EFFECT OF COMBINATION THERAPY ON VISIBLE/NON-VISIBLE SYMPTOMS, AND DISEASE BURDEN ASSOCIATED WITH SEVERE ROSACEA: RESULTS FROM A POST-HOC ANALYSIS OF A RANDOMIZED CONTROLLED TRIAL James Del Rosso, DO1, Mark Jackson, MD2, Sandra Johnson, MD3, Alison Harvey, PhD, MS4, Rajeev Chavda, MD5 1JDR Dermatology Research/Thomas Dermatology, Las Vegas, NV; 2University of Louisville, Louisville, KY; 3Johnson Dermatology Clinic, Fort Smith, AR; 4Galderma Laboratories, Fort Worth, TX; 5Galderma S.A., Tour-de-Peilz, Switzerland SYNOPSIS • Rosacea affects over 16 million people in the US with symptoms in central areas of the face, causing flushing, stinging/burning, chronic erythema, and inflammatory lesions, with major negative impact on quality of life.1-3 • The disease is chronic and inflammatory in nature and burden typically extends beyond visible symptoms with impact on emotional, social, and psychological aspects of life.4 • In a global survey (N=710) evaluating impact of symptoms associated with disease burden in rosacea, non-visible symptoms such as itching, burning/pain, and swelling were significantly associated with high disease burden.4 • Combination therapy with multiple mechanisms of actions are often used to manage symptoms associated with rosacea. However, limited numbers of controlled studies have evaluated impact of combination therapy vs monotherapy on reducing the signs and symptoms associated with rosacea.5 • Ivermectin and doxycycline are two well-established molecules with proven efficacy and safety, targeting different and complementary pathological features of severe rosacea.5 CONCLUSIONS Combined IVM and DMR improved both visible (eg, inflammatory lesions, erythema, flushing) and non-visible symptoms (eg, stinging/burning) of rosacea. Correlations between symptom reduction and DLQI suggest improvements in different aspects of quality of life. Results from this Phase 4 study and this post-hoc analysis emphasize the importance of targeting both the visible and nonvisible signs and symptoms in patients with severe rosacea with combination therapy to create best outcomes. Acknowledgements This research was sponsored by Galderma Laboratories, Fort Worth, TX; medical writing support was provided by Medicalwriters.com, New York, USA. References 1. Steinhoff M, Schmelz M, Schauber J. Facial erythema of rosacea - aetiology, different pathophysiologies and treatment options. Acta Derm Venereol. 2016;96(5):579–586. 2. Schaller M, Almeida LMC, Bewley A, et al. Recommendations for rosacea diagnosis, classification and management: Update from the global ROSacea COnsensus (ROSCO) 2019 panel. Br J Dermatol. August 2019. [Epub ahead of print] 3. National Rosacea Society Website. https://www.rosacea.org/rosacea-review/2010/winter/rosacea-now-estimated-to-affect-at-least-16-million-americans. Accessed November 21, 2019. 4. Tan J, Steinhoff M, Bewley A, Gieler U, Rives V. Characterizing high-burden rosacea subjects: a multivariate risk factor analysis from a global survey. J Dermatolog Treat. June 2019. [Epub ahead of print] 5. Schaller M, Kemeny L, Havlickova B, et al. A randomized phase 3b/4 study to evaluate concomitant use of topical ivermectin 1% cream and doxycycline 40 mg modified-release capsules versus topical ivermectin 1% cream and placebo in the treatment of severe rosacea. J Am Acad Dermatol. May 2019. [Epub ahead of print] OBJECTIVE The objective of this post-hoc analysis was to extend beyond the Phase 4 study looking at the efficacy and safety of ivermectin 1% cream (IVM) and doxycycline 40 mg modified release (DMR) in severe rosacea subjects, and highlight the impact on the visible and nonvisible symptoms associated with rosacea. The relationship between the following parameters was assessed: • Impact of changes in the visible and nonvisible symptoms of rosacea on Dermatology Life Quality Index (DLQI) • Impact of baseline lesion count, prior rosacea treatment, and disease duration on efficacy of combination therapy METHODS The study was a 12-week, multicenter, randomized, investigator-blinded, parallel-group comparative study in 273 adults with severe rosacea, as previously described (ClinicalTrials.gov number: NCT03075891).4 Post-hoc analyses assessed the correlations between change in visible symptoms, stinging/burning, flushing severity, and Dermatology Life Quality Index (DLQI) and impact on disease burden using the Spearman's rank correlation coefficient, Mann–Whitney U test, Chi-square test, or Kruskal–Wallis test. Analyses were performed on the combined treatment arms. RESULTS • Impact of stinging/burning and ‘clear’ vs ‘almost clear’ on DLQI • A significant correlation was seen between change in stinging/burning over 12 weeks and change in DLQI score over 12 weeks (0.325; P =.000). [Figures 1a–b] • Significant correlations between the reduction in stinging/burning and individual DLQI parameters were found for itchy, painful skin (0.338; P =.000), problems with partner/friends (0.247; P =.000), feeling embarrassed (0.243; P=.000), and social activities (0.237; P =.000). [Figure 1c] • Significant correlations between ‘clear’ or ‘almost clear’ and individual DLQI parameters were found for feeling embarrassed (0.253; P =.000), interference with shopping/home garden (-0.161; P =.012), and social activities (-0.143; P=.026). [Figure 1d] • Severity of flushing and impact on DLQI • There was a significant correlation between change in flushing severity over 12 weeks and DLQI change over 12 weeks (0.222; P=.001). [Figure 2] • Impact of baseline lesion count, previous oral and/or topical treatment, and disease duration on efficacy of combination therapy • At week 12, there was a significant and strong correlation (-0.727; P=.000) between lesion counts at baseline and absolute change in lesion counts for the combined treatment arms. [Figures 3 a–c] • In the combination therapy arm, the highest significant percent reduction in lesion count was observed for previous oral plus topical treatment (P=.016). [Figure 4] • Disease duration for up to 9 years was significantly correlated with highest percent change in lesion counts from baseline to week 12 in the combination therapy arm (P=.006). [Figure 5] Impact of Reaching ‘Clear ’ and ‘Almost Clear ’ on DLQI Figure 1: d) Spearmanʼs correlation between individual DLQI parameters and completely/almost clear by IGA (the three parameters with the strongest correlations are shown) IGA score at 12 weeks (LOCF) -0.253 0.000 243 Correlation coefficient How embarrassed or self-conscious have you been because of your skin? P-value N -0.161 0.012 243 Correlation coefficientHow much has your skin interfered with you going shopping or looking after your home or garden? P-value N -0.143 0.026 243 Correlation coefficient How much has your skin affected any social or leisure activities? P-value N Analyses were performed on the combined treatment arms. Impact of Reduction in Stinging/Burning on DLQI Figure 1: a) Change in DLQI score over 12 weeks versus change in stinging/burning score over 12 weeks in treatment groups combined. b) Spearmanʼs correlation for the change in DLQI score versus change in stinging/burning score in treatment arms combined. c) Spearmanʼs correlation between individual DLQI parameters and stinging/burning (the four parameters with strongest correlations are shown) 10 0 -10 -20 -30 -3 -2 -1 0 1 2 235 211 199 17 189 188 Change in stinging/burning over 12 weeks (LOCF) Ch an ge in D LQ I s co re o ve r 1 2 w ee ks b) a) c) Change in DLQI score over 12 weeks 0.325 0.000 243 Correlation coefficient Change in stinging/burning class over 12 weeks (LOCF) P-value N Change in stinging/burning over 12 weeks (LOCF) 0.338 0.000 243 Correlation coefficient Change in How itchy, sore, painful, or stinging has your skin been? P-value N 0.247 0.000 243 Correlation coefficientChange in How much has your skin created problems with your partner or any of your close friends or relatives? P-value N 0.243 0.000 243 Correlation coefficient Change in How embarrassed or self-conscious have you been because of your skin? P-value N 0.237 0.000 243 Correlation coefficient Change in How much has your skin affected any social or leisure activities? P-value N Analyses were performed on the combined treatment arms. Impact of Reduction in Severity of Flushing and DLQI Figure 2: Spearmanʼs correlation of flushing severity with DLQI Change in DLQI score over 12 weeks 0.222 0.001 216 Correlation coefficient Change in flushing severity class over 12 weeks (LOCF) P-value N Impact of Prior Rosacea Treatment on Efficacy of Combination Therapy Impact of Disease Duration on Efficacy of Combination Therapy Figure 4: Association between oral versus topical previous treatment and percentage change in lesion count from baseline to week 12 investigated by Kruskal–Wallis test in the individual treatment arms and in the treatment arms combined Figure 5: Association between disease duration and percentage change in lesion count from baseline to week 12 investigated by Kruskal–Wallis test Percentage change in lesion count from baseline to week 12 (LOCF) Treatment 8.273 2 0.016 Chi-square IVM+DMR df Asymp. sig. Percentage change in lesion count from baseline to week 12 (LOCF) Treatment 10.266 2 0.006 Chi-square IVM+DMR df Asymp. sig. Percentage change in lesion count from baseline to week 12 (LOCF) Treatment Mean NDuration of rosacea Standard deviation IVM+DMR -75.60–4 years 46 22.4 -86.55–9 years 39 17.8 -79.8≥10 years 50 22.9 -80.3Total 135 21.7 Percentage change in lesion count from baseline to week 12 (LOCF) Treatment Mean NOral, topical or both Standard deviation IVM+DMR -75.9Oral 6 17.6 -70.2Topical 31 25.1 -84.8Oral+topical 18 23.7 -75.6Total 55 24.5 Impact of Baseline Lesion Count on Efficacy of Combination Therapy Figure 3: a) Box plot of number of lesions at baseline versus absolute change in lesion count from baseline to week 12 for the treatment arms combined. Association between number of lesions at baseline versus absolute change in lesion count from baseline to week 12 in the treatment arms combined examined by b) Kruskal–Wallis test and c) Spearmanʼs correlation 25 0 -50 -25 -75 20–29 30–39 40–49 50–69 131 113 238 146 222 154 102 196 167 156 137 169 32194 Number of lesions (papules+pustules) at baseline Ab so lu te ch an ge in le si on co un ts ov er 1 2 w ee ks (L O CF ) Treatment arms: IVM+DMR and IVM+placebo Change in lesion counts over 12 weeks (LOCF) 138.206 3 0.000 Chi-square df Asymp. sig. Number of lesions (papules+pustules) at baseline -0.727 0.000 273 Correlation coefficient P-value N Change in lesion counts over 12 weeks (LOCF) a) b) c) Fig 6. Subject Visual Symptom Improvement from Baseline to Week 12 IVM+DMR Therapy Resulted in Notable Reduction in Visible Symptoms Baseline IGA=4; IL=55; CEA=2 Week 12 IGA=1; IL=1; CEA=1 d) SOO.P-RD113322-14 Steve Midgley Line