Treatment for Glabellar Lines
Gary D. Monheit, M.D.
Department of Dermatology
University of Alabama at Birmingham
Background: Botulinum toxin is used to treat glabellar lines, but the optimal dose of Reloxin™ (Dysport®) for this indication remains to be established.
Objectives: Evaluation of three doses of Reloxin to determine the efficacy and safety in treating glabellar lines.
Methods: Participants were given 20, 50, or 75-units Reloxin, or placebo, injected across the glabella. Follow-up was on Days 7, 30, 60, 90, and 120. Adverse events were also elicited by telephone on Day 3.
Results: From investigators’ and participants’ assessments at Day 30, there were significantly more responders in each Reloxin-treated group than placebo (p?0.001). The 50 unit dose was as effective as the 75 unit dose, with a similar duration. The most common adverse events were mild headache and nasopharyngitis, occurring similarly in all groups.
Conclusions: Reloxin (Dysport) treatment resulted in a significant improvement in glabellar lines, and the 50 unit dose was identified as optimal. All doses were well tolerated.
The growth of the aging population, especially over the last decade, has led to an increasing demand for procedures to reverse the appearance of aging, particularly around the face. This has resulted in the development of minimally invasive cosmetic procedures such as ablative and non-ablative resurfacing lasers, chemical peels and the use of chemodenervation and filling agents. Because none of these is risk-free, continuing research is necessary to provide the safest and most effective methods for treating the aging face.
The interplay of five factors is considered to produce “the aging face”(1), but it is primarily the interaction of two that produces the more significant lines: the skin and the underlying muscles. The skin factors are treated with many therapies, among them ablative and non-ablative resurfacing and injections for soft tissue augmentation. One treatment directed at the muscles underlying the hyperkinetic lines, is botulinum toxin type A (BoNT-A). This agent temporarily paralyzes specific muscle groups and has been shown to ameliorate lines and folds (2-14). Both clinical observation and objective measurement of glabellar folds, have shown that low doses of BoNT-A suppress the muscular activity of the glabellar area by temporarily paralyzing the procerus and corrugator supercilii muscles. Many studies regarding the cosmetic use of this agent have now been published (15).
Reports on the use of Dysport (Reloxin) in cosmetic treatments suggest that it is effective in reducing the severity of glabellar lines (15-23). The doses used in these studies ranged from approximately 40 to 100 units.
The aim of this study was to evaluate three doses of Reloxin, by comparison with placebo, to determine the optimal dose in reducing the severity of hyperfunctional glabellar lines, with a view to using this dose in subsequent studies. We also assessed the safety of Reloxin at the doses tested.
This was a multi-center, double-blind, placebo-controlled study.
Written informed consent was obtained from 373 adult male or female participants, who had moderate or severe vertical glabellar lines at maximum frown and who fulfilled the other entry criteria.
Women of childbearing age were required to take appropriate contraceptive measures throughout the study period.
Participants were treated on Day 0 with Reloxin (Ipsen Biopharm Ltd., Wrexham, UK) (20, 50, or 75 units), or placebo, injected at five sites (0.05 mL per injection site) across the procerus, corrugator, and orbicularis muscles. Both participants and investigators were blinded to the treatment. Following injection, participants remained under observation for 30 minutes and were contacted three days later, by telephone, to check for adverse events and concomitant medications. Follow-up visits were made on Days 7, 30, 60, 90, and 120.
A blood sample was taken on Screening and on Day 120 for testing to establish whether neutralizing antibodies to BoNT-A were present.
Primary outcome measures
Co-primary efficacy endpoints were selected: the investigators’ live assessment of glabellar lines at maximum frown at Day 30 and the participant’s self-assessment of change in severity of glabellar lines at Day 30. All other assessments were either secondary or exploratory and are described below.
On Days 30, 60, 90, and 120 after treatment, investigators evaluated the participants by comparing their appearance to validated scales of glabellar lines, at maximum frown and at rest. Each scale was comprised of four photographs graded 0 to 3: Grade 0 (none), Grade 1 (mild), Grade 2 (moderate), or Grade 3 (severe). The scales were established in conjunction with several clinical experts, Parexel International and Canfield Scientific Inc., who provided the reference photographs and assisted with the scale validation process and the analyses.
“Live” assessments of the participants were made and compared to the validated photographic scales. Live assessment was considered to have a number of advantages over photographic comparisons, in that the assessor was able to evaluate the level of effort being made by the participant in attempting to frown and the dynamic nature of frowning was taken into consideration. In addition, factors which could vary across the different centres, such as lighting and camera angle, could be discounted by the investigator when focusing on the frown lines. The nature of photographs tends to flatten the facial image of the face, including the lines, confounding the accuracy of assessments. Live assessment is most commonly used by physicians treating wrinkles and this is the method of choice for assessment.
A “responder” was defined as a participant who had a rating of none (0) or mild (1) glabellar lines at maximum frown at Day 30.
At each visit (except Screening) a photograph was taken of each participant. These were used to support the validity of study assessments, but not to assess the outcome per se. The Day 7 photograph was taken by an individual independent of the study assessments, in order that the study blinding was not broken. On completion of the study (Day 120) this photograph was assessed by the investigator for efficacy, using the validated 4-Point Photographic Scale.
Participants were asked to assign a score that best described their overall assessment of the change in severity of their glabellar lines on Days 30, 60, 90, and 120, compared to pretreatment
.They were asked: “How would you rate the change in the appearance of your glabellar lines compared with immediately before the injection?” using the following nine point scale.
+4 (complete improvement, about 100%)
+3 (marked improvement, about 75%)
+2 (moderate improvement, about 50%)
+1 (minimal improvement, about 25%)
-1 (slight worsening, about 25%)
-2 (moderate worsening, about 50%)
-3 (marked worsening, about 75%)
-4 (very marked worsening, about 100%)
A “responder” was defined as having a grade change of at least +2 (moderate improvement, about 50%) at Day 30.
Participants were also asked to assess the status of their glabellar lines using a 10 cm visual analogue scale (VAS). On this scale zero represented 'no glabellar lines' and ten represented 'severe glabellar lines'.
Continuous data were summarized by treatment group using descriptive statistics (number, mean, median, standard deviation, minimum, and maximum). Categorical data were summarized by treatment group using frequency tables (frequencies and percents). Ninety-five percent confidence intervals were constructed for proportions of successes.
Comparability of the investigators’ assessment of glabellar lines at baseline, across treatment groups, was confirmed using a Chi-squared test. Similarly, comparability among treatment groups for participants’ assessment at baseline using a VAS was confirmed with analysis of variance using an F-test.
A Mantel-Haenszel Chi-Square test, adjusting for age group (?50 years, >50 years) was performed to compare the proportion of responders between each Reloxin group and placebo, at each visit, for the investigators’ assessment and for the participants’ assessment.
Bonferroni’s adjustment was used for multiple comparisons and p ? = 0.0167 (0.05/4) was taken as the significance level.
The intent-to-treat (ITT) data set included all participants randomized and treated with study medication. This data set was used for the analysis of primary and secondary endpoints.
A modified ITT (MITT) data set was used for all exploratory analyses. This data set excluded 13 participants from the ITT population, due to protocol violation.
Exploratory Efficacy Endpoints
- Day 7 Efficacy Measurements
- Participant's Global Assessment of Appearance of Glabellar Lines using VAS
- Duration of Response Measured by Investigator's Assessment at Maximum Frown
- Duration of Response Measured by Participant's Assessment using the nine-point scale
- Subgroup Analyses
- Correlation Between Participant's Assessment of change in appearance of glabellar lines and change in Investigator's Assessment at Maximum Frown
- Correlation Between Participant's Assessment Using 9-point Scale and Using VAS
The demographic data relating to the ITT population is shown in Table 1. The majority of participants were female, under 50 years of age, and Caucasian. The non-Caucasian participants were predominantly Hispanic.
(N = 94)
(N = 91)
(N = 93)
(N = 95)
|Mean (S.D.)||42.5 ± 9.9||41.5 ± 9.7||41.9 ± 10.1||42.1± 10.3|
|Range||20 - 63||20 - 64||23 - 67||20 - 76|
|<50 years||73 (77.7%)||77 (84.6%)||74(79.6%)||74 (77.9%)|
|>50 years||21 (22.3%)||14 (15.4%)||19 (20.4%)||21 (22.1%)|
|Male||10 (10.6%)||10 (13.2%)||21 (22.6%)||17 (17.9%)|
|Female||84 (89.4%)||79 (86.8%)||72 (77.4%)||78 (82.1%)|
|Caucasian||70 (74.5%)||70 (76.9%)||64 (68.8%)||74 (77.9%)|
|Total Non-Caucasian||24 (25.5%)||21 (23.1%)||29 (31.2%)||21 (22.1%)|
At baseline, there were no significant differences between the investigators’ grading of the glabellar lines at maximum frown or at rest, across the treatment groups (Table 2).
(N = 94)
(N = 91)
(N = 93)
(N = 95)
lines at maximum frown
|Grade 2 (moderate)||42 (44.7%)||43 (47.3%)||35 (37.6%)||48 (50.5%)|
|Grade 3 (severe)||52 (55.3%)||48 (52.7%)||58 (62.4%)||47 (49.5%)|
lines at rest
|Grade 0 (none)||3 (3.2%)||5 (5.5%)||4 (4.3%)||5 (5.3%)|
|Grade 1(mild)||42 (44.7%)||35 (38.5%)||42 (45.2%)||40 (42.1%)|
|Grade 2(moderate)||44 (46.8%)||47 (51.6%)||41 (44.1%)||46 (48.4%)|
|Grade 3(severe)||5 (5.3%)||4 (4.4%)||6 (6.5%)||4 (4.2%)|
Effects of treatment – Investigators’ assessments
The investigators’ assessment of glabellar lines at maximum frown found that a there was a statistically significantly larger proportion of responders in each Reloxin treatment group compared with placebo at all time points (p < 0.001 for all comparisons) with the exception of 20 units at Days 90 (p = 0.004) and Day 120 (p = 0.071) (Figure 1).
At the time points beyond Day 30, the number of participants exhibiting a response declined slowly. By Day 120 after treatment, a significant proportion of participants (26% and 27%) continued to show a response in the groups treated with Reloxin 50 or 75 units, respectively. In the groups treated with placebo or Reloxin 20 units, the numbers of participants exhibiting a response was not significant.
In some responders, there was no decline in benefit by Day 120 after injection. None of these individuals was seen in the placebo-treated group. In the Reloxin-treated groups three individuals who received 20 units, eight who received 50 units and seven who received 75 units were considered to have had a sustained improvement at Day 120.
The investigators’ assessment of glabellar lines at rest also found that there was a statistically significantly larger proportion of responders in each Reloxin treatment group compared with placebo (p ? 0.001 to 0.01 for all comparisons). (Figure 2).
Effect of treatment – participants’ assessments
The participants’ assessments of change in severity of glabellar lines showed that there was a statistically significantly larger proportion of responders in each Reloxin treatment group compared with placebo at all time points (p?0.001 for all comparisons) with the exception of 20 units at Day 120 (p = 0.005) (Figure 3).
At the time points beyond 30 days, the number of participants exhibiting a response declined slowly. By Day 120 after treatment, a significant proportion of participants (20%, 42% and 51%) continued to show a response in the groups treated with Reloxin 20, 50 or 75 units respectively.
Some responders, estimated that there was no decline of benefit by Day 120 after injection. Two individuals in the placebo-treated group, made this observation. In the Reloxin-treated groups two individuals who received 20 units, seven who received 50 units and twelve who received 75 units considered that they had a sustained improvement at Day 120.
Similarly, the participants’ assessment of glabellar lines using a visual analogue scale, found that there was a statistically significantly greater reduction in the severity of these lines in the Reloxin treatment groups compared with placebo at all time points with the exception of 20 units and 50 units at Day 120, Figure 4.
Data on different subgroups were available by the end of the study, but due to the small numbers of participants in these groups, no formal analyses were possible. In general, greater benefit was seen in younger participants (< 50 years) compared with older (> 50 years), females compared with males, and Caucasians compared with non-Caucasians.
Correlation Between Participant and Investigators’ Assessments
It is clear from the assessments of benefit, that there was good agreement between the investigators’ ratings and those of the participants. That is there was good agreement between the assessment made by the investigators comparing appearance with a battery of reference photographs (static scaling) and the assessment of the participant relying on memory of their appearance at the start of the study (dynamic scaling).
The level of this agreement between the value on the participants’ 9-point scale and the investigators’ assessment at Day 30 was calculated. The correlation coefficient (rho) was 0.645, (p<0.001, n = 355, Spearman's correlation). A similar correlation was found for the assessments of all participants at all visits (rho = 0.645, p<0.001, n = 1417).
During the course of this study a number of participants reported at least one adverse event. The most common adverse events were headache and nasopharyngitis. The incidence of events, occurring in more than 5% of the participant population in any treatment group, is shown in Table 3 below.
(N = 94)
(N = 90)
(N = 95)
(N = 94)
|Any adverse event||52 (55.3%)||53 (58.9%)||64 (67.4%)||52 (55.3%)|
|Headache||10 (10.6%)||15 (16.7%)||19 (20.0%)||13 (13.8%)|
|Nasopharyngitis||8 (8.5%)||9 (10.0%)||8 (8.4%)||7 (7.4%)|
|Blood cholesterol increased||7 (7.4%)||5 (5.6%)||7 (7.4%)||5 (5.3%)|
|Back pain||5 (5.3%)||1 (1.1%)||4 (4.2%)||1 (1.1%)|
Following treatment, mild ptosis was reported in three participants (0.8 %) at Day 7. One participant (1 event) was in the group that received 50 units Reloxin and 2 participants (3 events) received 75 units. There was no evidence of any ptosis in any of the participants in the groups treated with placebo or 20 units Reloxin.
Four serious adverse events were reported, all of which were assessed as being unrelated to treatment. In the group who received 20 units Reloxin, one participant developed diverticulitis and one reported an unintended pregnancy. In the group receiving 50 units Reloxin, one participant reported an unintended pregnancy. Among the participants receiving 75 units Reloxin, one participant developed dehydration. No deaths were reported and no adverse event led to withdrawal of a participant from the study.
The overall incidence of adverse experiences for all Reloxin treatment groups was similar to that in the placebo-treated group.
There was no evidence of neutralizing antibodies in any participant at baseline or at follow-up.
Glabellar wrinkles may arise as a result of overactivity of the underlying corrugator supercilii, procerus and orbicularis oculi muscles. Their appearance is often unwelcome, especially in younger adults, but also in older subjects where they can create an impression of greater age, or the unintentional appearance of anger or worry.
A number of treatments are available to reduce the severity of these wrinkles (e.g. injections of collagen, silicone oil, or autologous fat; browlifts; surgical ligation; resurfacing). However, none of these treatments addresses the underlying problem of muscle overactivity.
Botulinum neurotoxin toxin type-A (BoNT-A) acts by chemodenervation of the underlying muscles and has been used to treat the severity of wrinkles since 1992 (24).
Many of the early studies were performed using the American botulinum toxin type-A (Botox®) (11; 25-27), but there have also been publications reporting the use of the UK product Dysport (Reloxin)(17; 22; 23; 28). In these studies the doses varied between 40 and 100 units Dysport. Data from the published studies also show that Dysport (Reloxin) is effective in reducing the appearance of wrinkles, although none of these investigations was placebo-controlled. The number of injection points also varied in these investigations, from two (17) to seven (22).
The aim of this prospective study was to determine the optimal dose of Reloxin in reducing the severity of glabellar lines, with a view to using this dose in subsequent Phase 3 trials.
In this study, all doses of Reloxin resulted in a statistically significantly better response in the appearance of glabellar frown lines compared with placebo. This effect was reported by both the investigators and the participants when evaluating glabellar lines at maximum frown or at rest.
The response was first observed at Day 7 and persisted throughout the study. This difference from placebo treatment was observed at all time points for the participants receiving Reloxin at 50 or 75 units, but in those receiving Reloxin at 20 units, the effect remained apparent at Day 90 but not at Day 120. In terms of the number of participants responding to treatment, 50 units Reloxin was found to be as effective as 75 units Reloxin, and a similar duration was observed in these two treatment groups.
The duration of benefit in this study was not formally captured between assessments. At 90 days after treatment, a large number of participants continued to show a response (47% at the highest dose, based on investigators’ assessments and 72% based on participants’ assessments). In a number of participants, this response continued to be observed at Day 120 (27% and 51%; investigators’ and participants’ assessments respectively) . This finding is similar to that reported in other publications. Ascher et al. (23) found that the mean duration of effect was 3.4 months after the injection of 40 units Dysport (Reloxin) into the glabellar region. Le Louarn (17; 29) also reported retreatment every 3 – 4 months, although this was extended to 7 – 10 months with later treatments, to avoid visible muscle atrophy.
In our study, we observed greater benefit in women than in men, which has also been reported by Pribitkin et al. (14). However, unlike our findings, these authors found no difference in the benefit in people of different ages.
The majority of subjects enrolled in our study were of Caucasian origin, with the second largest group being Hispanic. There was some evidence of greater perceived benefit in the Caucasian participants, although botulinum toxin treatment is also clearly effective in people of Hispanic origin (30). There are some suggestions that there may be differences in skin thickness between people of Asian origin compared with Caucasians (31), although there may also be a difference in the perception of the youthfulness in the Asian face (32). The small numbers of participants of non-Caucasian origin in our study mean that sub-group analyses could not be performed.
All doses of Reloxin were well tolerated in this study, although ptosis was observed in three participants (0.8 %). Ptosis has been reported by other investigators using botulinum toxin in the treatment of facial wrinkles, but Reloxin (Dysport) was not the product used in these studies. They found an incidence ranging from below 0.5% (33), to 3% (34). Feller et al. reported no ptosis in their study involving Dysport (Reloxin), at doses up to 70 units (22) and Ascher et al. found no ptosis when using Dysport (Reloxin) at doses up to 75 units (35)
The most commonly reported adverse event was headache, which occurred with a similar frequency in all treatment groups, including placebo. This was also observed by Feller et al. (22) who commented that this effect probably resulted from the injected volume exerting a greater pressure in the glabellar region.
From the data obtained in our study, the 50 unit dose was identified as the optimal dose with respect to efficacy, duration and safety profile. It is planned that this dose will be used for rigorous testing in Phase 3 trials.