Timely- versus Delayed-Use of TNFi’s: Which Approach Is Better?
Interview with Elaine Husni, MD, MPH
Given the serious damage that psoriatic arthritis (PsA) can cause if inadequately treated, choosing the right treatment approach is a significant decision. Slowing radiographic progression of PsA is a primary concern.
According to a new study presented at the annual meeting of the American College of Rheumatology (ACR), held in November, 2015, in San Francisco, California, beginning treatment with a tumor necrosis factor inhibitor (TNFi) right away appears to be more effective when compared with apremilast (Ortezla), a relatively new oral phosphodiesterase-4 inhibitor (PDE4i), for patients with severe disease.
However, gauging PsA severity can be tricky. There is no official biomarker or measure that informs practitioners whether the diagnosis is mild enough to warrant an initial treatment approach with apremilast. While it is important to make sure the disease is effectively inhibited, the treatment should avoid pharmaceutical overkill—the “kill a fly with a shotgun” effect.
“One of the big questions is ‘Do I do any harm by not providing top of the line treatment for somebody over time, versus starting slow and going up?’” This can be a real balancing act, said Elaine Husni, MD, MPH, of the Cleveland Clinic Foundation in Ohio. “So the best thing to do is look at patients that are treated both ways and evaluate the outcome to see which way is better.” For example, timely use with a TNFi or starting off treatment with apremilast.
The study evaluated patients with moderate to severe PsA (≥3 swollen joints and ≥3 tender joints) treated with TNFi’s (specifically adalimumab, etanercept, infliximab, or golimumab) and/or apremilast. Certolizumab was excluded, noted the investigators.1
After assessing the 1-year outcomes, the timely-use group achieved noticeably higher rates of ACR20 responses compared to the delayed-use group (70.4% versus 59.6%, respectively). The subpopulation of PsA patients with comorbid psoriasis showed similar results, as combined joint and skin response rates were noticeably higher for the timely-use group compared to the delayed-use group, (41.0% versus 30.0%, respectively).
The mean time spent with these ACR responses and skin responses (measured by psoriasis area and severity index score [PASI75]) also was noticeably longer for the timely-use patients compared to delayed-use patients, (0.60 and 0.35 years versus 0.48 and 0.24 years).
Be Pragmatic: Switch Medications
While the study operated on a 1-year time frame, patients were evaluated as early as 12 weeks to see whether they were achieving ACR20 scores while on the TNFi or apremilast. If patients were not improving on the TNFi, they were switched to a different TNFi. If patients were not improving on apremilast, they were switched to a TNFi.
According to Dr. Husni, this would seem like an appropriate strategy during the early phases of feeling out a treatment. Around 10 to 16 weeks is the “sweet spot” for assessing how a patient responds to the drug. Doing this too late can mean dire consequences because the PsA can cause irreversible damage, she noted. However, cycling medications too early can be counter-intuitive, as there needs to be enough time for the drug to build up in the system.
The study also calculated the number needed to treat (NNT) for timely- and delayed-use approaches, which showed similar findings as the other data sets, where TNFi drugs outperformed apremilast (1.42 versus 1.68, respectively). Secondary analysis with the subpopulation of patients with baseline co-occurring psoriasis showed a similar pattern of timely- beating out delayed-use (2.44 versus 3.33, respectively).
Apremilast: Favorable for Mild PsA
Apremilast may be a more favorable medication for milder forms of PsA, however. A Phase III trial examined apremilast’s safety profile through 156 weeks of continued use.2 The drug was deemed safe, but there are precautions to consider. Common side effects like gastrointestinal complications, nausea, headaches, and respiratory tract infections do exist. Weight loss (5%-10%) is also a fairly common concern. Rare cases of depression have been observed in trials as well, which means practitioners may want to re-evaluate using the therapy on a patient who has a history of depression and/or suicidal thoughts and behaviors. Doctors also should avoid mixing the drug with CYP450 enzyme inducers.
Overall, recent meta-analysis found that in NNT, apremilast stands at the back of the pack of agents. One study presented at this year’s ACR meeting detailed a systematic literature review of Phase III randomized, controlled clinical trials that directly compared NNT data amongst the 3 classes of arthritis drugs, including TNFi’s (adalimumab, certolizumab, etanercept, golimumab, and infliximab), PDE4’s (apremilast), and interleukin inhibitors (ustekinumab and secukinumab).3
Using a Bayesian network meta-analysis, the study indirectly compared the drugs’ efficacy for treating active PsA, evaluating ACR20 and PASI75 response rates at week 24. The drugs were split into various groups based on their FDA-approved dosages (eg. 75, 90, 150, 300 mg).
The resulting data presented an interesting hierarchy, where TNFi’s (adalimumab, golimumab, infliximab) and secukinumab; 300 mg—had the lowest, most favorable NNT levels per ACR20 and/or PASI75 responders, which remained true after reanalysis using biologic-naïve patients. Apremilast, by contrast, had the highest NNT per ACR20 responder, closest only to ustekinumab 45 mg. NNT rates per PASI 75 responder told a different story, however. While apremilast’s NNT was still far higher than most of the other TNFi’s, it was identical to patients taking etanercept, suggesting a comparable efficacy in skin symptom management.