Reduction of Inflammatory and Noninflammatory Lesions with Topical Tyrothricin 0.1% in the Treatment of Mild to Severe Acne Papulopustulosa: A Randomized Controlled Clinical Trial

Acne vulgaris is a disorder of the pilosebaceous unit, mostly affecting teenagers of both genders. The pathogenesis of acne is multifactorial. There are four primary pathogenic factors, which interact to produce acne lesions: (1) increased sebum production in the sebaceous glands, (2) alterations in the keratinization process, (3) Propionibacterium acnes follicular colonization, and (4) release of inflammatory mediators [1,2,3,4,5].

Hyperkeratinization leads to plugging of the follicular duct [5] which results in the clinically invisible microcomedones preceding acne lesion formation. They might develop into noninflammatory lesions (closed and open comedones) and into clinically visible inflamed papules, pustules, and nodules. Despite this differentiation, the inflammatory potential of acne vulgaris is given throughout the course of the disease; hence, acne is an inflammatory disease from its beginning onwards [3,6,7,8,9,10,11].

For the treatment of mild to severe papulopustular acne, topical treatments with benzoyl peroxide (BPO), azelaic acid, and retinoids as well as combinations of adapalene/BPO, clindamycin/BPO, and systemic antibiotics/adapalene are recommended [1]. These established combination products of antibiotic and antimicrobial agents aim at reducing the risk of developing antibiotic resistance and maintaining treatment efficacy in acne patients. Topical BPO has antimicrobial, comedolytic, sebosuppressive, and anti-inflammatory properties and does not form resistances [1,12].

One major concern in the treatment of acne is the growing rate of resistant cutaneous propionibacteria, especially seen in erythromycin and clindamycin treatments [13]. Searching for alternative treatment options in order to reduce antibiotic-induced resistance as well as developing new effective and tolerable therapies for acne patients have become main targets in acne research [1,4,14,15,16].

Antimicrobial peptides (AMPs) are regarded as alternatives to traditional antibiotics in the treatment of infectious and inflammatory skin diseases [15,17]. The so-called ‘nature's antibiotics' are cationic, amphiphilic peptides built up by 6-50 amino acids. They interact with anionic phospholipid membranes and show a broad-spectrum of antimicrobial activity against a wide range of pathogens including bacteria and fungi [17,18]. AMPs might contribute to overcoming the increasing problems of antibiotic resistance [19,20], based on their diversity, safety and mode of action by directly targeting and destroying membranes even in multidrug-resistant pathogens [15,19,20].

AMPs are part of the innate immune system of mammals, forming an initial barrier to pathogenic influences before the adaptive immune system develops [21,22]. In humans they are predominantly produced by keratinocytes in the upper skin layers [18,22]. In innate immune defense, AMPs play an important role in epithelial barrier protection [18]. AMPs also play a role in some inflammatory processes, as their release provides innate antibiotic-like action against infectious pathogens [22].

One of the oldest clinically used AMP is tyrothricin [17], produced by Bacillus brevis [23,24]. This polypeptide antibiotic substance consists of the two cyclic decapeptides, gramicidin S (22%) and tyrocidine A (78%) [24,25,26]. Both peptides have broad bactericidal activity against Gram-positive bacteria due to intercalation of the peptides into bacterial membranes [17,27].

Topical tyrothricin 0.1% has been successfully used for the treatment of small, superficial, and infectious wounds by promoting wound healing for more than 70 years [28,29]. Available evidence suggests that tyrothricin has good tolerability and no potential to induce bacterial resistance [25,30,31,32]. Recently, the antimicrobial effectiveness of tyrothricin against Propionibacterium acnes was demonstrated in vitro [33].

Therefore, this AMP might be a potential candidate for acne treatment. The purpose of this clinical study was to investigate the efficacy and tolerability of topical tyrothricin 0.1% in the treatment of mild to severe acne vulgaris of the papulopustular type compared to two established topical acne treatments.

This 25-day, exploratory, active comparator-controlled, randomized, observer-blind, intraindividual, monocenter clinical trial was conducted at the Charité - Universitätsmedizin Berlin, Department of Dermatology and Allergy, Clinical Research Center for Hair and Skin Science, Berlin, Germany. Eligible patients were recruited from October 2013 to February 2014; the last patient completed the study on March 13, 2014.

The randomization process consisted of two parts. First, all patients were randomized in a 1:1 ratio to receive topical tyrothricin 0.1% (Engelhard Arzneimittel GmbH & Co. KG, Niederdorfelden, Germany) and either topical BPO 5% (Aknefug® Oxid mild 5%; Dr. August Wolff GmbH & Co. KG Arzneimittel, Bielefeld, Germany) or the combination of clindamycin + BPO (Duac® Akne Ge; GlaxoSmithKline GmbH & Co. KG, Berlin, Germany) as comparator treatments on the face. Further, patients were also randomized in a 1:1 ratio to receive the respective comparator treatment on the right or left half-face. For allocation of the treatments, a computer-generated list was used. Randomization was performed successively using sealed randomization envelopes. Investigators involved in the study assessments were blinded to the identity and allocation of the treatments.

Male and female nonsmoking patients aged 18-25 years with a diagnosed mild to severe acne papulopustulosa, defined as an Investigator's Static Global Assessment (ISGA) score of 2-4 [34,35], were included after receiving thorough information on the study aim and procedure, and giving their written informed consent. Another inclusion criterion was that the number of inflammatory and noninflammatory acne lesions on one half-face was not greater than twice the number of the lesions on the other half-face. The main exclusion criteria were more than 2 nodulocystic lesions on each side of the face, treatment with systemic (4 weeks before randomization) or topical (2 weeks before randomization) anti-inflammatory drugs, treatment with drugs that are known to exacerbate acne (e.g. high doses of certain vitamins, haloperidol, halogens, lithium) for 3 months before randomization, smoothing or ablative procedures within 3 months before randomization, known hypersensitivity against tyrothricin, BPO or clindamycin, intensive UV exposure within 4 weeks before randomization, pregnancy, and lactation.

Prior to the study start, the protocol and the patient informed consent form were approved by the independent Ethics Committee of the State Office of Health and Social Affairs Berlin. The study was registered at the European Union Drug Regulating Authorities Clinical Trials Database (EudraCT 2013-001716-30; https://eudract.ema.europa.eu/). The study was conducted according to the principles of the Declaration of Helsinki (1996) and Good Clinical Practice Guidelines (1996). All patients provided written informed consent.

Treatments were applied according to product information once daily in the evening by an unblinded study nurse on the respective facial side of the patient, avoiding the median line to prevent translocation of investigational products. Further, the perimucosal skin around the eyes and lips was not treated. Patients were seen daily during the course of the study. The study products were applied without investigator access. ISGA scoring and lesion counting were performed by trained investigators on days 1 and 25.

The study outcomes were inflammatory, noninflammatory, and total lesion counts as well as ISGA scores per half face. These were obtained at baseline, days 4, 8, 12, 15, 18, 22 and at the end of the study (day 25). Furthermore, dermal tolerability and safety were assessed by recording adverse events (AEs) throughout the study from screening until day 25.

A sample size of 12 patients per treatment group was regarded as appropriate due to the exploratory study design. For all efficacy outcomes, comparisons between treatments were based on 95% CI for intraindividual differences or differences between treatments. Generalized estimating equations (GEE) for analysis concerning the development over the entire time course with day and site/treatment as intraindividual variables and lesion counts as the dependent variable were conducted. Interactions in terms of day and site/treatment were included into the GEE models to analyze different patterns of development over time. Tyrothricin 0.1% served as the reference parameter for the comparator treatments. All p values were considered to be descriptive. Analysis was done according to the originally assigned groups. IBM SPSS Statistics 22.0 was used for statistical analysis.

A total of 24 patients were randomly assigned to one of two treatment regimens (fig. 1). Using a split-face design, 12 of the 24 patients received tyrothricin 0.1% and BPO 5%, and 12 patients received tyrothricin 0.1% and a combination of clindamycin + BPO on each half-face. Twenty-three patients out of 24 completed the trial. In the tyrothricin 0.1% and clindamycin + BPO group, 11 of the 12 patients were analyzed; 1 patient was excluded from the study because of a persistent treatment-related AE on day 8. In the tyrothricin 0.1% and BPO 5% group, all patients were analyzed (fig. 1). Demographic and baseline clinical characteristics were comparable in both treatment groups (table 1).

At baseline, mean inflammatory lesion counts were 21.5 (SD 13.2) for tyrothricin 0.1% and 23.1 (SD 15.1) for clindamycin + BPO (table 2, fig. 2). Mean differences in inflammatory lesion counts between day 25 and baseline were -7.7 (95% CI: -11.7 to -3.7) for tyrothricin 0.1% and -12.3 (95% CI: -20.5 to -4.1) for clindamycin + BPO. Mean differences in noninflammatory lesion counts between day 25 and baseline were -6.5 (95% CI: -11.6 to -1.4) for tyrothricin 0.1% and -16.1 (95% CI: -24.6 to -7.6) for clindamycin + BPO. Mean baseline total lesion counts were 56.1 (SD 22.9) for tyrothricin 0.1% and 60.6 (SD 24.6) for clindamycin + BPO (table 2, fig. 2). Mean differences in total lesion count comparing day 25 with baseline were -14.2 (95% CI: -21.1 to -7.3) in tyrothricin 0.1% and -28.4 (95% CI: -39.9 to -16.9) in clindamycin + BPO (table 2, fig. 2). The ISGA score was reduced by 0.3 in tyrothricin 0.1% and 0.5 in clindamycin + BPO (table 2, fig. 2). Comparing both treatments, we found a statistically stronger reduction in inflammatory, noninflammatory, and total lesion count over time with clindamycin + BPO than with tyrothricin 0.1% (p = 0.038 and p = 0.022; table 3) in the GEE analysis.

At baseline, mean inflammatory lesion counts were 19.9 (SD 10.4) for BPO 5% (table 2, fig. 2). Mean differences in inflammatory lesion counts between day 25 and baseline were -10.2 (95% CI: -15.3 to -5.0) for BPO 5%. Mean baseline noninflammatory lesion counts were 34.5 (SD 15.9) for tyrothricin 0.1% and 35.5 (SD 17.0) for BPO 5%. Between day 25 and baseline, mean differences in noninflammatory lesion counts were -16.8 (95% CI: -26.2 to -7.3) for BPO 5% (table 2, fig. 2). For BPO 5%, the total lesion count baseline value was 55.4 (SD 23.4) and the mean differences comparing day 25 to baseline were -26.9 (95% CI: -39.6 to -14.2; table 2, fig. 2). The ISGA score was reduced by 0.3 in both treatments (table 2, fig. 2). Comparing both treatments, we found a statistically stronger reduction in noninflammatory (p = 0.018) and total lesion count (p = 0.045) over time in BPO 5% compared to tyrothricin 0.1%, whereas no statistically significant differences in inflammatory lesions were observed for BPO 5% compared to tyrothricin 0.1% (p = 0.327; table 3).

In dermal tolerability and safety assessments it was observed that a total of 15 out of 24 (62.5%) patients with tyrothricin 0.1% treatment, 9 of 12 (75.0%) patients with clindamycin + BPO treatment, and 11 of 12 (91.7%) patients with BPO 5% treatment reported product-related AEs including local intolerances (scaling, erythema, itching, burning, and stinging). The most common product-related AE was mild scaling. The incidence of product-related AEs was lower in tyrothricin 0.1% (n = 31 in 24 patients) than in the comparator products (n = 37 in 12 BPO 5% patients, and n = 20 in 12 clindamycin + BPO patients). One patient with tyrothricin 0.1% and clindamycin + BPO treatment withdrew because of a persistent product-related intolerance reaction expressed in a moderate edema against clindamycin and BPO (online suppl. material 1, www.karger.com/doi/10.1159/000439439). The two most frequent nonproduct-related AEs were headache (n = 5) and common cold (n = 4).

Acne papulopustulosa is one of the most frequent dermatological disorders worldwide [36,37]. Using a topical AMP could possibly be one option by linking their anti-inflammatory potential and their downregulatory capacity on defensins even in noninflammatory acne lesions [4,17,38].

This exploratory clinical study provides evidence that the topical AMP tyrothricin 0.1% reduces inflammatory and noninflammatory acne lesions. The main finding of this study was an overall statistically significant decline in the number of inflammatory and total lesions in all three treatments. The performance of clindamycin + BPO seems to be best. The efficacy of tyrothricin 0.1% and BPO 5% treatments seems to be similar over time in terms of inflammatory lesion count reduction.

Interestingly, the reduction of inflammatory acne lesions under the comparator treatments was comparable, suggesting that the key effect of both comparator study products might be related to BPO. The noninflammatory lesion count reductions demonstrated that even if clindamycin + BPO, followed by BPO 5%, showed the strongest effects, there seems to be a certain ‘comedolytic' effect of tyrothricin 0.1% as well. This suggests that the mode of action of the applied AMP might be able to reduce follicular obstruction possibly by its immuno-modulatory property [4,22]. Referring to acne pathogenesis, hyperproliferation of the follicular epithelium leads to the development of clinically nonvisible microcomedones, which are the precursor lesions of acne vulgaris. This leads to the hypothesis that even macroscopically noninflammatory lesions carry inflammatory potential [3,6,10]. Hence, the antimicrobial effect of tyrothricin might also be effective in the treatment of microcomedones.

Another finding of this study was the better tolerability using tyrothricin 0.1% with a lower number of local intolerances and AEs compared to clindamycin + BPO and BPO 5%. Taking into account that mild to severe acne papulopustulosa needs a long-term treatment for nearly the entire duration of adolescence, this could be an aspect for further development of an efficient topical AMP without the risk of developing resistance phenomena, as seen in topical antibiotic preparations [25,29].

The strengths of this study were the controlled and standardized treatment and measurement conditions as well as the active comparator design. As shown by Maruani et al. [39], missing head-to-head comparisons with active controls are major weaknesses of randomized clinical trials in acne today.

The limitations of our proof-of-concept study were the size of the investigational cohort and the short study duration of 25 days. Longer treatment periods are more appropriate to study long-term changes in acne trials. Based on the effects showed in our results, larger-scale and longer-term confirmatory studies seem to be justified.

Although the effects of tyrothricin 0.1% treatment were not as strong as in the established comparator treatments, it can be concluded that this AMP might be an alternative approach in the management of mild to severe acne papulopustulosa. It is not known to cause resistance and it seems to be more tolerable compared to established acne therapies. More research is needed to study the mode of action, the anti-inflammatory potential of tyrothricin, longer-term effects, and the efficacy of alternative formulations (e.g. combination products).

The authors would like to thank the whole study team at the Clinical Research Center for Hair and Skin Science for the excellent organization and conduct of this clinical trial. We thank Engelhard Arzneimittel GmbH & Co. KG, Niederdorfelden, Germany, for providing the educational grant and Atiye Saynab and Christiane Staiger for very productive scientific discussions.

Engelhard Arzneimittel GmbH & Co. KG, Niederdorfelden, Germany provided an unrestricted educational grant as well as the investigational products for conducting this investigator initiated clinical trial at the Charité - Universitätsmedizin Berlin, Germany. The funds of research provided by Engelhard Arzneimittel GmbH & Co. KG had no influence on the study design, data collection, and analysis.