Self‐perception of natural outcome, appearance, and emotional well‐being after OnabotulinumtoxinA treatment for upper facial lines: Post hoc analysis across age and gender

OnabotulinumtoxinA (onabotA) is indicated for upper facial lines (UFL). Fear of unnatural‐looking outcomes is a frequently reported treatment barrier.


| INTRODUC TI ON
The interest in minimally or non-invasive aesthetic procedures has grown over the past 2 decades, with botulinum toxin type A (BoNT/A) injections being the most commonly performed. 1As aesthetic medicine has grown in popularity, so has the diversity of the population seeking these procedures.While the majority of individuals receiving an aesthetic procedure are still women over 40, the interest and demand from millennials (defined as those born between 1981 and 1997) and men have risen substantially. 1 When seeking out these procedures, most desire a youthful and natural appearance, but gender-and age-specific anatomy, along with psychosocial and emotional differences, demand that special considerations be made when treating these patients to meet their aesthetic goals. 2,3e fear of unnatural-looking outcomes is a frequently reported barrier to receiving treatment; this trend appears to hold across age and gender. 2,4,5For many considering toxin treatments, natural could mean avoiding the "frozen look" that comes from improper dosing or overuse of BoNT/A.For others, the goal for treatment is to refresh or enhance ones' personal features yet still look familiar and not noticeably different. 6Achieving an aesthetically pleasing and natural look likely comes from several factors, including product placement, injection technique, and dosing. 7In the case of BoNT/A, one current tendency of practitioners is to use lower than labeled doses as a conservative measure for more subtle results; however, it is unclear whether this approach produces more natural and satisfactory outcomes from the patient perspective. 8Given that many patients seek treatment to improve their emotional well-being, it is also important to evaluate whether the resulting aesthetic outcomes fulfill these expectations. 9A better understanding of what constitutes a natural look and how patients perceive themselves after toxin treatment would help practitioners address patients' concerns and communicate their mutual goal of aesthetically pleasing results.
Forehead lines (FHL), glabellar lines (GL), and crow's feet lines (CFL), collectively referred to as upper facial lines (UFL), are among the main areas of concern for aesthetically conscious individuals. 51][12] While the safety, efficacy, and positive impact of onabotA treatment for UFL have been previously established, there are limited clinical data that specifically examine satisfaction with natural-looking results from the patient perspective. 10,11As self-perceived naturalness is a key motivator, but also a potential objection to pursue onabotA treatment, patient-reported outcome (PRO) measurements are well suited to address this knowledge gap.This post hoc analysis of PROs from 2 phase 3 clinical trials aimed to evaluate participant-perceived satisfaction with a natural look, satisfaction with the treatment effect, whether the treatment met expectations, and the psychological impact of onabotA treatment for UFL.These outcomes were assessed among all participants who met the primary efficacy endpoint, as well as subsets of millennials and men, using the Facial Line Satisfaction Questionnaire (FLSQ) scale that was developed and validated according to FDA PRO guidance. 13

| Study design
This post hoc analysis pooled data from two, 12-month, randomized, placebo-controlled phase 3 studies (study 1 [NCT02261467] and study 2 [NCT02261493]) that evaluated onabotA for the treatment of UFL. 10,11Detailed inclusion and exclusion criteria were previously described. 10,11Briefly, eligible participants were botulinum toxin naive men or women aged 18 years or older.Participants selfidentified as either female or male.In both studies, participants must have had investigator-and self-assessed moderate to severe FHL at maximum eyebrow elevation and investigator-assessed moderate to severe GL at maximum frown, based on the Facial Wrinkle Scale with Photonumeric Guide (FWS; 0 = none; 3 = severe).In study 2, participants must have also had moderate to severe CFL at maximum smile as assessed by the investigator using the FWS.
In study 1, participants were randomized to receive either placebo or 40 U onabotA (20 U in FHL, 20 U in GL).In study 2, participants were randomized to receive either placebo, 40 U, or 64 U onabotA (20 U in FHL, 20 U in GL, with or without 24 U in CFL).Both studies were composed of a 6-month double-blind period (treatment cycle 1), followed by a 6-month open-label treatment period in which participants could receive up to two additional treatments with on-abotA (treatment cycles 2 and 3).Retreatment criteria for additional cycles were previously described. 10,11Follow-up evaluations occurred on Days 7, 14, and 30, and every 30 days after each treatment in all treatment cycles.Both studies were conducted following Good Clinical Practice guidelines and the principles of the Declaration of Helsinki.All participants provided written informed consent.

| PRO measurements
The FLSQ is a proprietary questionnaire developed and validated according to FDA guidelines for PRO measures, which assesses treatment satisfaction and psychological impact as perceived by patients. 13The FLSQ was developed with input from participants with FHL, GL, and CFL during face-to-face interviews that included concept elicitation (i.e., asking what concepts are most relevant to their UFL and associated treatment) and cognitive debriefing (i.e., asking if the measure is comprehensive and that the instructions, items, and response options are well understood and clear).
Stand-alone FLSQ items were used to assess satisfaction with receiving a natural look (Item 4), satisfaction with treatment effect (Item 5), and treatment meeting or exceeding their expectations (Item 11).Items 4 and 5 are rated on a 5-point Likert scale (very dissatisfied to very satisfied), while Item 11 is rated on a 3-point Likert scale (worse than expected, met expectations, better than expected).Responders for Items 4 and 5 were classified as those who answered "mostly" or "very satisfied," while responders for Item 11 were those who answered that treatment "met expectations" or was "better than expected." The FLSQ Impact Domain measures self-perceived appearance and emotional impacts associated with UFLs.It includes the participant's evaluation of five items regarding their facial lines: looking tired, looking angry, feeling older, feeling unhappy about facial lines, and negatively affecting self-esteem.Each of these items were rated on a 5-point scale (from "not at all" to "a lot/extremely").The FLSQ Impact Domain scores are a sum of the 5 individual item scores transformed to a scale of 0-100, where a higher score indicates a greater negative psychological impact from facial lines.FLSQ Impact Domain responders were defined as participants achieving a ≥20point improvement from baseline, among those who reported negative impact at baseline (defined as a baseline score of ≥20-points).This is an anchor-based responder definition previously proposed to be the smallest amount of change on the FLSQ Impact Domain considered meaningful to participants. 13

| Statistical analysis
Analyses were performed on the pooled modified intention-totreat (mITT) populations from both studies who achieved at least a 2-grade improvement from baseline in both investigator and participant FWS ratings of FHL severity at Day 30 post-treatment 1, and who met the criteria for additional treatments (FWS responder set).Subgroup analysis was performed for millennials (born between 1981 and 1997, age 18-34 at the time of study randomization) and men.[Corrections made on 28 December 2023, after first online publication: 'UFL' in the previous sentence has been corrected to 'FHL' in this version.]Because age data were collected without a birth year, a derived birthday calculated by subtracting reported age (age multiplied by 365.25 days) from the study randomization date was used to define millennial participants.Response rates were calculated by visit for the FWS responder set, millennials, and men.The response rate for all items was also analyzed at each primary time point for those in the FWS responder set who achieved a 3-grade improvement in FWS rating, according to both the participant and investigator.The 3-grade stratification was not performed for the millennial or men subgroups because of the low participant count.Wald limits without continuity correction were used to calculate the 95% confidence interval (CI) for each response rate.
Logistic regression models for repeated measures with generalized estimating equations were used to explore the association between variables (millennials vs. non-millennials, men vs. women, or repeated treatment over 12 months) at prespecified time points: Day 30 for FLSQ Items 4,11 and the FLSQ Impact Domain, and Day 60 for FLSQ Item 5. 12 Day 60 was chosen for Follow-up Item 5 based on previously established trends that showed peak participant satisfaction was achieved for Item 5 at this timepoint, after peak clinical efficacy was observed at Day 30. 14Each model was adjusted by participant FWS rating of FHL severity at max eyebrow elevation and its own endpoint baseline value except for FLSQ Item 11, which does not have a corresponding baseline item.A two-sided p-value less than 0.05 was considered to be statistically significant.

| Participant demographics
This analysis pooled participants from the two studies that received at least 1 treatment of 40 U or 64 U onabotA, achieved at least a 2-grade improvement from baseline in both investigator and participant FWS ratings of FHL severity at Day 30 post-treatment, and who met the criteria for additional treatments.Out of 921 participants randomized to receive onabotA in the 2 studies, a total of 458 met these criteria, which included 94 millennials and 43 men (Table 1).
Most participants in the overall analysis population were female (90.6%) and White (88.4%).A higher proportion of millennials had both investigator and participant FWS ratings of moderate, while a higher proportion of men had FWS ratings of severe at baseline.The safety data for the entire population and the millennial and men subgroups have been previously reported, and thus, no safety analysis was performed here. 10,11,15,16

| FLSQ Item 4: Satisfaction with natural look
The majority of participants (>80%) were FLSQ Item 4 responders throughout the 12-month period, being mostly/very satisfied with achieving natural-looking outcomes (Figure 1).On Day 30 in the first treatment cycle, 90.5% of the FWS responder set, 94.6% of millennials, and 85.7% of men were satisfied with natural-looking outcomes.
Of the FWS responder set, 94.0% of those who achieved a 3-grade improvement reported satisfaction with the treatment giving natural results.At this timepoint, millennials had 3.30-times higher odds of being satisfied with the treatment giving natural results than non-millennials (OR = 3.30, 95% CI [1.30, 8.41], p = 0.01).There were no statistically significant differences in the FLSQ Item 4 responder rates between men and women and having more than 1 treatment did not significantly impact whether individuals were satisfied with natural outcomes over the 12-month period (Table 2).However, it is important to note that the number of participants was considerably smaller by the third treatment cycle (Table 3).Representative images of a 56-year-old woman, a 35-year-old millennial woman, and a 63-year-old man who replied "very satisfied" to FLSQ Item 4 at treatment 1 Day 30 are shown in Figure 2.

| FLSQ Item 5: Satisfaction with treatment effect
The majority of participants (>80%) were FLSQ Item 5 responders throughout the 12-month period, defined as being satisfied or very satisfied with the treatment effect on their facial lines (Figure 3).On Day 60 in the first treatment cycle, 89.8% of the FWS responder set, 95.7% of millennials, and 83.3% of men were satisfied with the treatment effect.Of the FWS responder set, 92.8% of those who achieved a 3-grade improvement in FWS rating reported satisfaction with the treatment effect.Millennials had 2.22-times higher odds of being satisfied with the treatment effect than non-millennials (OR = 2.22, 95% CI [1.14, 4.33] p = 0.02).There were no significant differences between men and women; repeat treatments over the 12-month period did not significantly impact satisfaction with the treatment effect (Table 2).

| FLSQ Item 11: Met expectations
The proportion of FLSQ Item 11 responders, those reporting that their treatment expectations were met or were exceeded, was high (>90%) throughout the study period for all subgroups (Figure 4).On Day 30 in the first treatment cycle, 96.0% of the FWS responder set, 98.9% of millennials, and 92.9% of men reported that treatment expectations were met.Of the FWS responder set, 100% of those who achieved a 3-grade improvement in FWS rating reported that the treatment met expectations.There were no significant differences in whether the treatment expectations were met between millennials and non-millennials, or between men and women (Table 2).
Additionally, repeat treatments over the 12 months did not affect whether the treatment met expectations (Table 2).

| FLSQ Impact Domain: Self-perceived appearance and emotional impact
Across all subgroups, 39.5%-87.5% of participants throughout the study were FLSQ Impact Domain responders, exhibiting a ≥20-point improvement from baseline (Figure 5).The FLSQ Impact Domain  F I G U R E 3 FLSQ Item 5 Responder Rate.The FLSQ Item 5 responder rate among the FWS responder set, millennials, and men.Responders were those who answered "mostly satisfied" or "very satisfied" to FLSQ Item 5, which addressed satisfaction with the treatment effect.The sample size for each timepoint is presented in Table 3.

FWS Responder Set Millennials Men
This post hoc analysis of two phase 3 studies examined satisfaction with natural-looking outcomes, satisfaction with the treatment effect on facial lines, whether the treatment met expectations, and self-perception of appearance and emotional well-being after onlabel dosing of onabotA treatment for UFL using items from the validated PRO measure, the FLSQ.A high satisfaction rate with achieving a natural look after onabotA treatment was observed in participants achieving a clinically meaningful ≥2-grade change (from severe/moderate to none/mild) on investigator-assessed FWS scores, including in millennials and men.These participants were also highly satisfied with the onabotA treatment effect, and many reported improvements in appearance-related psychological impacts of UFL.High satisfaction on these PROs were sustained after a single treatment and were maintained or improved with repeated onabotA treatment.Altogether, these results demonstrate that treatment with onabotA at the on-label doses studied here is The FLSQ Item 11 responder rate among the FWS responder set, millennials, and men.Responders were those who answered "met expectations" or "exceeds expectations" to FLSQ Item 11, which addressed whether the effect of treatment met expectations.The sample size for each timepoint is presented in Table 3.In these studies, the injection sites targeting UFLs were rationally selected, taking into account the interactions between the elevator muscles in the frontalis and the opposing depressor muscles of the glabellar complex. 17Co-treatment of UFLs may create a balanced and harmonious effect, thus providing more natural-looking results, higher satisfaction with treatment, and greater improvements in self-perceived appearance and emotional well-being. 18Moreover, in other studies higher doses given in the glabellar region (20-80 U) were shown to increase the duration of effect while maintaining participant satisfaction with natural results. 19Together, the results shown here and elsewhere suggest that there may be considerable flexibility in the relationship between dose and natural outcomes.
The FLSQ Impact Domain is a composite of five items related to how facial lines make the participants look and feel (look tired, look angry, feel older, feel unhappy, and negative effects on selfesteem).Therefore, the responder rate should correspond to the treatment effect on facial lines, trending upward with the improvement in UFL severity and downward as the visible effects decline.
The FLSQ Impact Domain responder rates significantly increased through the second and third treatment cycles, indicating that more participants experienced improved self-perception and related emotional impact after repeat onabotA treatments (Table 2).
It is plausible that after becoming familiar with the treatment, participants' perceptions may change over time as they realize that they still look natural, adjust to their appearance after treatment, and develop a heightened sense of self-esteem.Recent studies have demonstrated increased self-reported ratings of social functioning after onabotA treatment, supporting this hypothesis. 20rthermore, as confidence transfers over to how one presents themselves, self-perceptions can both affect, and be affected by, how one is seen by others.Future work incorporating observerreported outcomes will help clarify what represents a natural, satisfactory outcome from aesthetic procedures.
Understanding the millennial perspective on toxin treatments is important as the frequency of aesthetic procedures in this age group continues to rise.Here, the baseline FWS severity was more likely to be moderate for the millennial subgroup, which was expected given that their younger skin likely has experienced less physiological aging.Younger adults may look to aesthetic treatment as a preventative measure to maintain a youthful, flawless appearance rather than as a treatment for present aging. 2,9Despite millennials having less severe wrinkles, use of the on-label dose rather than the trend of using fewer units, may yield a longer duration of response, reducing the frequency of treatments needed to maintain satisfactory results. 21Millennials had higher odds than non-millennials of being satisfied with natural-looking outcomes and with the effect of treatment with 40 U (FHL/GL) or 64 U (FHL/GL/CFL) of onabotA, also supporting the use of on-label dosing in this population.Digital engagement and social media use is a significant part of the millennial lifestyle, and the Internet and social media are increasingly used as sources of information related to aesthetic procedures. 5,22It is plausible that the de-stigmatization of aesthetic treatments increased awareness of accessible procedures, and a particular focus on personal appearances prompted by social media has impacted perceptions of beauty among this generation. 23In the authors' view, this could influence their goals and expectations for these treatments, affecting their experience and satisfaction with the results.
Results from this study help address the knowledge gap about outcomes specific to men after onabotA treatment.Compared to the women in this study, in men, there was no significant difference in satisfaction with a natural look, the effect of treatment, and meeting expectations.However, men had significantly lower odds of perceiving a positive emotional impact after treatment when compared to women, as indicated by the lower FLSQ Impact Domain responder rate, despite men having similar FLSQ Impact Domain scores as all FWS responders at baseline.One possibility for this difference is that men require a higher dose to achieve similar results as compared to women, owing to differences in facial anatomy, larger muscle mass, and a tendency toward a higher severity of wrinkles. 3th regard to investigator-assessed efficacy, 64 U dosing in men is currently supported with sufficient clinical evidence, but psychosocial outcomes might be improved with even higher doses as seen in previous studies with higher doses given to treat GLs. 24,25Future studies with men should evaluate the relationship between dose and psychological impact after onabotA treatment.
It is important to consider the results in the context of the study's limitations.The phase 3 studies included in this analysis were

F I G U R E 2
Images of (A) a 56-year-old woman, (B) a 35-year-old millennial woman, and (C) a 55-year-old man, before and after onabotA treatment, at rest (top) or at maximum eyebrow elevation (bottom).

5 FLSQ
Impact Domain Responder Rate.The FLSQ Impact Domain responder rate among the FWS responder set, millennials, and men.Responders were those who achieved ≥20-point improvement from baseline on the FLSQ Impact Domain.The sample size for each timepoint is presented in Table3.with self-perceived natural looking and satisfactory outcomes across gender and age.A ≥2-grade improvement on the FWS at Day 30 post-treatment was chosen as evaluation criteria because this translates to a considerable physical improvement based on investigator judgment.Day 30 is when the peak visible effect of onabotA is present, and thus is the point where the results are most likely to look unnatural or frozen.The consistently high response rate for FLSQ Item 4 across all time points shows that most individuals were satisfied with their natural outcomes.This was corroborated by high satisfaction with the effect of treatment (Item 5) and met expectations (Item 11), which together suggests that the treatment fulfilled the primary treatment goals of the participants.Furthermore, outcomes were similarly high for those achieving a 3-grade improvement on the FWS (a change from severe to none), which shows that even substantial changes in FHL severity can produce natural and satisfactory results from the participant perspective.[Corrections made on 28 December 2023, after first online publication: 'UFL' in the previous sentence has been corrected to 'FHL' in this version.]As per the phase 3 studies for onabotA, participants included in this analysis were treatment naive.When addressing fears of unnatural results in first-time patients, some providers may recommend starting with lower doses of onabotA than what is indicated for UFL.The high patient-reported satisfaction with natural results shown here with the on-label dosing scheme of 40 U (FHL/ GL) or 64 U (FHL/GL/CFL) provides evidence that such conservative dosing may not be necessary.Along with the dose, there are other important considerations for achieving natural-looking results.