Efficacy and Safety of a Hyaluronic Acid–Containing Cream in the Treatment of Chronic, Venous, or Mixed-Origin Leg Ulcers: A Prospective, Multicenter Randomized Controlled Trial

Chronic venous insufficiency is a condition that occurs when the venous wall and/or valves in the veins of the legs are not working effectively, making it difficult for blood to return to the heart from the legs. It commonly results in aching or pain in the lower extremities, swelling, a feeling of heaviness, and skin changes, with chronic ulceration as the most severe manifestation.^1-3 Prevalence estimates of chronic venous insufficiency vary widely, but the condition is one of the most common vascular diseases in Western populations, affecting as many as 40% of women and 17% of men, with open or healed venous leg ulcers present in more than 1% of the population.⁴ The socioeconomic effect of leg ulcers is major; the individual’s ability to participate in social and occupational activities is impaired and there is a substantial financial burden on individuals, families, caregivers, and the health care system.^2,3,5

Most chronic leg ulcers are of venous origin. However, an increasing proportion of ulcers are of mixed (arterial and venous) origin and are associated with less frequent causes, including infections, diabetes mellitus, and vasculitis.^2,6 The overall prognosis of venous leg ulcers is relatively poor, marked by a high frequency of delayed healing and recurrence.¹ Leg ulcer-related disability is likely to result in loss of productivity of subjects in the workforce and may necessitate early retirement.¹ Venous ulcers are most often seen in women and are seldom seen in men or women younger than 60 years. Chronic leg ulcers that do not heal within 1 year, despite optimal therapy, are considered to be therapy resistant.²

Appropriate wound and skin management of both venous and mixed-origin leg ulcers is important in order to cleanse and debride the wound to remove dead skin, fibrin remnants, and any other debris that may impede the healing process.² Appropriate dressings used under compression bandages help to promote faster healing and prevent adherence of the bandage to the ulcer. However, no specific type of dressing has been shown to be superior.

Hyaluronic acid (hyaluronan, HA), a large, nonsulfated mucopolysaccharide of the class of glycosaminoglycans (GAGs) and an essential component of the extracellular matrix of the skin and other connective tissues, is a linear polysaccharide consisting of repetitive chains of disaccharide units of N-acetyl-D-glucosamine and D-glucuronate.⁷ The specific hygroscopic, rheologic, and viscoelastic properties of HA, which impart critical chemical and physical characteristics to the extracellular matrix, are responsible for the role of HA in tissue hydration⁷ and underpin the use of HA as a topical cream or cream-impregnated gauze as a rational and effective therapeutic approach to treat subjects with chronic wounds. Endogenous HA is generally complexed with other GAGs, collagen, and elastin, binding with water into a large viscoelastic macromolecular aggregate that forms a viscous complex that encapsulates cells, fills open spaces in the fiber network, and lubricates the moveable tissue layers, stabilizing intracellular structures and protecting the tissue against mechanical stress.^7-9 Hyaluronic acid plays a complex and important role at all stages of the wound healing process, including organization of the granulation tissue matrix, cell migration and proliferation, modulation of the inflammatory response, reepithelialization, and angiogenesis.^7-13

A cream containing 0.2% HA as an active ingredient has been developed and marketed for the topical treatment of noninfected wounds, (exuding or superinfected), including leg ulcers for more than 20 years. The 0.2% HA product (ialuset cream*) is biocompatible and creates a moist environment around the wound that promotes optimal conditions for healing and supports the reepithelialization process. An HA gauze pad preparation is also available.

The value of topical application of HA in improving the wound-healing process and reducing time to healing has been substantiated in in vitro, in vivo, and clinical studies.^12-19 Cream and gauze dressing formulations of HA are as well tolerated as reference or neutral comparators, are highly rated for ease of use and treatment satisfaction by both subjects and practitioners, and have been shown to be effective in treating subjects with venous or mixed-origin leg ulcers.^14-20 Although the topical application of HA is a therapy approach routinely used to promote tissue healing in the management of chronic wounds, data to base clinical decisions on the use of HA in the healing of chronic leg ulcers of venous or mixed origin are mostly limited to short-term studies.²¹

This study describes the results of a randomized controlled trial that compared the use of an HA cream with that of a neutral comparator cream in the treatment of subjects with chronic leg ulcers of venous or mixed (venous and arterial) origin. A parallel study evaluated the use of an HA–containing gauze pad in a similar subject population and will be reported separately.

Design and objectives

This was a confirmatory, prospective, multicenter, multinational, parallel-group, randomized, double-blind clinical investigation comparing the HA cream (0.2% HA) and a neutral comparator. The objective of the study was to confirm the superiority of a local application of HA cream as compared to a neutral comparator cream in the treatment of subjects with chronic leg ulcers of venous or mixed (venous and arterial) origin.

The trial was conducted in accordance with the Declaration of Helsinki and its modifications, the rules of Good Clinical Practice, and the International Organization for Standardization (EN) 14155 requirements as well as conformed to the Consolidated Standards of Reporting Trials (CONSORT) 2010 extension for randomized controlled trials. Approval for the protocol and protocol amendments was obtained from the Bioethics Committee of the Regional Medical Chamber, Lodz (Poland), and written informed consent was obtained from all subjects before starting the trial.

Project management and monitoring of the study were carried out by the local contract research organization, CROMSOURCE, in Poland.

Subject population

Adult males and females 18 years or older were eligible for inclusion in the study if the following criteria were met: one or more chronic leg ulcers of venous (varicose or postthrombotic) or mixed (venous and arterial) origin with a predominance of venous origin, with a duration of at least 2 months but less than 4 years; the target ulcer area was at least 5 cm² but no more than 40 cm²,with less than 50% of necrotic tissue; an arteriovenous Doppler examination showing superficial or profound venous reflux; and/or a well-documented history of deep venous thrombosis and/or clinical evidence of postthrombotic syndrome with chronic edema and lipodermatosclerosis. Other inclusion criteria included daily use of an appropriate compression system, albuminemia of 25 g/L or higher, and a negative pregnancy test together with the use of reliable contraception for women of childbearing age.

The main exclusion criteria were as follows: the presence of a nonvascular ulcer or a general cause (eg, hematologic cause), ankle-brachial index less than 0.8 or higher than 1.2 and/or dominant significant arterial insufficiency, diabetes mellitus of any type, hepatic or renal failure, presence of wound infection, an ulcer with exposed tendon or bone or due to malignancy, a recent history of venous thrombosis, and ongoing treatment with drugs known to adversely affect the healing process.

Study protocol

Subjects were randomized in a 1:1 ratio using a centralized randomization system to have an investigator-selected target ulcer treated once daily locally with either a cream containing HA 0.2% or a neutral comparator cream for a maximum period of 20 weeks (active treatment period) or until complete ulcer healing was achieved, whichever occurred first. The trial was stratified on target ulcer size (≤ 20 cm² or > 20 cm²).

The active treatment (HA cream) contained HA 0.2% intended for topical use and was supplied for the study in 100-g tubes. The neutral comparator cream contained the same ingredients, with the exception of HA, and had visual and physical characteristics identical to those of the active cream.

The randomization list was prepared using validated software (SAS Institute Inc) by the Data Management and Statistics Department of the sponsor and stored in electronic form in a secure directory to ensure confidentiality in full respect of standard operating procedures. The study was double-blind; treatment allocation was kept hidden to the subject, investigator, sponsor, contract research organization team, and central assessor. Strict procedures were followed throughout the study to maintain blinding.

Treatments under scope were used in conjunction with standard local therapy for both groups, with standard of care as recommended by the French National Health Authority²²; that is, sterile saline was used before each application to clean the target ulcer, together with the use of surgical debridement or local anesthesia if necessary. Wound debridement, dressing, and optimized compression were only applied by experienced personnel with knowledge of the assessment and treatment of subjects with leg ulcers. Subjects with nontarget ulcers were treated according to the same standard-of-care procedures. If clinical evidence of infection was present, therapy with systemic antibiotics was considered. However, the use of topical antimicrobial drugs and antiseptic agents was prohibited.

After wound cleansing, the HA cream or neutral comparator cream was applied directly to the target ulcer once daily by the study nurse or authorized study personnel either at the subject's home or clinic. Following cream application, the wound area was covered with a sterile gauze dressing and an appropriate long-stretch graduated elastic bandage with stirrup (BIFLEX 16+ PRACTIC bandage), which was provided to all sites; this regimen was applied to all subjects according to the standard of care. Adapted and efficient compression was required to be worn daily during the entire investigation period. The compression was set up before subjects arose in the morning and was removed in the evening prior to going to bed. Its daily usage was confirmed by the study nurse when visiting the subjects at home or by the study personnel at each in-clinic visit. Standard of care (compression socks or compression bandages) was continued for subjects discontinuing the trial.

Medications or therapies prohibited during the entire trial period included treatment known to delay the healing process (eg, high-dose systemic corticosteroids, cytostatic drugs, immunosuppressants), the use of proteolytic enzymes for wound debridement, other topical treatments of the target ulcer, vein surgery of the affected leg, and any other systemic or general therapy that could positively or negatively influence the healing of the target ulcer.

Data collection

A total of 8 visits were carried out by the same investigator for each subject, consisting of screening/inclusion (visit 0); randomization and first treatment (visit 1); weeks 4, 8, 12, 16, and 20 of treatment (visits 2–6); and a final evaluation (follow-up) visit (visit 7) at day 162 (week 23) or 3 weeks after complete healing if it occurred before week 20. An unscheduled visit to confirm healing could be arranged by request of the study nurse if that person thought complete healing had been achieved between scheduled visits. The healing status of the target ulcer was assessed and recorded at each study visit after treatment was initiated. The target ulcer was evaluated by the investigator and documented using standardized digital photography after cleansing and wound preparation were completed at randomization and before the Investigational Medical Device (IMD) was applied at each study visit and at the follow-up visit. During each evaluation visit, the aspect of periulcerous skin was assessed by the investigator, and evidence of edema, purpura, erythema, oozing, and/or maceration was identified on a four-grade scale of intensity: none (0), mild (1), moderate (2), and severe (3).

Adherence with the study treatment, defined as the number (%) of daily applications of the IMD as compared with the theoretical number, was checked at each visit during the treatment phase. This was based on the subject's and nurse’s records in study-dedicated diaries in which daily medication renewal, compression bandage, and any oral analgesic use were recorded.

Primary and secondary endpoints

The primary efficacy endpoint was ulcer healing, defined as 100% reepithelialization of the wound area at 20 weeks or at any earlier visit if healing occurred before week 20, as evaluated by the central blinded assessor and confirmed 3 weeks after initial healing achievement. The clinical assessment of the primary efficacy variable was centralized and independently performed by an experienced, blinded assessor judging via standardized photography of the target ulcer.

Secondary efficacy endpoints included the percentage of completely healed target ulcers as assessed by the investigator, and at all other scheduled study visits as assessed by the central blinded assessor, the target ulcer area relative to baseline at each study visit, the condition of the periulcerous skin, the total amount of analgesics used, the rate of infection of the target ulcer, the adherence with treatment, the time to achieve complete healing as centrally assessed, and pain intensity as self-assessed by the subject on a visual analog scale (VAS) at each study visit.

Safety endpoints

The incidence of adverse events (AEs) and treatment-emergent AEs (TEAEs) occurring at any time during the trial, regardless of relationship to the IMD, were recorded, together with vital signs at each study visit, and physical examination at screening, randomization, and final visit. Adverse events and serious adverse events (SAEs) were coded using the Medical Dictionary for Regulatory Activities (MedDRA) version 20.1. The frequency of TEAEs, as well as serious TEAEs, and the proportion of subjects experiencing TEAEs, were calculated.

Statistical analysis

It was estimated that the percentage of subjects expected to have completely healed target ulcers at week 20 or the earliest visit when healing was complete and confirmed after 3 weeks was 43% for the HA cream group and 20% for the neutral comparator group. Therefore, a sample size of 84 evaluable subjects in each group (168 in total) was calculated to have greater than 90% power to detect a difference of 23% between the HA cream group and the neutral comparator group in terms of successes, according to the assessment of the central blinded assessor judging in favor of the HA cream group (superiority study), when using a χ² test with a two-sided α set to equal .05.

The proportion of completely healed target ulcers in the primary analysis was compared between treatment groups by means of the Cochran-Mantel-Haenszel χ² test, considering target ulcer size at baseline (≤ 20 cm² or > 20 cm²) as a stratification factor. For the safety analysis, the difference between the proportions of subjects in treatment groups was analyzed using the χ² test.

Twenty centers in Poland experienced in the assessment and management of chronic leg ulcers participated in the study between June 13, 2017, and April 17, 2019. A total of 199 European subjects were screened, and 168 (85 in the HA cream group and 83 in the neutral cream group) were eventually enrolled in the study and received at least 1 application of the IMD (safety analysis set). Of these subjects, 164 were included in the full analysis set (FAS; all subjects in the safety analysis set who had at least 1 postbaseline efficacy assessment), with 83 in the HA cream group and 81 in the neutral cream group.

Almost all subjects (96.4%) were receiving other concomitant medications, most commonly bioflavonoids (33.3%), platelet aggregation inhibitors (32.7%), and angiotensin-converting enzyme inhibitors (32.1%). The most frequently used medications were acetylsalicylic acid (26.8%), spironolactone (22.0%), and pentoxifylline (21.4%).

The demographic and target ulcer characteristics of subjects overall were similar between groups (Table 1).

A total of 144 subjects (HA cream group, 70; neutral cream group, 74) completed the study. Twenty-six subjects (HA cream group, n =17; neutral cream group, n =9) discontinued treatment. Of those, 14 subjects (HA cream group, n =9; neutral cream group, n =5) withdrew from the study for personal reasons. Three subjects (HA cream group, n =2; neutral cream group, n =1) withdrew for AEs and 3 (HA cream group, n =1; neutral cream group, n =2) for SAEs. Three subjects in the HA cream group withdrew because of a protocol violation (n =1) or were lost to follow-up (n =2), and 3 subjects (HA cream group, n =2; neutral cream group, n =1) withdrew for other reasons. One subject in each group was lost to follow-up. A total of 5 subjects (HA cream group, n =4; neutral cream group, n =1) had at least 1 major protocol deviation, most commonly for prohibited concomitant medication.

Adherence to treatment was almost 100% in both groups, as expected, because daily medication change was performed by the study nurse.

Efficacy

At study completion, the proportion of subjects in the FAS with centrally assessed complete healing of the target ulcer at week 20 or at any control visit and confirmed 3 weeks later (primary efficacy endpoint) was substantially higher in the HA cream group (31.3%) than in the neutral cream group (14.8%; P =.009) (Table 2). There was no significant difference in treatment effect for complete ulcer healing between strata (baseline ulcer size ≤20 cm² vs >20 cm²; Breslow-Day P =.637) (Table 2). There was a similar trend for primary efficacy performance in the per-protocol set (PPS) (data not shown). Pictures of the target ulcer area taken before treatment, at inclusion, and at selected time points during the study in 2 participants, representative of the big ulcer (>20 cm²) and of the small ulcer (≤20 cm²) subgroups, are shown in Figure 1A and Figure 1B, respectively.

Similarly, the proportion of subjects in the FAS with complete healing of the target ulcer and confirmed 3 weeks later was higher in the HA cream group in comparison with the neutral comparator group when assessed by the investigator at the site (28.9% vs. 14.8%, P =.029) (Table 2). The results obtained in the per-protocol set (PPS) for the primary efficacy performance and all secondary endpoints were consistent with those in the FAS (data not shown).

Cumulative Kaplan-Meier estimates of the probability of complete healing over the entire study period showed that a statistically significantly higher proportion of subjects in the HA cream group achieved complete ulcer healing as compared with the neutral cream group when centrally assessed (P =.006) and according to the assessment of the investigator at the site (P =.017) (Figure 2). Achievement of complete healing of target ulcers was significantly better in the HA cream group by week 8 (P =.017), but the difference between groups was only of borderline significance at the final follow-up visit (P =.067). In the subgroup of subjects in the HA cream group who achieved complete healing of the target ulcer, the median time to healing was approximately 12 weeks, according to the central assessor.

Logistic regression analysis of explanatory variables on ulcer healing showed that target ulcer size at baseline (≤ 20 cm² vs > 20 cm²), treatment (HA cream vs neutral cream), and prior target ulcer duration at baseline were all significantly associated with ulcer healing (all P <.01). Other potentially confounding variables tested in the model were subject age (years), baseline pain (VAS score), sex, baseline weight, and body mass index, and clinical site; none of these were found to have a significant effect on the primary outcome. Although the reduction in residual target ulcer area relative to baseline was numerically higher in the HA cream group at all time points, the difference did not reach statistical significance (Figure 3).

There was no statistically significant difference between groups for the condition of the periulcerous skin, with the exception of a slightly lower incidence of purpura and erythema in the neutral cream group at week 16. However, it should be noted that the condition of the periulcerous skin was, overall, more compromised in the HA cream group at baseline.

There was no statistically significant difference between groups in the reported use of analgesics or in the proportion of subjects with reported infection of the target ulcer, and no correlation between the achievement of complete healing and concomitant use of pentoxifylline was observed in either group (data not shown).

The mean VAS score for pain intensity diminished over time during the study period, in a similar manner in both groups. Ulcer pain was mild, on average, at the final visit in both groups, with more than 50% reduction as compared to pretreatment levels. Results obtained in the PPS were fully consistent with those in the FAS (data not shown).

Safety

A total of 64 TEAEs were reported by 34 subjects (40.0%) in the HA cream group, and 84 were reported by 38 subjects (45.8%) in the neutral cream group (Table 3, Supplementary File). Most TEAEs were mild to moderate in severity.

The most commonly affected system organ classes were skin and subcutaneous tissue disorders and infections and infestations, occurring in 21.2% and 11.8% of subjects, respectively, in the HA cream group and in 16.9% and 15.7% of subjects, respectively, in the neutral cream group.

There were 11 treatment-emergent SAEs (Table 3). There was no statistically significant difference between groups in the proportion of subjects reporting at least 1 TEAE, serious TEAEs, or AEs leading to study discontinuation. Furthermore, there was no statistically significant difference between groups in the proportion of subjects reporting at least one AE involving a device, and there was no statistically significant difference between groups either in terms of TEAEs severity or relationship to treatment.

Treatment-related TEAEs were reported in 10 subjects (11.8%) in the HA cream group (12 events) and 7 subjects (8.4%, 8 events) in the neutral cream group. Ulcer pain was the most frequently reported TEAE (3 AEs in 3 subjects in both groups). Treatment-related dermatitis (3 subjects), erythema (3 subjects), pruritus (1 subject), and rash (1 subject) were also occasionally reported in one or both treatment groups.

One case of target ulcer infection was reported in the HA cream-treated group; the event was mild in severity and resolved without consequence.

In this multicenter study conducted in Poland, 85 subjects with chronic leg ulcers of venous or mixed origin were treated with HA cream vs 83 with a neutral comparator. Subjects treated with the HA 0.2% cream daily for 20 weeks were more likely to experience complete healing, with an effect on ulcer size apparent after 8 weeks. The reduction in ulcer size was most significant in smaller ulcers (≤20 cm²) and in those of shorter duration. To date, the use of topical HA in subjects with vascular leg ulcers has largely been empirical or in studies of shorter duration. To the authors’ knowledge, this study is the first to investigate the application of an HA cream in the context of a randomized controlled trial in which subjects were followed up for more than 20 weeks. The study design was rigorous, with appropriately selected endpoints that were consistent with recent recommendations of scientific and regulatory bodies.^23-25 To minimize bias, the test and comparator treatments were randomly allocated using a standard central randomization system, and access to the randomization code information was strictly regulated and monitored, and the clinical assessment of the primary efficacy variable was centrally and independently performed by an experienced, blinded assessor judging clinical results via standardized photography of the target ulcer. All personnel involved in the trial were blinded to the treatment, and the presentation, packaging, and labeling of the HA cream and neutral comparator cream were fully indistinguishable.

In terms of ulcer size, presence of necrotic tissue, target ulcer duration, ankle-brachial pressure index albuminemia, and other demographic and target ulcer characteristics at baseline, the subject population was comparable to that of other studies in the indication.^14-17,19 However, subjects with leg ulcers of mixed origin, for whom treatment is more difficult, were allowed in this study.

By 23 weeks, the primary efficacy endpoint of complete ulcer healing in the FAS, as measured by a blinded central assessor on standard digital images of the target ulcer using validated image analysis software, was 31.3% in the HA cream group, compared with 14.8% in the neutral cream group, a clinically and statistically significant difference that was achieved regardless of ulcer size (≤20 cm² or >20 cm²) at baseline. Results in the PPS and when assessed by the investigator were fully consistent with the primary endpoint findings. For subjects with a completely healed target ulcer, the median time to complete healing was significantly shorter (12 weeks) in the HA cream group than in the neutral cream group (16 weeks).

No new or critical safety signals were identified in the study, and the results support the established favorable safety profile of this HA cream in the treatment of subjects with acute and chronic wounds.

Compression therapy to reduce the ambulant venous hyperpressure remains the mainstay for the management of venous ulcers, with the aim of transforming the chronic wound into an acute wound that can heal through the regular process of wound healing, a coordinated cascade of cell proliferation, migration, and differentiation.¹² However, subjects with compromised arterial circulation require a lower level of compression to avoid the risk of pressure-related complications, including critical ischemia and tissue necrosis.²⁶

A number of therapeutic interventions combined with standard or modified compression therapy have been investigated to reduce the time required for ulcer healing. Pharmaceutical treatments, surgical interventions, the use of negative pressure devices, electrical stimulation therapy, physiotherapy, and topical agents have been investigated with varying degrees of success.^2,27 However, high-quality evidence to support specific interventions in venous leg ulcers is lacking because of factors such as small sample sizes, short follow-up, study design limitations, lack of appropriate control procedures, and inconsistencies in reporting measurements.^24,25

Hyaluronic acid has a number of chemical and physical properties that explain its ability to elicit tissue healing in chronic wounds. The high molecular weight and negative charge of HA confer a capacity to bind large amounts of water, essential for regulating tissue hydration. In combination with collagen and elastin, HA contributes to the stabilization of intracellular structures via the formation of a viscoelastic network.^28,29 Hyaluronic acid also helps protect against the deleterious effects of the free radicals that play a key role in skin aging and is involved in nutrient exchanges essential for wound healing and biologic processes, such as cell differentiation and motility.^12,29,30 Finally, as a component of the extracellular matrix, the dynamic tridimensional structure of HA creates favorable physiologic conditions for proliferation, migration, and structuring of the epidermis, all processes involved in wound healing.^7,8,28,29

The topical application of exogenous HA as an HA-containing cream or HA-impregnated dressings has long been routine to encourage tissue healing of chronic wounds, including leg ulcers.¹⁹ Although the exact mechanisms of wound healing are still under investigation with the aim of improving chronic wound management, the results obtained in this trial are in line with the current understanding of the multifaceted role of endogenous HA in the wound repair process and support the hypothesis that HA delivered as a cream formulation is effective in promoting and accelerating the wound healing process and in reducing pain in leg ulcers of venous or mixed origin.

Although a number of appropriate measures were implemented to minimize bias in this randomized controlled study, there remain several limitations to consider. Stratification by ulcer duration was not undertaken at baseline; however, as a known potential confounding variable, it was analyzed as an explanatory variable during the logistic regression analysis. In addition, the cost-effectiveness of the treatment was not assessed. Because all subjects were recruited in Poland, the generalizability of the results to other countries should consider differences in health care system resourcing and organization in other countries.

Treatment of chronic leg ulcers of venous or mixed (venous or arterial) origin with HA cream is safe, well tolerated, and results in a higher rate of healing than a neutral comparator cream. From the results, HA is a major structural component of the extracellular matrix and based on the results of the study appears to have a major role in wound healing.

The authors thank Ray Hill, an independent medical writer, who provided technical writing support funded by IBSA Institut Biochimique S.A. (Lugano, Switzerland). The authors thank Simone Cazzaniga, a member of Centro Studi GISED, who measured target ulcer areas using a validated image analysis system in blind conditions. The authors thank all the CROMSOURCE (Verona, Italy) staff involved in coordinating and managing the clinical trial. The authors thank all the clinical research sites that participated in the study.

Authors: Jacek Mikosinski, MD, PhD¹; Anna Di Landro, MD²; Karolina Kasztalska-Kazmierczak, PhD³; Emilie Soriano, MSc, PharmD⁴; Carol Caverzasio, MSc⁵; Daniela Binelli, MStat⁵; Bruno Falissard, MD, PhD⁶; and Olivier Dereure, MD, PhD⁷

Affiliations: ¹Nzoz Mikomed, ul., Łódź, Poland; ²Centro Studi GISED, Bergamo, Italy; ³CROMSOURCE, Verona, Italy; ⁴IBSA Pharma SAS (previously Laboratoires GENEVRIER), Antibes, France; ⁵IBSA Institut Biochimique S.A., Pambio-Noranco, Lugano, Switzerland; ⁶Director, Centre de Recherche en Epidemiologie et Santé des Populations, Maison de Solenn, 75679 Paris, France; ⁷Department of Dermatology, Hospital Saint-Eloi, Montpellier, France

Correspondence: Carol Caverzasio, MSc, IBSA Institut Biochimique S.A., Via del Piano, 29, 6915 Pambio-Noranco (TI), Switzerland; carol.caverzasio@ibsa.ch

Disclosure: The study was sponsored and funded by IBSA Institut Biochimique S.A. CC and DB are employees of IBSA. ES was an employee of Laboratoires Genevrier at the time the manuscript was conceived and prepared. KK is an employee of CROMSOURCE. Centro Studi GISED received a grant from IBSA, ADL is a member of GISED. JM was the coordinating investigator of the study. BF and OD were members of the data review board of the study, and both received a grant from IBSA Pharma SAS.