In-vivo Effectiveness of Adapalene 0.1%/Benzoyl Peroxide 2.5% Gel on Antibiotic-sensitive and Resistant Propionibacterium acnes

Abstract

Background: A gel combination treatment containing a retinoid (adapalene 0.1%) and an antimicrobial (benzoyl peroxide 2.5%) has been shown to be an effective treatment for acne vulgaris, addressing three of the four pathogenic factors (hyperkeratinization, Propionibacterium acnes proliferation, inflammation) without contributing to the incidence of Propionibacterium acnes antibiotic resistance as neither the retinoid nor benzoyl peroxide creates selective pressure for resistance. Objective: To evaluate the effectiveness of an adapalene-benzoyl peroxide gel combination in reducing antibiotic-sensitive and resistant strains of Propionibacterium acnes on the facial skin of volunteers. Methods: This four-week, open-label, single-center study included 30 healthy adults with high facial Propionibacterium acnes populations [>10⁴ colony-forming units per square centimeter of skin (CFU/cm²)] and presence of subpopulations resistant to erythromycin, tetracycline, and clindamycin. The gel was applied once daily to the forehead. Cultures for total and antibiotic-resistant Propionibacterium acnes were obtained from the forehead area at screening, Baseline, Week 2, and Week 4. Results: Total Propionibacterium acnes counts decreased by 1.1 log₁₀ CFU/cm² after two weeks of treatment, and by 1.6 log10 CFU/cm² after four weeks. All subjects had strains resistant to each of the five antibiotics at baseline. Mean counts of erythromycin and clindamycin resistant Propionibacterium acnes were high at baseline (5.37 and 5.28 log₁₀ CFU/cm², respectively) and decreased by ≥2.1 log₁₀ by Week 4 (P<0.001). Mean counts of strains resistant to tetracyclines were lower at baseline (3.8 to 4.2 CFU/cm²) and decreased by 1.9 (tetracycline), 2.4 (minocycline), and 1.3 (doxycycline) log₁₀ CFU/cm² by Week 4 (P<0.001). Limitations: Although limited in scope, the results of the present study demonstrate that the fixed-dose combination gel containing adapalene 0.1% and benzoyl peroxide 2.5% effectively inhibited both antibiotic-susceptible and antibiotic-resistant Propionibacterium acnes. In addition to reducing population densities, therapy with adapalene-benzoyl peroxide eradicated some resistant strains entirely in some individual subjects. Conclusion: Topical adapalene-benzoyl peroxide gel effectively reduced skin colonization by antibiotic-sensitive and antibiotic-resistant Propionibacterium acnes after four weeks. This trial was registered with ClinicalTrials.gov (http://clinicaltrials.gov/), registry number NCT00907101.

The pathogenesis of acne is multifactorial and includes sebaceous follicle obstruction, excessive sebum production due to hormonal stimulation of sebaceous glands, proliferation of Propionibacterium acnes, which produces chemotactic factors and proinflammatory mediators that, in turn, generate an inflammatory response, followed by follicular rupture and extension of inflammation into the dermis, resulting in the formation of inflammatory lesions.^{1, 2}

As P. acnes proliferation in sebaceous follicles plays a central role in the pathophysiology of inflammation in acne, a primary mode of action for benzoyl peroxides (BPOs) and topical antibiotics is through inhibition of P. acnes in vivo as demonstrated by a 1.0 to 2.0 logarithmic (base 10) reduction.^3–7 Strains of P. acnes that are insensitive or resistant to antibiotics commonly used in the treatment of acne can also be found on the skin of acne patients worldwide.^8–16 In these studies, higher minimum inhibitory concentration (MICs) were found for erythromycin followed by clindamycin, and most erythromycin-resistant strains also showed varying levels of insensitivity to clindamycin.⁸ Resistance to tetracycline hydrochloride is also common, and resistance to doxycycline and minocycline is increasing.⁴

Although the clinical significance of P. acnes resistance to antibiotics still remains to be fully elucidated, it is generally believed that the antibacterial activity of antibiotics plays an important role in their efficacy in acne, and that the high prevalence of antibiotic-resistant P. acnes on the skin of acne patients may compromise the efficacy of oral or topical antimicrobial treatment for acne.^4,8,9,11,17

Based on the trend of decreased sensitivity to the antibiotics typically used in acne, as well as the trend in decreased efficacy and even nonresponse to these agents, new research has focused on 1) the development of optimum methods of use of antibiotics in acne management to preserve the utility of these drugs (i.e., in combination with BPO to help minimize the development of resistance), and 2) the development of nonantibiotic therapies to potentially minimize the risk of community antibiotic resistance.^{18, 19}

In 2003 and 2009, the Global Alliance to Improve Outcomes in Acne issued recommendations stressing that, in order to effectively address as many acne pathogenic factors as possible while minimizing side effects, a rational approach to acne management was to use multiple agents at the same time during treatment (concomitant therapy).^{20, 21} Some of the treatment recommendations include lower concentrations of BPO (2.5%, 5%) and topical retinoids (adaplene 0.1%, tretinoin microsphere 0.04%, tazarotene 0.05%).

One such commercially available, antibiotic-free, therapeutic option for the treatment of acne is a fixed-dose, once-daily gel combining adapalene 0.1%, a well-tolerated and efficacious topical retinoid, and BPO 2.5%, a well-established nonantibiotic antimicrobial agent (Epiduo® Gel, Galderma Laboratories, L.P., Fort Worth, Texas). Three randomized, double-blind, vehicle- and placebo-controlled clinical trials have established the safety and efficacy of adapalene-BPO gel in the treatment of acne and have shown that this fixed-dose combination gel is significantly more effective than the corresponding monotherapies of BPO alone or adapalene alone.^22–24 Retinoids such as adapalene help normalize the abnormal desquamation of follicular keratinocytes (comedogenesis) and have some anti-inflammatory properties.²⁵ Adapalene itself possesses anticomedogenic and anti-inflammatory properties.^26–29 The antimicrobial BPO, an oxidizing agent with antibacterial and keratolytic effects, is widely used in acne treatment for its activities in decreasing the bacterial population of P. acnes.^30–33 Thus far, no microbial resistance to BPO has been described. In addition, the nonclinical and clinical safety profile of BPO has been well established.³⁴

Because neither the retinoids nor BPO create selective pressure for resistance, this combination may also be expected to decrease the incidence of epidermal bacterial resistance relative to antibiotics. This study was conducted to assess the ability of adapalene 0.1%/benzoyl peroxide 2.5% gel (adapalene-BPO gel) to reduce the numbers of sensitive and resistant strains of P. acnes in subjects who are colonized with these bacteria.

Methods

This study was conducted in June and July 2009 in the United States in accordance with the ethical principles originating from the Declaration of Helsinki and Good Clinical Practices and in compliance with all regulatory requirements. The protocol was reviewed and approved by an institutional review board. Prior to the performance of any study procedures, written informed consent was obtained from all participants.

Subjects. The volunteers selected for the study were healthy adult men and women 18 years of age and older who showed a high degree of fluorescence of the facial skin under a Wood's lamp indicating the presence of high levels of P. acnes. Because the objective of this study was to evaluate pattern of P. acnes sensitivity and resistance to antibiotics, the primary inclusion criteria focused on the presence of insensitive or resistant bacteria on the skin of the panel volunteers, rather than whether or not they displayed active acne. Subjects were further screened to ensure that none of them had used any form of topical or systemic antibiotics within four weeks prior to enrollment. They were instructed not to use any antimicrobial topical products (e.g., soaps, medicated shampoos, acne preparations) or systemic antibiotics for the duration of the study. Baseline samples taken from the facial skin (forehead) were to confirm P. acnes counts of >10⁴ CFU/cm². In addition, subjects presented with P. acnes with high MICs to erythromycin (>512µg/mL), varying high MICs to clindamycin (8 to >512µg/mL), and MICs greater than 2µg/mL for tetracycline.

Adapalene-BPO gel application. Adapalene-BPO gel was applied topically once a day to the entire forehead area for 28 days. Once-daily treatment was applied by each subject under supervision by a technician at the study center in a standardized manner (Monday–Friday); application on weekends was unsupervised. Prior to each daily application, the subjects washed their face, patted dry, and allowed to dry thoroughly. They were instructed to use enough test material to cover the forehead from temple to temple. After the application was complete, they were instructed to wash their hands.

Collection and processing of bacterial samples. Quantitative cultures were collected from the surface of the forehead at screening, baseline, Week 2, and Week 4 using a modification of the detergent scrub method of Williamson and Kligman.³⁵,³⁶ The forehead was cleansed of surface aerobic bacteria by thoroughly wiping the area for 30 seconds with sterile gauze soaked with 0.1% Triton-X-100 (Sigma-Aldrich, Inc., St. Louis, Missouri) to remove surface debris and bacteria. The surface area to be cultured (3.8cm²) was then delineated by a sterile plastic template held firmly to the skin. A cotton-tipped swab was first moistened in 2mL of wash solution (Bacto Letheen Broth, Difco, Sparks, Maryland) and used to swab the forehead area for 30 seconds. The swab tip was then placed into 2mL of wash solution, and the bacteria were released by wringing the swab on the side of the tube. The procedure was repeated on the same skin area with the same swab. The swab was then broken into the 2mL of wash solution. The sample was serially diluted using 0.05% Tween-80 (Sigma-Aldrich, Inc.) (buffered with 0.075M phosphate buffer, pH 7.9) in four 10-fold dilutions. Using a micropipettor, 0.05mL aliquots of each dilution were placed onto plates of Brucella agar (Neogen, Lansing, Michigan) supplemented with yeast extract, dextrose, and cysteine, with and without antibiotics, each on duplicate plates. Serial antibiotic incorporation plates with 1, 2, 4, 8, 16, 32, 64, 128, and 512µg/mL were made fresh the day before subjects were to be cultured. Cultures were placed in an anaerobic jar with BBL Gas Pak Plus anaerobic system envelope and were incubated anaerobically at 35 to 37°C for seven days. Representative colonies on each medium were counted at the dilution that contained between 10 and 100 CFUs, and the number of viable P. acnes colonies was expressed as the average (of both plates) of the log₁₀ CFU/cm² skin.

In the study screening phase, culture samples were plated in Brucella agar with incorporation of erythromycin 8µg/mL and tetracycline hydrochloride 2µg/mL. All antibiotic incorporation plates were prepared the day prior to use to ensure that the antibiotic had not degraded (tetracyclines degrade after 2–4 days). Another check on the stability of the incorporated antibiotics was the plating of strains known to be sensitive to tetracyclines.

Statistical analysis. The primary analysis was conducted on P. acnes log₁₀ CFU/cm² testing the null hypothesis that the change from Baseline at Week 2 and Week 4 was zero versus the alternative hypothesis that the change from Baseline was less than zero. The Shapiro-Wilk normality test was performed on the difference between quantitative bacteriology measurements at Baseline and each post-baseline visit. If the difference between quantitative bacteriology measurements at the two time points was normally distributed, a t-test on the difference was performed. If the difference was not normally distributed, then the Wilcoxon signed rank test on the difference was used. Additionally, the percent change from baseline in P. acnes log₁₀ CFU/cm² was calculated using the formula (10^x−10y)/10^y × 100, where 10x was the log₁₀ CFU/cm² at Week 2 or Week 4, and 10^y was the log₁₀ CFU/cm² at Baseline.

Results

A total of 44 suitable subjects were microbiologically screened. Of these, 30 met the microbiological entry criteria of >10⁴ CFU/cm² skin with prospectively defined levels of resistance and all were included in this analysis. The 30 subjects entered into the study (22 male and 8 female) were all Caucasian except for one subject who was African American. Age ranged from 19 to 61 years (average 36 years). All 30 subjects completed the 28-day treatment to the forehead with 100-percent adherence, and all attended the Week 2 and Week 4 visits. None of the subjects had used topical or oral antibiotics within four weeks prior to the screening visit.

MICs of P. acnes strains at Baseline. This panel of 30 subjects had strains of P. acnes both sensitive and insensitive to all five antibiotics tested (Table 1). All subjects had strains with high MICs to erythromycin (≥512µg/mL) and varying sensitivity to clindamycin (ranging from 8 to 1,024µg/mL), similar to those seen in published series.² In the tetracycline family, all subjects had strains of P. acnes with low MICs ranging from 2 to 16µg/mL. Thus, all 30 subjects had populations of multiple antibiotic-resistant P. acnes.

Table 1.

Number and percent of subjects with antibiotic-resistant P. acnes strains identified at baseline, by level of resistance

CATEGORY	ERYTHROMYCIN	CLINDAMYCIN	TETRACYCLINE	MINOCYCLINE	DOXYCYCLINE
MIC ≥512µg/mL	30(100%)	6 (20%)	0	0	0
MIC 128–256µg/mL	0	10(33%)	0	0	0
MIC32–64µg/mL	0	1 (3%)	0	0	0
MIC 2–16µg/mL	0	13 (43%)	30(100%)	30(100%)	30(100%)

Population density of total and antibiotic-resistant P. acnes. Daily application of adapalene-BPO gel resulted in significant reductions in total P. acnes counts and in counts of all antibiotic-resistant strains after two weeks and four weeks of treatment (Figure 1 and Table 2). The results of the quantitative bacteria cultures demonstrated a significant decline in the mean total P. acnes counts present at Week 2 and Week 4 with a reduction of 1.1 log₁₀ CFU/cm² (80% reduction) at Week 2 (P<0.001) and 1.6 log₁₀ CFU/cm² (93% reduction) at Week 4 (P<0.001).

Figure 1.

Effect of adapalene-BPO gel on total and antibiotic-resistant populations of P. acnes

Table 2.

Mean reduction (log₁₀ and percent) in total P. acnes and in resistant subpopulations during treatment with adapalene-BPO

	BASELINE VALUES		CHANGE AT WEEK 2			CHANGE AT WEEK 4
	N	MEAN (SD)	N	LOG10 MEAN (SD)	PERCENT MEAN (SD)	N	LOG10 MEAN (SD)	PERCENT MEAN (SD)
Total P. acnes	30	6.7 (0.28)	30	−1.1 (0.81)*	−80% (21.6)*	30	−1.6(0.78)*	−93% (8.1)*
Erythromycin-resistant P. acnes	30	5.4 (0.85)	30	−1.7(1.02)*	−89% (20.4)*	26	−2.2 (0.86)*	−97% (8.9)*
Clindamycin-resistant P. acnes	30	5.3 (0.96)	30	−1.7(1.22)*	−82% (32.1)*	27	−2.1 (1.15)*	−92% (19.7)*
Tetracycline-resistant P. acnes	30	4.2(1.34)	26	−1.3(0.97)*	−79% (25.7)*	16	−1.9(0.92)*	−92% (15.6)*
Minocycline-resistant P. acnes	30	4.0(1.47)	25	−1.6(1.19)*	−85% (22.6)*	13	−2.4(1.29)*	−97% (3.7)*
Doxycycline-resistant P. acnes	30	3.8(1.41)	27	−1.1 (1.21)*	−67% (28.0)*	22	−1.3(0.75)*	−88% (14.5)*

^*All P<0.001 by t-test comparing change from baseline to zero (if change from baseline was normally distributed) or by Wilcoxon signed rank test comparing change from baseline to zero (if the change from baseline was not normally distributed)

Baseline mean counts of resistant P. acnes were highest for erythromycin and clindamycin (5.37 and 5.28 log₁₀ CFU m², respectively) and these counts decreased by 2.2 (97% reduction) and 2.1 log₁₀ (92% reduction), respectively, after four weeks of treatment with adapalene-BPO gel (P<0.001). Baseline mean counts of strains resistant to tetracycline, minocycline, or doxycycline were lower (3.8–4.2 CFU/cm²) and were reduced by 1.9 (92% reduction; tetracycline), 2.4 (97% reduction; minocycline), and 1.3 (88% reduction; doxycycline) log₁₀ CFU/cm² by Week 4 (P<0.001). Thus the number of erythromycin- and clindamycin-resistant bacteria was reduced by several orders of magnitude reduction, while the bacteria resistant to one or more of the tetracyclines decreased to levels close to total eradication.

Conclusion

Although limited in scope, the results of the present study demonstrate that the fixed-dose combination gel containing adapalene 0.1% and BPO 2.5% effectively inhibited both antibiotic-susceptible and antibiotic-resistant P. acnes, reducing the bacterial populations by 97 percent after four weeks of application. In addition to reducing population densities, therapy with adapalene-BPO eradicated some resistant strains entirely in some individual subjects.

With the growing incidence of antibiotic-resistant P. acnes on the skin of patients with acne, adapalene-BPO gel provides clinicians and patients with a combination product that is effective against acne, is well tolerated, and is also effective against antibiotic-resistant P. acnes. This makes adapalene-BPO gel a therapeutic option for treating acne in patients with resistant P. acnes while minimizing the development of further antibiotic resistance.

Acknowledgment

The authors would like to acknowledge the role of Christiane Baud, PhD, in preparation of the manuscript.