Long‐term efficacy and safety of a hyaluronic acid dermal filler based on Tri‐Hyal technology on restoration of midface volume

Art Filler Volume (AFV) is a hyaluronic acid (HA)‐based filler formulated with “Tri‐Hyal” technology, a unique combination of three sizes of HA chains. This study assessed AFV efficacy and safety over 18 months when used to restore midface volume.


| INTRODUC TI ON
Age-related aesthetic changes to the face, which generally begin in a person's late 30s, may adversely impact other people's perception of the subject as well as undermining the person's self-esteem, body image, and confidence. 1,2 These changes arise from reductions in the thickness of the skin, and in the volume of bone, muscle, and superficial and deep fat pads. 1 Total and dermal collagen, and elastin content also decline. 3 The loss of volume tends to be most marked in the deeper layers, following age-related reductions in subcutaneous fat and bone. 3 The cumulative effect of these age-related changes can profoundly affect facial appearance. 1,[3][4][5] For instance, the loss of volume can result in: reduced midface projection, such as a diminution in the prominence and width of the malar eminence; lead to a heartshaped face becoming rectangular or pear-shaped; and deepening of the nasolabial fold. 1,3-5 A crescent-shaped hollow beneath the lower edge of the orbicularis oculi muscle, lowering of the malar fat pad and cheek skin, which results in a visually apparent segmentation of facial compartments, and reduced jaw angle are also common agerelated changes. 1,[3][4][5] Subjects may also show age-related dermatological alterations, including loss of skin elasticity, increased skin roughness, and xerosis. 1,[3][4][5] Injectable hyaluronic acid (HA) fillers restore facial volume, which reduces the appearance of sagging skin and skin folds. 1,[5][6][7][8][9][10] Art Filler® Volume is a high G' HA filler with good cohesivity which can be injected in the hypodermis or supra-periosteal for a volumiz-

| Treatment
AFV was injected bilaterally into the cheeks. A maximum of 1.8 mL AFV was injected into each cheek (total 3.6 mL) on Day 0 (D0).
Subjects were evaluated at D21 when, if deemed necessary by the subject and clinician, a retouch could be performed using a maximum of 1.2 mL AFV per cheek. No further reinjections into the cheeks were permitted during the study. Subjects were evaluated at D42, D90 (3 months), D180 (6 months), D270 (9 months), D360 (12 months), D450 (15 months), and D540 (18 months). Figure 1 summarizes the study design.

| Assessment
The primary assessment was the change in volume of each treated cheek (right and left) overall on D21 after the first AFV injection based on the Medicis Midface Volume Scale (MMVS) score. 11 Each side was considered as its own control and there was no comparison between the two cheeks, even if there was an asymmetry. The MMVS score rates midfacial appearance using a 4-point scale: (1) fairly full midface, cheek prominence projected beyond the infraorbital rim at 45° view; (2) mild loss of fullness in the midface area, flatness of the midface, cheek prominence at or behind infraorbital rim; (3) moderate loss of fullness with slight hollowing below malar prominence, the presence of the nasojugal groove extending past the mideye; (4) substantial loss of fullness in the midface area, clearly apparent hollowing below the malar promence. 11 "Satisfactory" volume restoration was defined as a reduction of ≥1 point compared with baseline in the MMVS score. 11 If a retouch was deemed necessary by the subject and clinician on D21, the subject was also assessed on D42. The persistence of the correction was evaluated at D90, D180, D270, D360, D450, and D540.
Local and general adverse events were recorded at each study visit. Investigators and subjects assessed outcomes on the Global Aesthetic Improvement Scale (GAIS) at each clinic visit. Subjects also completed a validated self-esteem questionnaire (Echelle Toulousaine de l'Estime de Soi [ETES]), at baseline, at D180, and at D360.

| Ethical approval
Ethical approval was provided by the committee of "Île-de-France VI Comité de protection des personnes," at Pitié-Salpêtrière Hospital, Paris. Subjects provided informed consent for the study and use of images. The study was conducted in accordance with ISO 14155:2011.

| Inclusion and noninclusion criteria
The study enrolled males or females aged 19 years or older, with a Fitzpatrick Phototype of I-IV and a MMVS score of 3 or 4, who did not plan to undergo a facial cosmetic procedure during the whole study.
Noninclusion criteria included a lifetime history of facial injections, implants, or treatments that employed nonabsorbable fillers or a skin-retaining device on the face, such as mesh, gold wire, liquid silicone, or other particulate material. The subjects could not have had any corrective facial aesthetic injections (botulinum toxin or HA facial filler) in the cheeks for the last 12 months. Subjects who underwent medical treatment by laser, ultrasound, deep chemical peeling, or facial dermabrasion in the last 3 months or who planned to undergo such treatment during the study, people with a history of hypertrophic, keloid or dyschromic scarring, multiple severe allergies or anaphylactic shock, or hypersensitivity to any ingredient were not included.

| Statistical analyses
Satisfactory volume restoration was defined as an improvement of ≥1 point on the MMVS 11 compared with baseline, assessed separately for the right and left cheeks and overall (both cheeks). A satisfactory volume correction rate of 95% was expected at D21 (D42 for subjects who received a second injection). To detect this rate with an accuracy of 5% (95% confidence interval), 73 cheeks would need to be injected. Therefore, the study aimed to assess at least 100 cheeks.
The MMVS score of each injected area, measured at D21 (3 weeks) or D42 (6 weeks, if retouched at 3 weeks) after the first injection, defined the optimal aesthetic score for a given target area. The maintenance of this score for treated areas was calculated at D90, D180, D270, D360, D450, and D540 without any reinjection according to the difference in the MMVS score be-  was assessed using the Wilcoxon test. Analyses were repeated using the PPP.
The GAIS score has 7 levels (from +3 for very strongly improved to −3 for very strongly worsened) and was assessed at each time point for each cheek and overall, on both sides. This analysis was performed in the PPP. Changes in the ETES score at D180 and D360 were calculated using subjects with these values at baseline and the given time point.

| Demographics
The study enrolled 79 healthy Caucasian subjects, 7 males and 72 females, with a mean age of 54.8 ± 9.7 years. The study ran between September 2016 and July 2018. Table 1 summarizes the baseline demographics and MMVS scores.

| Volume injected
The mean volume for the first injection was 1.1 ± 0.4 mL per side (maximum authorized volume was 1.8 mL per side). Twenty-five midface zones required a second injection at D21 (14 injections on the left and 11 injections on the right side), a total of 16% across 158 midface zone injected. The mean total volume injected during the study was 1.1 ± 0.5 mL per side at baseline and 0.7 mL for reinjection at D21. The maximum injected volume was 2.4 per side.

| Initial efficacy
In the ITT population, mean MMVS scores ± SD for the left cheek improved significantly from 3.2 ± 0.4 at baseline to 1.8 ± 0.7 at D21 and 1.7 ± 0.6 at D42, including those that received a retouch (p < 0.0001). For the right cheek, the mean score improved significantly from 3.2 ± 0.4 at baseline to 1.8 ± 0.6 at D21 and 1.7 ± 0.6 at D42 (p < 0.0001). The mean overall score improved significantly from 3.2 ± 0.4 to 1.8 ± 0.6 and 1.7 ± 0.6 at D21 and D42, respectively

| Persistence
The persistence rate was calculated based on the maintenance of success rate at each time point compared to D21/D42. This parameter revealed 11% in all ITT analyses (left, right, and both sides) at 18 months (D540) (data not shown). In addition, we revealed that persistence was almost the same between groups without or with touch up at D21 or the global population (both groups) in the PPP, 22% (N = 51), 33% (N = 15), and 24% (N = 66), respectively ( Figure 3B). An example of before and after treatment images is shown in Figure 4.
According to the GAIS evaluated by subjects and investigators, the proportion rating facial fullness as "improved" or "much" or "very much improved" was 90% at D21 (100% at D42 for the retouched group) and 68% at D540 (18 months) ( Tables 2 and 3). ETES score at D180 and D360 showed no significant changes compared with baseline.

| DISCUSS ION
This study provides evidence that a HA filler based on Tri-Hyal technology is effective and well tolerated when used for facial midface volume restoration. These data are similar to published data. 1,[5][6][7][8][9][10] At D21, more than 90% of those who received AFV showed an im- with no difference between the two sides of the face. These data demonstrate that more than 80% of the subjects had a "fairly full midface" or only "mild loss of fullness" after treatment while they all showed "moderate to severe loss of fullness" before the treatment.
A small proportion of patients (6% at D21 or D42 in the retouch group) were still rated as demonstrating "moderate loss of fullness" after treatment. This is due to partial correction of the patients with Other studies confirm the long-term ability of HA fillers (including Juvéderm® Voluma®, Belotero® Volume and Emervel HA) to restore facial volume loss and the published data are similar to the data obtained in this study. 1,7,13 One study reported that 3 months and 2 years after HA treatment, 92.8% and 79.0% of subjects, respectively, rated their cheek volume as "improved" or "much improved" on the GAIS; in this study, at 6 months, the investigator rating was 96% and subject rating 86%, while at D540 both were 68%. 1 In another study, a comparison of two HA fillers (Belotero® Volume In the current study 61% and 22% subjects showed ≥1 grade improvement in the MMVS score at 12 and 18 months; this difference may be due to differences in average injection volumes, which were 1.1 mL in this study and 2 mL in the previous study. 7 Moreover, 78% and 65% of the subjects in the same study showed an improvement on the GAIS at 18 months which is comparable to the 68% observed in the present study. 7 TA B L E 3 Evolution of the Global Aesthetic Improvement Scale performed by the subject for both cheeks.  (14) 18 (13)  HA dermal fillers. 12 The improvement was apparent in several facial areas, including 66.6% in the cheeks and 58.6% improvement in the cheekbones. The volume of injection was also higher in this study; the mean volume injected was 7.4 mL for the full face, 2.2 mL in the cheeks, and 1.8 mL in the cheekbones. 13 Other studies have explored the parallels between assessment of the longevity of clinical outcomes and histological durability of HA fillers in tissues. 14 Plus) has been shown to display greater correction longevity after injection compared to a monophasic HA filler (Perlane®). 16 Although the different classes of HA filler behave differently immediately after their injection into the dermis, their histological patterns are predictable. 17 The intrinsic physicochemical properties of HA fillers, associated with different manufacturing technologies, may also affect HA filler longevity in the dermis.
AFV was well tolerated, which again is consistent with other studies assessing HA fillers used to restore facial volume. 1,7,13 In this study, the investigators respected the "less is more" concept to avoid high volume injections which directly influence the inflammatory reactions of highly cross-linked fillers. 18 The most common adverse events associated with HA injections include erythema, swelling, bruising, tenderness, firmness, and lumps and bumps. 1,8,13 Most adverse events in previous studies were mild to moderate and lasted less than 2 weeks, 1 which is similar to the results of this study.
The study had certain limitations. The study was a post market study aimed to mirror naturalistic practice and blinding investigators and subjects would be difficult. The study was, therefore,  [19][20][21][22][23] Facial rejuvenation and aesthetic interventions need to be racially sensitive. 22 Aesthetic outcomes depend on the expertise of the treating practitioner, but the numbers of subjects in each center were inadequate to discern any difference between sites or practitioners.

| CON CLUS ION
AFV injections produced ≥1 grade improvement in MMVS scores in almost all subjects in the right and left cheeks and overall (both cheeks) after 3 weeks. AFV shows a sustained objective and subjective benefit that persists in some subjects for at least 18 months, often without retouching. The sustained improvement was consistent in both cheeks, and across multiple assessment methods, suggesting the findings are robust and clinically meaningful. AFV was well tolerated. These findings suggest that AFV is a safe and effective filler to rejuvenate cheek volume.

ACK N OWLED G M ENTS
The authors wish to show their appreciation to the project managers

DATA AVA I L A B I L I T Y S TAT E M E N T
Research data are not shared.

E TH I C S S TATEM ENT
Ethical approval was provided by the committee of "Île-de-France VI Comité de protection des personnes", at Pitié-Salpêtrière Hospital, Paris. Subjects provided informed consent for the study and use of images. The study was conducted in accordance with ISO 14155:2011.