PrabotulinumtoxinA for the treatment of glabellar lines in adults, 65 years of age and older: The fourth in a series of post hoc analyses of the phase III clinical study data

Background: Despite the absence of Phase III clinical trial data to support their use, 3.4% of all neurotoxin treatments performed internationally for esthetic purposes in


| INTRODUC TI ON
by age, data have typically been stratified by age <50 years and ≥50 years. [7][8][9][10][11] As extracted from the FDA-approved prescribing information and listed below, few patients over the age of 65 participated in any of the pivotal placebo-controlled clinical studies conducted with other botulinum toxins for the treatment of glabellar lines. [3][4][5][6] Participants in this age group for this indication included Ltd.). 6 In comparison, a total of 70 participants 65 years of age and older were treated with 20 U prabotulinumtoxinA (Jeuveau; Evolus Inc.) in the three single-dose placebo-controlled Phase III glabellar line clinical studies. 12,13 Given the dearth of data with other toxins in patients 65 years of age and older, the authors thought it would be of interest to explore outcomes for the 70 patients in this population who had been treated with prabotulinumtox-inA compared with similarly treated patients less than 65 years of age. Post hoc analyses of data pooled from the Phase III studies were undertaken to better comprehend the efficacy and safety of prabotulinumtoxinA for the treatment of glabellar lines in this understudied population.
Of note, the approach taken and the endpoints examined in this latest subgroup analysis were intentionally similar to those taken and examined in the three previous post hoc analyses, which have included investigations of outcomes in prabotulinumtoxinA-treated patients with skin of color, millennials, and males. 14-16 Because of the age-based nature of both analyses, our paper most closely parallels that of the millennial analysis. However, the two populations under investigation are markedly different. While the current analysis is focused on those 65 years of age and older, millennials at the time that the studies were conducted had a maximum age of 33 years while the comparator cohort was 33 years of age and older. 15 2 | MATERIAL S AND ME THODS

| Conduct of the original studies
The EV-001 and EV-002 studies were conducted in parallel in the U.S. between January 2015 and September 2015 12 ; the EVB-003 study was conducted in, Germany, France, Canada, Sweden, and the United Kingdom between June 2015 and April 2016. 13 All three were designed as multicenter, randomized, double-blind, placebocontrolled Phase III studies of 150 days' duration; the EVB-003 study also had an active-control arm. All participants were adults, at least 18 years of age, who had moderate to severe glabellar lines at maximum frown; those in the EVB-003 study also had to have found that their glabellar lines had an important psychological impact (i.e., on mood, anxiety, and/or depressive symptoms). All provided written informed consent before participating in these studies. Those and investigator assessment of adverse events.

| Statistical methods of the post hoc analyses
Similar in methodology to previously reported subgroup analyses of prabotulinumtoxinA-treated patients, 14-16 data extracted from these single-dose glabellar line studies were pooled and summarized-in this case, for patients less than 65 years of age and those 65 years of age and older. Safety data are reported for all 737 prabotulinumtoxinA-treated participants. Of these, four patients withdrew after receiving treatment before any post-baseline data had been collected, and thus, efficacy data are presented for the 733 participants who had post-baseline data. For efficacy endpoints, the proportion of responders between the two cohorts at each post-treatment assessment visit was compared using the Fisher's exact test; exact binomial limits for the absolute differences were calculated for the two-sided exact 95% confidence intervals.
Kaplan-Meier methodology was used to calculate the durability of response (number of days at which 50% of patients had stopped responding) for the two age groups. As in each of the previous subgroup analyses, [14][15][16] the key efficacy endpoint was the percentage of responders with a ≥1-point improvement from baseline based on the GLS at maximum frown by investigator assessment. The other efficacy endpoint based on the GLS was the percentage of responders with a post-baseline score of 0 or 1 (none or mild) at maximum frown. Non-GLS endpoints included the percentages of positive responders on the GAIS (i.e., those with a post-baseline score of improved or much improved) and of positive responders on the SSS (i.e., those with a post-baseline score of satisfied or very satisfied). Treatment-related adverse events, defined by the investigator as possibly, probably or definitely treatment related, were summarized including all treatment-related events, most common treatment-related events (i.e., headache), and treatment-related adverse events of particular interest for this type of treatment and indication (e.g., ptosis). Statistical analyses of adverse event data were not performed.

| Patient disposition and demographics
Of the 737 prabotulinumtoxinA-treated patients, 70 (9.5%) were 65 years of age or older; 667 (90.5%) were less than 65 years of age ( US studies, less than half of the 492 prabotulinumtoxinA-treated participants in those studies had had previous exposure to a botulinum toxin: 45.3% of those 65 years of age or older; and 38.7% of those less than 65 years of age. Of the 70 prabotulinumtoxinA-treated patients who were 65 years of age or older, 53 (75.7%) were from the U.S., eight (11.4%) were from France, five (7.1%) were from Germany, three (4.3%) were from Canada, and one (1.4%) was from the UK.

| Efficacy
Efficacy data were available for 733 of the 737 (99.5%) prabotulinumtoxinA-treated patients, including 70 patients who were 65 years of age or older and 663 who were less than 65 years of age; four patients, all of whom were less than 65 years, had no post-baseline data. Representative photographs of glabellar lines of a patient aged 65 years at rest and at maximum frown taken at baseline and following treatment with 20 U prabotulinumtoxinA are presented in Figure 1A-H.
While comparatively less than that of younger patients at most time points, those 65 years of age and older also exhibited strong responder rates-most notably at each of Days 7, 14, and 30-for each of the efficacy endpoints examined (Figures 2-6). As outlined below, as might be expected given the differences observed in the severity of baseline GLS scores, differences between the two cohorts were most marked based on the percentage of responders who achieved a postbaseline score of 0 or 1 (none or mild) on the GLS at maximum frown.

| Responders on the GLS
The percentages of prabotulinumtoxinA-treated patients who experienced a ≥1-point improvement on the GLS at maximum frown by investigator assessment were numerically lower at five out of the seven time points for patients who were 65 years of age and older compared with those less than 65 ( Figure 2); the absolute mean difference across all visits was −2.7%. None of the differences observed reached statistical significance. Responder rates for this endpoint were 50% or greater for both cohorts at each of Days 2, 7, 14, 30, 90, and 120. The absolute difference in responder rates was only markedly greater for younger patients at Day 7 and Day 120, by 7.5% and 7.3%, respectively; otherwise, differences between the two cohorts were unremarkable. For this key endpoint, the durability of response (i.e., the number of days at which 50% of patients had stopped responding) was estimated to be 142 days (95% CI 121, TA B L E 1 Baseline demographic and glabellar line characteristics. 148) for patients who were 65 years of age and older compared with 149 days (95% CI 148, 150) for those less than 65.
Differences between the percentages of prabotulinumtoxinAtreated patients who experienced a post-baseline score of 0 or 1 (none or mild) on the GLS at maximum frown by investigator assessment were numerically lower at six out of the seven time points for patients age 65 years and older compared with those less than 65 ( Figure 3). Differences between the two cohorts were most pro-

| Safety
The overall incidence of treatment-related adverse events was higher among younger patients: 13.8% in patients less than 65 years of age versus 11.4% in those 65 years of age and older ( Table 2).
Headache, which was reported in 5.7% of patients 65 years of age and older, and in 9.7% of those less than 65 years, was the most common treatment-related event. Among patients 65 years of age and older, two patients experienced treatment-related eyelid ptosis

| DISCUSS ION
It has been suggested that, in general, botulinum toxins as a group are less effective in older patients. 17 Indeed, lower response rates have typically been observed in older patients when Phase III glabellar line clinical trial results with other toxins were stratified by age less than 50 years and greater than or equal to 50 years. [7][8][9][10][11] Of the many neurophysiological changes that occur as we age, one potential explanation for the drop in efficacy of botulinum toxins may be related to changes that occur at the level of the neuromuscular junction. As reviewed by Jang and Van Remmen, 18  to focal denervation of the neuromuscular junction has been identified as a key event in these types of age-related end plate changes.
The Internet has provided a forum to perpetuate a "myth" that neuromodulators stop working in the elderly, compounded by the limited content of the FDA labels. Some opinion leaders have suggested that lower doses of botulinum toxin may be more appropriate for older patients, [19][20][21][22] which may have been in an effort to limit potential adverse events. However, these suggested lower doses may further limit the effectiveness of treatment in an older population.
An alternative approach would be to use a higher dose of a toxin, which ideally has a narrow field of effect to limit adverse events. In were not statistically significant at any visit post Day 30 for any of the five efficacy endpoints examined. Table 3 Not only did 3/4 of patients aged 65 to 81 years, who had been treated with 20 U prabotulinumtoxinA, achieve treatment success at Day 30 for the more rigorous GLS endpoint assessed (i.e., GLS score of 0 or 1 at maximum frown), but they did so without incurring a disproportionate percentage of treatment-related adverse events.

As summarized in
That is, compared with 13.8% of prabotulinumtoxinA-treated patients less than 65 years, only 11.4% of those 65 years of age and older experienced treatment-related adverse events. As a point of comparison, 9.0% of all 211 patients who were treated with placebo in these three studies also experienced adverse events that the investigator had determined to be treatment related (data on file, Evolus); as in the case of prabotulinumtoxinA-treated patients, most of these events were headache. Importantly, compared with younger patients, prabotulinumtoxinA-treated patients who were 65 years of age and older also did not experience a large increase in the rate of treatment-related ptosis events-a complication of special interest that had been thought to be a greater risk for older patients. 19 The limited sample size, the pooling of patients across three studies, and the post hoc nature of these analyses are all acknowledged limitations of this effort to understand the efficacy and safety of 20 U prabotulinumtoxinA in patients 65 years of age and older.

ACK N OWLED G M ENTS
The authors acknowledge the dedicated work of all investigators in The authors also gratefully acknowledge the contribu-

E TH I C S S TATEM ENT
All patients provided written informed consent prior to entering a study. As stated in manuscript reference 12 regarding the conduct

PH OTO CO N S ENT S TATEM ENT
Please be aware that no separate photo consent forms for these studies exist. Rather the patient featured in Figure 1A Avelar, on behalf of Evolus Inc. that signed study informed consent was obtained for this patient allowing for this type of use of the patient's photographs.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data available upon request.