Background/Purpose: Apremilast (APR), a phosphodiesterase
4 inhibitor, helps regulate the immune responses in psoriatic arthritis (PsA).
PALACE 1-3 compared APR efficacy/safety with placebo in patients with active
PsA despite prior conventional disease-modifying antirheumatic drugs (DMARDs)
and/or biologics. We report the safety of APR treatment during Weeks 0 to ≤104.

Methods: Patients were randomized (1:1:1) to placebo,
APR 30 mg BID (APR30), or APR 20 mg BID (APR20) stratified by baseline DMARD use
(yes/no). The placebo-controlled phase continued to Week 24, with an early
escape option at Week 16. Double-blind APR treatment continued to Week 52; patients
could continue to receive APR during an open-label, long-term treatment phase.

Results: 1,493 patients were
randomized and received ≥1 dose of study medication (placebo: n=495; APR30:
n=497; APR20: n=501). A total of 1,441 (1,209.3 patient-years) patients
received APR in Weeks 0 to ≤52, with 1,028 (907.7 patient-years)
continuing treatment during Weeks >52 to ≤104. Of the 1,441 patients
randomized, 698 (48.4%) had a full 2 years of exposure. During Weeks 0 to ≤52,
adverse events (AEs) occurring in ≥5% of APR-exposed patients were
diarrhea, nausea, headache, upper respiratory tract infection, and
nasopharyngitis (Table). Most diarrhea and nausea events were reported within
the first 2 weeks of treatment and usually resolved in 4 weeks without medical
intervention. Most AEs were mild or moderate in severity during the Weeks 0 to ≤104
APR-exposure period; no increase was seen in the incidence and severity of AEs
with longer term exposure. During Weeks >52 to ≤104, diarrhea (2.9%),
nausea (1.8%), and headache (3.0%) occurred at lower rates vs. Weeks 0 to
≤52 (Table). Reported serious AEs (SAEs)
with APR30 between 0 to ≤52 and >52 to ≤104 weeks were comparable at 6.5% and 6.3%,
respectively. Reported SAEs were higher with APR20 at 7.5% between >52 to ≤104 weeks
compared with 5.6% between 0 to ≤52 weeks with no cluster indicating any specific organ
involvement. Most SAEs were reported by 1
patient each. The cardiac, malignant neoplasm, opportunistic infection, or
psychiatric disorder related SAEs were comparable between 0 to ≤52 and 
52 to ≤104 weeks, and no cases of tuberculosis (new or reactivation) were
reported with either APR dose. Discontinuations due to AEs occurred at a lower
rate (2.3%) during Weeks >52 to ≤104 than during Weeks 0 to ≤52
(7.5%). Marked laboratory abnormalities were
infrequent, and most returned to baseline with continued treatment or were
associated with a concurrent medical condition.

Conclusion: APR demonstrated a favorable safety
profile and was well tolerated for up to 104 weeks. These
data continue to support the lack of a need for specific laboratory monitoring
with APR. No new safety concerns were identified with long-term exposure to
apremilast.