Safety and Efficacy of VP-102, a Proprietary, Drug-Device Combination Product Containing Cantharidin, 0.7% (w/v), in Children and Adults With Molluscum Contagiosum: Two Phase 3 Randomized Clinical Trials

Abstract

Importance: Molluscum contagiosum (MC) is a common viral skin infection that primarily affects children. Cantharidin, a topical vesicant, has a long history of use for MC in compounded formulations, but the safety and efficacy of doses, regimens, and application methods have not been demonstrated in large-scale trials.

Objective: To determine the safety and efficacy of VP-102, a drug-device combination containing cantharidin, 0.7% (w/v), compared with vehicle in individuals with MC.

Design, setting, and participants: Two phase 3, randomized, double-blind, vehicle-controlled trials of identical design (Cantharidin Application in Molluscum Patients [CAMP-1 and CAMP-2]) were conducted in 31 centers across the US. A total of 528 individuals aged 2 years or older with MC participated. CAMP-1 was conducted from March 21 to November 26, 2018, and CAMP-2 was conducted from February 14 to September 26, 2018.

Interventions: Participants were randomized (3:2) to topical application of VP-102 or vehicle to all treatable lesions every 21 days until complete lesion clearance or up to 4 treatments.

Main outcomes and measures: The primary efficacy outcome was the proportion of VP-102-treated participants achieving complete clearance of all MC lesions (baseline and new) compared with those who received the vehicle at the end-of-study visit on day 84. Intent-to-treat analysis was conducted for the efficacy population. Secondary efficacy outcomes included the proportion of participants achieving complete clearance of lesions at days 21, 42, and 63. Safety outcomes included assessment of adverse events, including expected local skin reactions.

Results: Of the 528 participants enrolled, 527 received treatment (CAMP-1, n = 265; CAMP-2, n = 262). A total of 267 of 527 participants (50.7%) were male; mean (SD) ages for CAMP-1 and CAMP-2 were 7.5 (5.3) years and 7.4 (8.0) years for the VP-102 groups and 6.3 (4.7) years and 7.3 (6.7) years for the vehicle groups. Treatment with VP-102 demonstrated superior efficacy to vehicle in the percentage of participants with complete clearance of MC lesions at the end of the study visit for CAMP-1 (VP-102: 46.3% vs vehicle: 17.9%; P < .001) and CAMP-2 (VP-102: 54.0% vs vehicle: 13.4%; P < .001). Adverse events were observed in 99% (CAMP-1) and 95% (CAMP-2) of VP-102-treated participants and 73% (CAMP-1) and 66% (CAMP-2) of vehicle-treated participants. The most common adverse events included application site vesicles, pain, pruritus, erythema, and scab. Most adverse events were mild or moderate in severity.

Conclusions and relevance: In the 2 phase 3 trials reported herein, VP-102 was statistically significantly superior to vehicle in achieving complete clearance of MC lesions at the end of the study visit in both trials, with adverse events that were generally mild to moderate and confined to application sites. These findings show that VP-102 is potentially an effective and safe treatment for MC, a common skin condition with no US Food and Drug Administration-approved treatments.

Trial registrations: ClinicalTrials.gov Identifiers: NCT03377790 and NCT03377803.

Conflict of interest statement

Conflict of Interest Disclosures: Dr Eichenfield reported receiving grants from Verrica Pharmaceuticals Inc during the conduct of the study as well as personal fees and other support from Verrica Pharmaceuticals Inc outside the submitted work. Dr McFalda reported being a paid advisory board member for Paidion/Verrica Pharmaceuticals Inc during the conduct of the study. Dr Brabec reported receiving research site funding from Verrica Pharmaceuticals Inc during the conduct of the study. Dr Siegfried reported receiving research site funding paid to the institution from Verrica Pharmaceuticals Inc during the conduct of the study; personal fees from Verrica Pharmaceuticals Inc and Novan; personal fees, consultant fees, and research fees paid to the institution from Regeneron; consultant fees and research fees paid to the institution from Sanofi; consultant fees and research fees paid to the institution from Lilly; grant funding paid to the institution and fees for being on the data safety monitoring board from Pfizer; personal fees from Leo; research fees paid to the institution from Janssen; and research fees paid to the institution from Stiefel outside the submitted work. Dr Kwong reported receiving personal fees from Verrica Pharmaceuticals Inc during the conduct of the study; personal fees from Verrica Pharmaceuticals Inc outside the submitted work; and serving as a paid advisory board consultant for Verrica Pharmaceuticals Inc. Dr McBride reported receiving grants from Paidion Research Inc and grants from Verrica Pharmaceuticals Inc outside the submitted work and was paid, as a consultant, by the sponsor company for this study to run the analyses for this study and to perform quality control for the results. Dr Rieger reported being an employee of Verrica Pharmaceuticals Inc during the study, receiving company stock from Verrica Pharmaceuticals Inc, and having a patent to Verrica Pharmaceuticals Inc; he was also an employee of PBM Capital Group outside the submitted work; Ms Willson is an employee of Verrica Pharmaceuticals Inc and holds company stock. Dr Davidson reported personal fees and other support from Verrica Pharmaceuticals Inc during the conduct of the study and holds company stock; in addition, Dr Davidson had a patent to WO2018226894A1 pending, a patent to WO2018232277A1 pending, a patent to WO2016100732A3 pending, a patent to WO2016118633A1 pending, a patent to WO2015027111A1 pending, and patent number 2014308690 issued. Dr Burnett reported receiving personal fees from Verrica Pharmaceuticals Inc during the conduct of the study. No other disclosures were reported.

Figures

Figure 1.. CONSORT (Consolidated Standards of Reporting Trials) Flow Diagram of Participant Disposition in Cantharidin Application in Molluscum Patients (CAMP-1 and CAMP-2) Trials (Intent-to-Treat Population)

After randomization, one participant in the VP-102 group was found not to have met eligibility requirements and was not treated, although the participant was included in the intent-to-treat efficacy analyses. Two separate participants were not linked to housemates who were in the VP-102 group and were incorrectly randomized to the vehicle group. Both participants received treatment with VP-102 (consistent with their housemate’s group) and were included in safety analyses.

Figure 2.. Percentage of Participants With Complete Clearance of Baseline and New Molluscum Contagiosum Lesions After Treatments With VP-102 or Vehicle (Statistical Comparisons at Day 21, 42, 63, or 84)

Complete clearance of all treatable lesions (baseline and new) at individual time points after treatment, treated as binary end points and tested with a Pearson χ2 test. All statistical tests were 2-sided with a significance level of α = .05. P values were completed for the intent-to-treat population at individual time points, where independent assessments were performed at days 21, 42, 63, and 84, and not meant to imply statistical significance between those time points. Individual lesions were not tracked, and total lesion counts included any treatable (baseline and new) lesions that remained on the day/visit after baseline. If participant lesions cleared before the end of study (EOS), they were required to stay clear until the EOS visit to be counted as completely clear. At the EOS visit, 46.3% of participants in the VP-102 group and 17.9% in the vehicle group were completely clear of all lesions for Cantharidin Application in Molluscum Patients (CAMP)-1 (P < .001). In CAMP-2, 54.0% of VP-102 participants vs 13.4% of the vehicle participants were completely clear at the EOS visit (P < .001). While the methods for the trials were identical, clinical sites and participants were not the same between trials, and thus there is natural variability in results for each group from one trial to the other. aP < .05. bP < .01. cP < .001

Figure 3.. Percentage Change From Baseline in Lesion Counts After Treatment With VP-102 or Vehicle (Statistical Comparisons at Day 21, 42, 63, or 84)

Lesion counts were taken at each time point/visit (baseline, day 21/visit 2, day 42/visit 3, day 63/visit 4, and day 84/the end of study (EOS) visit, intent-to-treat population). Lesion counts at each visit after baseline were compared with baseline to determine percentage change. Percentages were analyzed with an analysis of covariance model. VP-102 showed significant differences (P < .001) in percentage change in lesion count from baseline compared with vehicle at each time point/visit.