ethnicity, age, body mass index (BMI), and Fitzpatrick skin type were made using the AT (as treated) population and tested for significance, using P<0.05 as the cutoff level. Efficacy was evaluated by the LFGS, Perioral Lines at Rest Severity Scale (POL),5 pGAI scale (Table 1), and (4) iGAI scale (Table 1). The POL is a validated 4-point scale for the most severe perioral line with the patient’s mouth at rest.
The LFGS was also used to assess treatment efficacy in the upper and lower lips separately at each visit.
Primary Efficacy
The primary efficacy endpoint was change from baseline to day 56 in overall LFGS with both lips together. The goal was to disprove the null hypothesis, that test device was inferior to comparator by more than 0.50 units. Two-sided hypothesis testing was conducted for all inferential analyses with P-values <0.05 considered statistically significant.
It was assumed that the LFGS data were normally distributed. The 95% confidence interval (CI) for the difference in mean values between the treatment groups (test device minus comparator) was calculated. If the lower limit of the CI was above the prespecified non-inferiority limit of -0.50, the null hypothesis was rejected to support the claim that test device was non-inferior to comparator. These calculations were performed for both the PP and mITT populations. The results for the PP population were considered as definitive and those for the mITT population as supportive.
Secondary Efficacy
Endpoints were (1) percent of responders with at least a 1-grade increase from baseline on the overall LFGS of both lips together on day 56, (2) percent of responders that achieved ≥1-point improvement [decrease in severity] from baseline on the overall POL severity scale of both lips together on day 84, (3) the change from baseline to day 84 in overall LFGS of both lips together, and (4) change from baseline to day 168 in overall LFGS of both lips together.
Other efficacy variables were (1) pGAI, iGAI, and swelling at each visit, (2) percent of patients with treatment success (upper and lower lips separately) at day 56 where responders had ≥1-grade increase from baseline on the LFGS post augmentation, (3) percent of responders (upper and lower lips separately) on the POL severity scale at day 84, where responders had ≥1-point improvement from baseline, and (4) patient satisfaction with the lips visual analog scale (0 = very unsatisfied to 100 = very satisfied).
Safety
For the AT population, treatment-emergent AEs (TEAEs), serious AEs, and AEs of special interest (vision changes or AEs events due to embolic or ischemic cause) were monitored. Other safety assessments included lip function, sensation, texture, firmness, symmetry, and movement/function, all evaluated by the blinded evaluating investigator before injection and at follow-up visits. Lip function was assessed by the patient’s ability to sip liquid through a straw. Lip sensation was assessed by (1) the monofilament test, a patient’s ability to feel the sensation of a 0.4G monofilament at 3 points on the upper lip and 3 points on the lower lip; and (2) the cotton wisp test, a patient’s ability to feel the sensation of a cotton wisp at 3 points on the upper lip and 3 points on the lower lip. Lip texture, firmness, and symmetry were assessed as normal or abnormal.
Lip movement/function was evaluated by the patient’s ability to pucker lips, blow, and pronounce words that began with the letter “W” such as water, work, week, and wind.
Ease of Use
Both devices were evaluated for ease of use (0 = not easy, 10 = most easy).
The LFGS was also used to assess treatment efficacy in the upper and lower lips separately at each visit.
Primary Efficacy
The primary efficacy endpoint was change from baseline to day 56 in overall LFGS with both lips together. The goal was to disprove the null hypothesis, that test device was inferior to comparator by more than 0.50 units. Two-sided hypothesis testing was conducted for all inferential analyses with P-values <0.05 considered statistically significant.
It was assumed that the LFGS data were normally distributed. The 95% confidence interval (CI) for the difference in mean values between the treatment groups (test device minus comparator) was calculated. If the lower limit of the CI was above the prespecified non-inferiority limit of -0.50, the null hypothesis was rejected to support the claim that test device was non-inferior to comparator. These calculations were performed for both the PP and mITT populations. The results for the PP population were considered as definitive and those for the mITT population as supportive.
Secondary Efficacy
Endpoints were (1) percent of responders with at least a 1-grade increase from baseline on the overall LFGS of both lips together on day 56, (2) percent of responders that achieved ≥1-point improvement [decrease in severity] from baseline on the overall POL severity scale of both lips together on day 84, (3) the change from baseline to day 84 in overall LFGS of both lips together, and (4) change from baseline to day 168 in overall LFGS of both lips together.
Other efficacy variables were (1) pGAI, iGAI, and swelling at each visit, (2) percent of patients with treatment success (upper and lower lips separately) at day 56 where responders had ≥1-grade increase from baseline on the LFGS post augmentation, (3) percent of responders (upper and lower lips separately) on the POL severity scale at day 84, where responders had ≥1-point improvement from baseline, and (4) patient satisfaction with the lips visual analog scale (0 = very unsatisfied to 100 = very satisfied).
Safety
For the AT population, treatment-emergent AEs (TEAEs), serious AEs, and AEs of special interest (vision changes or AEs events due to embolic or ischemic cause) were monitored. Other safety assessments included lip function, sensation, texture, firmness, symmetry, and movement/function, all evaluated by the blinded evaluating investigator before injection and at follow-up visits. Lip function was assessed by the patient’s ability to sip liquid through a straw. Lip sensation was assessed by (1) the monofilament test, a patient’s ability to feel the sensation of a 0.4G monofilament at 3 points on the upper lip and 3 points on the lower lip; and (2) the cotton wisp test, a patient’s ability to feel the sensation of a cotton wisp at 3 points on the upper lip and 3 points on the lower lip. Lip texture, firmness, and symmetry were assessed as normal or abnormal.
Lip movement/function was evaluated by the patient’s ability to pucker lips, blow, and pronounce words that began with the letter “W” such as water, work, week, and wind.
Ease of Use
Both devices were evaluated for ease of use (0 = not easy, 10 = most easy).
RESULTS
Of the 158 randomized patients, 141 (89.2%) completed the study. The most frequent reasons for discontinuation were withdrawal of consent in the test device group (n = 6, 7.5%) and lost to follow-up in the comparator group (n = 7, 9.0%).
Demographics
Demographics of the test device and comparator groups were compared using the AT population. Comparisons for gender, ethnicity, age, body mass index (BMI), and Fitzpatrick skin type showed that differences between the two groups were not significant except for ethnicity (P=0.044) and age (P=0.048), which were of borderline significance.
Primary Efficacy
In the PP population, the mean change from baseline to day 56 in overall LFGS with both lips together was 1.52 for the test device and 1.53 for comparator (Figure 1). The lower limit of the 95% CI for test device minus comparator was -0.33, which is higher than the pre-specified non-inferiority margin of -0.50.
In the mITT population, the lower limit of the 95% CI test device minus comparator was -0.26, which is also higher than -0.50.
Secondary Efficacy
In the following comparisons, numerical results are limited to those of the PP population due to spatial constraints. Results for the mITT population were similar in all cases.
The percent of PP patients with at least a 1-grade increase from baseline on the overall LFGS of both lips together on day 56 was 90.7% with test device and 92.7% with comparator.
Demographics
Demographics of the test device and comparator groups were compared using the AT population. Comparisons for gender, ethnicity, age, body mass index (BMI), and Fitzpatrick skin type showed that differences between the two groups were not significant except for ethnicity (P=0.044) and age (P=0.048), which were of borderline significance.
Primary Efficacy
In the PP population, the mean change from baseline to day 56 in overall LFGS with both lips together was 1.52 for the test device and 1.53 for comparator (Figure 1). The lower limit of the 95% CI for test device minus comparator was -0.33, which is higher than the pre-specified non-inferiority margin of -0.50.
In the mITT population, the lower limit of the 95% CI test device minus comparator was -0.26, which is also higher than -0.50.
Secondary Efficacy
In the following comparisons, numerical results are limited to those of the PP population due to spatial constraints. Results for the mITT population were similar in all cases.
The percent of PP patients with at least a 1-grade increase from baseline on the overall LFGS of both lips together on day 56 was 90.7% with test device and 92.7% with comparator.
