Novel Polymeric Tazarotene 0.045% Lotion for Moderate-to-Severe Acne: Pooled Phase 3 Analysis by Race/Ethnicity

July 2020 | Volume 19 | Issue 7 | Original Article | 727 | Copyright © July 2020


Published online June 22, 2020

Neal Bhatia MD,a Jonathan S. Weiss MD,b Neil Sadick MD,c Fran E. Cook-Bolden MD,d Stephen K. Tyring MD PhD,e Eric Guenin PharmD PhD MPH,f Anya Loncaric MS,g Susan Harris MSh

aTherapeutics Clinical Research, San Diego, CA bGeorgia Dermatology Partners, and Gwinnett Clinical Research Center, Inc., Snellville, GA cWeill Cornell Medical College, New York, NY; Sadick Dermatology, New York, NY dFran E. Cook-Bolden, MD, PLLC; Mount Sinai Hospital Center, New York, NY EUniversity of Texas Health Science Center, Houston, TX fOrtho Dermatologics,* Bridgewater, NJ gBausch Health US, LLC,* Petaluma, CA hBausch Health US, LLC,* Bridgewater, NJ *Bausch Health US, LLC is an affiliate of Bausch Health Companies Inc. Ortho Dermatologics is a division of Bausch Health US, LLC.

patients with darker skin tones, PIH is cited as the reason for seeking dermatological care.12

In these particular patients, using aggressive, non-irritating treatment is the recommended strategy to treat acne while reducing the risk of PIH and keloid scarring.10 The first line of treatment for mild-to-moderate acne in patients of color is topical retinoids6,10—these disrupt desquamation pathways and inhibit multiple inflammatory pathways, which is important for the prevention of secondary PIH.13

One challenge of traditional formulations of topical retinoids— such as tazarotene, adapalene, and tretinoin—has been achieving uniform and consistent application with low rates of irritation, which may preclude effectiveness in real-world settings.14 The development of tazarotene 0.045% lotion formulation utilizing polymeric emulsion technology improves topical delivery of a drug by suspending the active agent(s) with a polymer in a hydrating emulsion of solvents, emollients, and humectants without using surfactants. This new formulation can thereby reduce irritation and uniformly distribute these microscopic droplets across the skin surface in an aesthetically pleasing, easily spreadable lotion.15 The adaptation of this technology to topical retinoids has the potential to reduce skin irritation while maintaining efficacy at a lower dosage, which may have particular benefits for patients with darker skin tones.12,15

In a phase 2 trial in participants aged ≥12 years with moderateto- severe acne, tazarotene 0.045% lotion provided numerically lower lesion counts and higher rates of treatment success compared with tazarotene 0.1% cream.16 Promisingly, the incidence of treatment-emergent adverse events (TEAEs) was nearly 2-fold lower in the tazarotene 0.045% lotion group compared with the tazarotene 0.1% cream group.16 Furthermore, two identical phase 3 double-blind, randomized, vehicle-controlled 12-week clinical studies demonstrated tazarotene 0.045% lotion was efficacious versus vehicle and well tolerated in participants with moderate-to-severe acne.17 Pooled, post hoc analyses from these phase 3 studies were conducted to examine the potential effects of race and ethnicity on the efficacy and safety of tazarotene 0.045% lotion.

METHODS

Study Design and Participants
This pooled analysis includes data from NCT03168334 and NCT03168321, which were previously described.17 Both trials were identical, multicenter, double-blind, randomized, vehicle-controlled, parallel-group phase 3 studies conducted at 89 study centers in the United States and Canada. Eligible participants were aged ≥9 years with Evaluator’s Global Severity Score (EGSS) of 3 (moderate) or 4 (severe), and had facial acne inflammatory lesion counts between 20-50, facial acne noninflammatory lesion counts between 25-100, and ≤2 facial nodules. Participants were randomized (1:1) to tazarotene 0.045% lotion or vehicle lotion, applied to the face once daily for 12 weeks. Studies were conducted in accordance with the International Conference on Harmonization, the Declaration of Helsinki, Good Clinical Practice Guidelines, and local regulations. All participants or their legal guardians provided written informed consent. Studies were approved by relevant independent ethics committees or institutional review boards at each study site.

Study Assessments
Efficacy and safety assessments were performed at each study visit (baseline and at weeks 2, 4, 8, and 12). Blinded evaluators determined EGSS and measured the number of noninflammatory and inflammatory lesions as efficacy assessments. Treatment success was defined as the proportion of participants achieving ≥2-grade reduction from baseline in EGSS and a score of ‘clear’ (0) or ‘almost clear’ (1). Investigator-assessed cutaneous safety (scaling, erythema, hypopigmentation, hyperpigmentation) and participant-assessed tolerability (itching, burning, stinging) were evaluated using a 4-point scale where 0=none and 3=severe. Adverse events (AEs) and serious adverse events (SAEs) were also monitored throughout the study.

Statistical and Subgroup Analyses
The co-primary endpoints for the two phase 3 studies comprised absolute change from baseline to week 12 in mean inflammatory and noninflammatory lesion counts and the proportion of participants achieving treatment success at week 12. The intent-to-treat (ITT) population was defined as all participants who were randomized and received study drug. The safety population included all randomized participants who used study medication or vehicle at least once with a minimum of one post-baseline evaluation.

For this pooled post hoc analysis, data from a subset of participants were evaluated based upon self-reported race (white or black) and ethnicity (Hispanic or non-Hispanic). Race and ethnicity were not mutually exclusive. Least-squares (LS) mean percent changes from baseline in inflammatory and noninflammatory lesion counts at week 12 and treatment success at week 12 were analyzed for each subgroup.

An analysis of covariance (ANCOVA) was performed to test for superiority for lesion count data between treatment groups. Initial analyses of mean percent changes from baseline in noninflammatory and inflammatory lesion counts indicated significant skewness. To address this, a nonparametric method was used to rank transform data prior to performing ANCOVA, with factor of treatment and the respective baseline lesion count as a covariate. Logistic regressions (using Firth’s Penalized Likelihood) were performed to analyze treatment success, with