48-3.26).

The researchers then compared outcomes among patients who took the older TNF-inhibitors to those who took differing doses of secukinumab (Cosentyx), a new medication that is an interleukin-17A inhibitor rather than a TNF inhibitor. The older TNF-inhibitors fared better than any dose of secukinumab (which has yet to be approved in the US for the treatment of psoriatic arthritis) but not significantly so. Users of the older TNF-inhibitors were 1.90 times as likely to achieve ACR20 as secukinumab 75 users (95% CI, 0.95-3.78), 1.10 times as likely to achieve it as secukinumab 150 mg users (95% CI, 0.58-2.09) and 1.21 times as likely to achieve it as secukinumab 300 mg users (95% CI, 0.63-2.29).

"Our meta-analysis aimed to answer a commonly encountered question in clinical practice: What would be the first biological agent of choice after the patients fail or do not tolerate synthetic DMARDs/NSAIDs?” the study authors wrote in Seminars in Arthritis and Rheumatism. “We were able to demonstrate that patients who received older TNF-inhibitors and secukinumab at the dose of 150 mg and 300 mg weekly were more likely to achieve an ACR20 response at week 12 to 24 compared with those who received apremilast, certolizumab or ustekinumab.”

Sensitivity analysis suggested that the findings of greater efficacy among older TNF-inhibitors, when compared as a whole to newer ones, are statistically robust (p<0.001 for all comparisons). That said, the study authors noted a number of limitations in their work, including the inherent inferiority of using indirect comparisons instead of head-to-head randomized trials. They also struggled to explain their findings by suggesting any reason why older TNF-inhibitors would be consistently superior to newer ones.