Local Pharmacokinetics and Tolerability of Topically‐Applied Bimatoprost to the Scalp of Male Patients With Androgenetic Alopecia

ABSTRACT

Androgenetic alopecia (AGA) is one of the most common causes of hair loss in the male population, characterized by progressive baldness. This study was designed to evaluate the tolerability and local skin pharmacokinetics of bimatoprost in two different formulations in male subjects with AGA. Bimatoprost 1% Formulations A and B were applied topically once daily to the scalp and lateral trunk in the same subjects for 14 days. One skin biopsy from each application site was obtained on Day 14 after the final drug application. Eleven participants were dosed in this study, and their skin samples were analyzed for bimatoprost and its acid metabolite. Both formulations achieved at least a 10‐fold higher bimatoprost concentration in subcutaneous fat, where the hair papilla is typically situated, than the in vitro concentration needed for maximal increased hair growth (42 ng/g) in isolated hair follicles. Both formulations showed a significant decrease in bimatoprost concentration with increasing distance from the skin surface as expected for a topically applied formulation with no significant differences between Formulations A and B. Bimatoprost acid concentrations in most samples were below the lower limit of quantitation (0.8 ng/g) for both formulations. All adverse events reported were mild except for increased serum transaminases in one participant, which was moderate and not considered treatment related. In conclusion, there were no significant differences in bimatoprost concentration between Formulations A and B in the epidermis/dermis layer of the scalp samples from the same study participants, with levels above efficacious concentrations. Treatment with bimatoprost 1% Formulations A and B was well tolerated.

Study Highlights

• What is the current knowledge on the topic?

  • ○
    Bimatoprost is a prostamide originally developed for glaucoma that has shown hair‐growth promoting effects and was approved for eyelashes growth. For androgenetic alopecia (AGA), topical treatment options remain limited, and formulation performance at the hair follicle site is not well‐defined.
• What question did this study address?

  • ○
    The study asked whether two topical 1% bimatoprost formulations differ in target engagement at relevant depth of skin for hair growth and dermal tolerability in men with AGA, and whether either formulation achieves therapeutically relevant drug levels at the presumed follicular target site.
• What does this study add to our knowledge?

  • ○
    Using skin biopsies, the study showed that both formulations delivered bimatoprost into the scalp, and topical dosing produced a steep concentration gradient across the skin layers but still managed to maintain biologically active levels of bimatoprost (based on in vitro activity) deep into subcutaneous fat which is deeper than where hair follicles typically sit (in epidermis/dermis). This study also adds to our knowledge by providing high‐resolution, layer‐specific drug distribution data in human skin after topical application. Using careful skin biopsy collection with meticulous contamination control, the study enabled bioanalysis of drug concentrations across distinct skin layers, rather than relying on whole‐skin or less granular measurements. This approach offers a more detailed understanding of where the drug localizes within the skin, which is especially valuable because the findings are derived from human subjects. Together, these data improve our understanding of topical drug penetration and distribution in a clinical setting.
• How might this change clinical pharmacology or translational science?

  • ○
    These findings support the use of local skin pharmacokinetics to compare topical formulations and help select formulations likely to reach the follicular target. The work strengthens translational development of bimatoprost for AGA by linking formulation design, tissue exposure, and tolerability.

1. Introduction

Hair loss or alopecia is a prevalent condition that many people suffer from at some point in their lives. There are three main types of alopecia, androgenetic alopecia (AGA), alopecia areata (AA), and chemotherapy‐induced alopecia (CIA) [1]. AGA, also known as common baldness, is the most common cause of progressive hair loss. Although all genders can be affected, it is reported more commonly in males as early as puberty and incidence increases with age [2]. Norwood–Hamilton classification is widely used to classify hair loss and has four different types (II–V) with several subdivisions. Briefly, in Type II, frontotemporal hairline recession is not greater than 2 cm anterior to the midcoronal line, where the recession reaches the midcoronal line in Type III, and recedes beyond that line in Type IV. In Type V, the vertex is included in the area of denudation [3].

Bimatoprost is a synthetic prostaglandin analog that was initially developed as an eye drop to patients with elevated intraocular pressure due to its potent ocular hypotensive activity. It is currently marketed as LUMIGAN (0.01% bimatoprost ophthalmic solution) in many countries to reduce elevated intraocular pressure in patients with glaucoma or ocular hypertension by increasing the outflow of aqueous humor [4].

During the clinical development for bimatoprost ophthalmic solution, eyelash growth was spontaneously reported in some patients receiving topical eyedrops of bimatoprost solution 0.03%. The mechanisms by which bimatoprost increases growth and darkness of eyelashes have been proposed to occur by its influences on the eyelash hair cycle and follicles. Preclinical studies have shown that increases in eyelash growth may result from triggering transition of the hair cycle stage from telogen (resting) to anagen (new growth) and prolonging anagen, while eyelash thickening may be associated with larger dermal papilla and hair bulb diameters in early‐anagen follicles [5]. Increases in hair length and darkness have been associated with a prolonged duration of the anagen phase and melanogenesis (melanin synthesis), respectively [6]. In a Phase 2 study, Caucasian females with idiopathic eyelash hypotrichosis received bimatoprost ophthalmic solution of 0.005%, 0.015%, or 0.03% [7]. The change from baseline in eyelash length at Months 3 was 0.74, 0.92, and 1.36 mm and change from baseline in upper eyelash thickness at Months 3 was 12.75, 14.67, and 13.69 mm² across the dose groups, respectively [8]. In a Phase 3 clinical trial that tested bimatoprost for hypertrichosis, a Global Eyelash Assessment (GEA) score was utilized on a 4‐point ordinal scale (1 = minimal; 2 = moderate; 3 = marked; 4 = very marked). The proportion of subjects with at least a 1‐grade improvement from baseline GEA score was statistically significantly greater in the bimatoprost compared with the vehicle group, at the primary timepoint at Month 4. Subsequently, this Phase 3 trial supported approval of bimatoprost solution 0.03% as LATISSE in the United States for the treatment of eyelashes hypotrichosis by increasing their growth and prominence, including length, thickness, and darkness. Based on the clinical efficacy in eyelash growth, and an established safety profile of bimatoprost [9], a clinical development program was initiated to determine the safety and efficacy of a topical bimatoprost to treat AGA in male population.

To optimize clinical performance, a new formulation (Formulation B) containing glyceryl monooleate and oleic acid was developed after the original topical formulation (Formulation A) was tested clinically. Systemic pharmacokinetics for Formulation B were evaluated in a Phase 1 dose‐ascending study at concentrations up to 3%, applied once daily for 28 days [10].

The current study (NCT02848300) was a formulation comparison for the pharmacokinetics of bimatoprost, and it was designed to assess the local skin pharmacokinetics and dermal tolerability of topical bimatoprost 1% Formulations A and B applied once daily to the scalp and lateral trunk for 14 days. The study data will be used to develop a better understanding of the local pharmacokinetics of bimatoprost following administration of Formulation B to assist in the design of future clinical studies.

2. Methods

2.1. Patients

Eligible participants were male patients 18–49 years old with mild to moderate AGA of the vertex area of scalp (modified Norwood–Hamilton scale classification III, IV, or V) who met the following criteria: participant was in good health determined by medical history, 12‐lead electrocardiogram (ECG), physical examination, vital signs, and clinical laboratory tests (laboratory results outside of reference range had to be assessed as acceptable by the investigator); able to follow study instructions and willingness to complete all required procedures and visits; and completed written informed consent and written documentation in accordance with the relevant country and local privacy requirements, where applicable. Patients were excluded from the study if they had an uncontrolled systemic disease; history of significant cardiovascular disease; scarring disease, infection, or abnormality of the scalp or hair shaft or systemic disease that could cause temporary hair loss (e.g., seborrheic dermatitis, psoriatic dermatitis, alopecia areata, cicatricial alopecia, uncontrolled hyperthyroidism/hypothyroidism, tinea infections, and genetic disorders); patients who have received hair transplants or have had scalp reductions. In addition, any other form of bimatoprost product (other than the study treatment) (e.g., bimatoprost ophthalmic solution) was prohibited during the study. In addition, participants did not use topical drugs or products (e.g., sunscreen) on the treated areas other than the study drug, and study participants did not use any FDA‐approved therapy for the treatment of osteoporosis.

2.2. Study Design and Treatment

This was a Phase 1, single center, open‐label study to evaluate skin pharmacokinetics and dermal tolerability following 14 days of once daily topical administration of 1% bimatoprost Formulations A and B in male patients with AGA. As shown in Figure 1, each of the 11 study participants received Formulations A and B to the scalp and lateral trunk as follows:

• Bimatoprost 1% (Formulation A), approximately 0.06 mL administered to a small surface area (~10 cm²) on the left occipital area of the scalp and the left lateral trunk.

• Bimatoprost 1% (Formulation B), approximately 0.06 mL administered to a small surface area (~10 cm²) on the right occipital area of the scalp and the right lateral trunk.

FIGURE 1.

Study drug administration (and biopsy) sites. ⊗ in red is Formulation A, and ⊗ in blue is Formulation B.

The treatment area was cleaned and dried prior to application of study treatment, and study drugs were applied by study staff once daily for 14 days at the same time of day in the morning, ±30 min. For precise dosing and sampling, each patient received tiny, semipermanent ink dots on the left and right opposing sides of the occipital scalp and lateral trunk skin during the baseline visit to indicate the dosing areas. Each formulation (0.06 mL) was drawn in a pipette for application to the patient's scalp or trunk at the ink dot in a dropwise fashion to distribute the study drug evenly over the target application area avoiding any runoff. Study staff in charge of application used their gloved middle and index fingers to evenly spread the study drug to the periphery of the target application area, ensuring the target application area was completely covered. The patient was instructed to allow the application area to air dry.

The investigator obtained approval of the study protocol from an institutional review board (IRB) prior to study initiation (Supporting Information). Study procedures were performed in accordance with the ICH E6 guideline for GCP and the Declaration of Helsinki, or the applicable laws and regulations of the country in which the research was conducted, whichever afforded the greater protection to the individual. Written informed consent was obtained from all patients before enrollment.

2.3. Safety

Safety analyses were performed on all study participants who received at least one dose of study treatment. The safety parameters included adverse events and clinical laboratory, vital sign, 12‐lead ECG parameters, and local dermal tolerability assessment. The safety endpoint was the percentage of participants with at least one treatment‐emergent adverse event.

2.4. Biopsy Collection

Skin biopsies were collected from each study participant to determine bimatoprost and bimatoprost acid concentrations. Scalp and trunk hair were shaved clean prior to sampling to ensure the effectiveness of tape stripping for removal of the stratum corneum layer and to minimize sample contamination.

Location of the biopsy site within the 10 cm² drug‐treated area was determined to ensure sufficient room for tape striping. The biopsy areas were anesthetized via subcutaneous injection of 1 mL 1% lidocaine with epinephrine. Tape stripping was performed using D‐squame disks (Cuderm Corporation, Dallas, TX, USA) and repeated 20 times from the same designated areas. All tape strips were discarded.

Biopsies from the application sites of Formulations A and B were collected between 3 and 4 h after final dosing on Day 14. The timepoint was selected based on time to peak drug concentration observed in dermal pharmacokinetics studies of bimatoprost following topical administration for 15 days in minipigs for Formulations A and B (data on file, skin biopsies were collected at 0.5, 4, 12, and 24 h posttreatment). The biopsies from the application sites were extracted from Formulation B first, and Formulation A second. Skin biopsy samples were obtained by a 4‐mm punch and were immediately frozen. One skin biopsy from each of the four administration sites was collected per study participant for the determination of local concentrations of bimatoprost and bimatoprost acid.

2.5. Sample Processing and Liquid Chromatography–Tandem Mass Spectrometry (LC–MS/MS) Analysis

The biopsies were frozen at the clinical site and shipped to the sponsor for processing. Each biopsy was horizontally sectioned using a cryostat parallel to the skin surface, from dermis to subcutaneous fat, into sections of 100 μm thickness collected as five sections per tube for LC–MS/MS bioanalysis. Sections of 15 μm were also collected between pharmacokinetic sectioning every 500 μm for histological evaluation to enable identification of transition from dermis to subcutaneous fat layer for each biopsy sample. Samples were then shipped to Keystone Bioanalytical Inc. (North Wales, PA, USA). Concentrations of bimatoprost and bimatoprost acid were determined by LC–MS/MS for 377 samples from 36 biopsy samples collected from nine study participants (1 study participant withdrew consent and terminated the study early). The lower limit of quantitation (LLOQ) for the sectioned samples was 0.08 ng/g for bimatoprost and 0.8 ng/g for bimatoprost acid.

Each sample of pooled sections collected for LC–MS/MS bioanalysis was analyzed for concentrations of bimatoprost and its acid metabolite, and the resulting tissue concentrations were calculated.

An LC–MS/MS method was developed for the quantification of bimatoprost, bimatoprost acid and the internal standards (bimatoprost‐d⁴ and bimatoprost acid‐d⁴) (Cayman Chemical, Ann Arbor, MI, USA) in human skin. Chromatographic separation of bimatoprost and bimatoprost acid was achieved on a Unison UK‐C18 (50 × 2 mm, 3 μm) column with gradient elution using acetonitrile and 2.5 mM ammonium bicarbonate in water as mobile phases. Mobile phase composition started at 15% acetonitrile, raised to 35% over 2.5 min to a final composition of 100% acetonitrile at 3.5 min. The mobile phase composition was held at 100% acetonitrile for 2.0 min before returning to 15% acetonitrile.

The following transition ions were used for mass spectrometric quantification with a polarity switch using a Sciex API 5500 (Framingham, MA, USA):

• Bimatoprost → m/z 416.3 → 362.3 (positive ion)

• Bimatoprost‐d⁴ (IS) → m/z 420.3 → 366.3 (positive ion)

• Bimatoprost acid → m/z 387.4 → 193.0 (negative ion)

• Bimatoprost acid‐d⁴ (IS) → m/z 391.4 → 197.0 (negative ion)

Solutions of bimatoprost and bimatoprost acid at a concentration range of 5–5000 pg/mL and 0.5–500 pg/mL, respectively, were used as calibration solutions. A linear regression (weighted 1/x ²) was applied to a plot of the peak area ratio versus concentration for the standards to obtain the calibration curve.

Blank control human skin sections were spiked with bimatoprost and bimatoprost acid to give target QC sample concentrations of 3, 30, 375 pg/mL of bimatoprost and 30, 300, 3750 pg/mL of bimatoprost acid in human skin. Additional dilution QC samples were prepared and used at concentrations of 75,000, 187,500, and 375,000 pg/mL of bimatoprost in five batches. All QC samples were stored at −70°C.

Human skin samples were prepared for analysis by first soaking in 1 mL 50:50 ethanol/water with gentle shaking for 12 h at 8°C. Samples were stored at −70°C until analysis. Analytes were isolated from an aliquot of the skin soaking solution by liquid–liquid extraction using MTBE. After vortexing and centrifugation, the aqueous phase was frozen in a dry ice bath and the organic phase was collected and dried under a nitrogen gas stream. The sample was reconstituted in 100 μL of 20:80 ethanol/water, then injected for LC–MS/MS analysis.

2.6. Matrix‐Assisted Laser Desorption/Ionization (MALDI)‐MS Imaging Method

The 4 skin biopsy samples from one participant were embedded in 2% carboxymethyl cellulose and sent to ImaBiotech (Lille, France) for MALDI‐MS Imaging. Representative 10 μm vertical sections were collected from each sample using a MicromHM560 (Thermo Scientific, Waltham, MA, USA) cryostat set to −20°C, and mounted onto Indium Tin Oxide coated glass slides desiccated, treated by spraying a solution of 40 mg/mL 2,5‐dihydroxybenzoic acid (DHB) and 2 mg/mL lithium trifluoroacetate in 1:1 methanol/water and analyzed using a 7 T‐MALDI‐FTICR (Solarix, Bruker Daltonics, Bremen, Germany) in positive mode with a Continuous Accumulation of Selected Ions (CASI)‐mode in order to improve the analytical sensitivity. The bimatoprost lithium adduct (M + Li)⁺, m/z 422.2881 was monitored to map the distribution of bimatoprost in the skin sections. The resulting MALDI‐MS imaging data were analyzed using DataAnalysis 4.1 software (Bruker Daltonics, Bremen, Germany). Adjacent sections were collected and H&E stained to generate histology images for comparison.

2.7. Data Analysis

The safety parameters included adverse events and clinical laboratory, vital sign, 12‐lead ECG parameters, and local dermal tolerability assessment. For each safety parameter, the last nonmissing observation collected before the first dose of study treatment was used as the baseline value for the corresponding safety analyses. Continuous variables were summarized by number of study participants and mean, standard deviation from mean (SD), median, minimum, and maximum values. Categorical variables were summarized by frequency counts and percentages of study participants.

Skin sample concentrations of bimatoprost and bimatoprost acid were summarized by formulation and by biopsy location, plotted as a function of depth from the skin surface, and descriptive statistics calculated when applicable. Phoenix WinNonlin version 6.3 (Princeton, NJ, USA) and GraphPad Prism (San Diego, CA, USA) software were used for the graphing of pharmacokinetics data when applicable.

For the LC–MS/MS method, means, standard deviations, inter‐batch precision (%CV), and accuracy (%RE) of bimatoprost and bimatoprost acid concentrations of quality control samples and calibrators were calculated using Excel 2016. Means and standard deviations of the calibration curve parameters were also calculated by Excel.

3. Results

3.1. Patients' Demographics

Demographic and baseline characteristics are presented in Table 1. The mean age of the safety population was 38.3 (range 29–49) years. The majority of study subjects were White and not Hispanic/Latino. The mean weight was 99.37 kg, and the mean height was 178.26 cm. As specified in the inclusion criteria (Supporting Information), all study participants were male and had baseline AGA assessed as the Norwood–Hamilton classification of III, IV, or V.

TABLE 1.

Demographics and other baseline characteristics.

Parameters	Total (N = 11)
Age (years)
Mean (SD)	38.3 (6.6)
Minimum, maximum	29, 49
Race, n (%)
White	9 (81.8)
Non‐White	2 (18.2)
Black or African American	1 (9.1)
Native Hawaiian or Other Pacific Islander	1 (9.1)
Weight (kg)
Mean (SD)	99.37 (17.31)
Median	95.30
Minimum, maximum	73.7, 132.0
Height (cm)
Mean (SD)	178.26 (6.62)
Median	177.80
Minimum, maximum	167.6, 188.0
Modified Norwood–Hamilton scale class, n (%)
III	4 (36.4)
IV	2 (18.2)
V	5 (45.5)
Fitzpatrick Skin Phototype assessment, n (%)
I	0
II	2 (18.2)
III	5 (45.5)
IV	2 (18.2)
V	2 (18.2)

Note: The entire study course in demographic analysis started from the screening visit through the study exit up to the posttreatment follow‐up visit. Study participants who participated in the study anytime during the study course are included in the analyses.

Abbreviations: N, number; SD, standard deviation.

3.2. Safety

Table 2 provides an overview of the safety outcomes on the safety population, which was defined as all study participants who received at least 1 dose of study treatment. No deaths or other serious adverse events (SAEs) were reported during the study. There were no discontinuations from study or treatment due to adverse events. Adverse events were reported in 4 of 11 (36.4%) study participants during the entire study course. All the adverse events were treatment‐emergent adverse events (TEAE) (i.e., adverse events that began or worsened on or after treatment started). No study participant had treatment‐related TEAEs. All adverse events reported were of mild severity except increased transaminases in one study participant, which was moderate severity. This event happened on Day 28 (2 weeks after the last dose of the study treatment.) and was not considered treatment‐related. Alanine and aspartate aminotransferase levels were elevated (131 and 172 U/L, respectively) during a follow‐up visit on Day 40 but appeared to approach normalization (73 and 46 U/L, respectively) on Day 55.

TABLE 2.

Overall summary of adverse events in the safety population.

	Treatment period a (N = 11), n (%)	Posttreatment period b (N = 11), n (%)	Entire study course c (N = 11), n (%)
Study participants with at least 1 AE	2 (18.2)	2 (20.0)	4 (36.4)
Study participants with at least 1 TEAE d	2 (18.2)	2 (20.0)	4 (36.4)
Study participants with at least 1 treatment‐related TEAE d	0	0	0
Study participants with at least 1 serious TEAE e	0	0	0
Deaths f	0	0	0
Study participants with at least 1 SAE f	0	0	0
Study participants with at least 1 adverse event leading to discontinuation f	0	0	0

Note: Study participants were counted only once within each category.

Abbreviations: AE, adverse event; SAE, serious adverse event; TEAE, treatment‐emergent adverse event.

^a The treatment period started.

^b The posttreatment period started after Day 14 visit through Day 28/Early Exit visit (if applicable).

^c The entire study course in the adverse event analysis started from the receipt of the first dose of study medication through the study exit up to the posttreatment follow‐up visit.

^d Events that began or worsened on or after the receipt of the first dose of study medication.

^e Events that occurred or worsened on or after the receipt of the first dose of study medication.

^f Events that occurred at any time.

3.3. LC–MS/MS Quantitative Analysis

The qualified LC–MS/MS method was used for the quantification of bimatoprost and its acid metabolite from 377 human skin samples. Samples were analyzed in 10 batches for bimatoprost and five batches for bimatoprost acid. The method with an assay range of 5–5000 pg/mL for bimatoprost and 0.5–500 pg/mL for bimatoprost acid met predetermined acceptance criteria. All samples, regardless of skin layer, were analyzed with the same bioanalytical method.

To accurately represent the amount of bimatoprost and bimatoprost acid in human skin, a correction factor of 6.28 was applied to normalize the concentrations, calculated based on the approximate volume of the biopsy punch at 4 mm diameter and 0.5 mm depth. Therefore, the LLOQ was 0.08 ng/g and the upper limit of quantification (ULOQ) was 80 ng/g for bimatoprost in human skin. The LLOQ was 0.8 ng/g and the ULOQ was 800 ng/g for bimatoprost acid in human skin.

3.4. Pharmacokinetics, Staining, and MALDI‐MS Imaging

Skin biopsies were collected from 9 participants. All dosing time deviations occurred within 1–8 min earlier than the protocol‐allowable window (±30 min) except for one patient on Day 2 and another patient on Day 8 (All dosing time deviations occurred within 1–8 min earlier than the protocol‐allowable window (±30 min)). Since participants were dosed daily for 14 days and the samples were collected on the last day of dosing, it can be assumed that steady‐state was reached in all nine participants included in this analysis and that these deviations did not have any significant impact on the pharmacokinetics results. Due to individual differences in skin layer thickness and intersubject variability in pharmacokinetics (e.g., local drug metabolism), formulation performance comparison is more reliable based on data collected from within the same participant rather than comparing mean data across all participants.

For histological assessment, 7–14 sections were obtained from each biopsy depending on the sample thickness (one section every 500 μm). Histology images for the left occipital scalp biopsy of one subject are shown in Figure 2, beginning with the deeper subcutaneous fat layer [1] and extending, as the sections get closer to the skin surface, to the last section collected in the dermis [7]. Several hair follicles/shafts are visible in these cross‐section images (starting in Image 3 through the dermis layer). Sections between these histology sections were analyzed for concentrations of bimatoprost and bimatoprost acid.

FIGURE 2.

Patient skin biopsy collected from the left occipital scalp. Individual H&Es from each biopsy are shown in order, beginning with the deeper subcutaneous fat layer [1] and extending to the last section collected through the dermis, just underneath the epidermis [7].

Mean concentrations of bimatoprost from all participants across various depths from the epidermis surface for both formulations and dosing sites (scalp and trunk) are shown in Figure 3. Both formulations showed a significant decrease in bimatoprost concentration with increasing distance from the skin surface as would be expected for a topically‐applied formulation. Other than a few data points in the scalp sample comparison, overall Formulation B appeared to provide a slightly greater bimatoprost exposure in both the scalp and trunk biopsies across the depth of the biopsied skin samples (Figure 3).

FIGURE 3.

Mean concentrations of bimatoprost relative to the distance from the epidermis surface in samples from scalp (A) and trunk (B) after administration of Formulation A and B. Data points marked in red are where the hair follicle is approximately located.

The ratios of bimatoprost concentration of Formulation B relative to Formulation A in scalp samples in individual study subjects suggested no significant penetration differences between the two formulations in the epidermis/dermis layer. However, in subcutaneous fat, bimatoprost concentration was trending higher following the administration of Formulation B compared to Formulation A with a median ratio of ~2, although the values are more scattered across two orders of magnitude than in epidermis/dermis (Figure 4).

FIGURE 4.

Ratios of bimatoprost concentrations in scalp samples following topical administration of Formulation B relative to Formulation A.

Most of the samples had bimatoprost acid concentrations below LLOQ (BLQ) for both formulations. The measurable concentrations of bimatoprost acid in skin sections were several orders of magnitude lower than concentrations of bimatoprost and are presumed to be pharmacologically negligible.

For the biopsy samples collected from one patient, H&E staining and MALDI (at 50 μm spatial resolution) analysis were performed. Bimatoprost was detected in the epidermis and in the uppermost dermis of the scalp and trunk. It was particularly detected in the hair shafts located in the epidermis layer. However, bimatoprost could not be detected in the hair shaft located in the dermis/subcutaneous fat layers due to inadequate assay sensitivity (Figure 5). Lack of higher signal for Bimatoprost at the edges of the biopsy section at the dermis and fat layers compared to the center of the image indicated that the measures taken to avoid contamination during biopsy collection and sectioning appeared to be effective, at least to the extent that these methods could demonstrate.

FIGURE 5.

Bimatoprost distribution in treated tissue (sample from one patient).

4. Discussion

AGA in men is a common condition characterized by hairline recession associated with thinning and/or balding of the crown and/or vertex areas of the scalp [11]. Men with AGA are impacted emotionally and psychosocially by their condition [12]. There is no cure for AGA, and minoxidil (Rogaine/Regaine) and finasteride are the only FDA‐approved pharmacological interventions [13]. Cosmetic products, which disguise the area of hair loss, to more invasive surgical techniques, such as hair transplantation and scalp reduction can be used to enhance the appearance of patients [14], but often lead to poor and unsatisfying results or are invasive and carry risks.

As the first member of the prostamide drug class, bimatoprost is a synthetic compound similar to PGF2α in both its pharmacological properties and structure [15]. Studies have shown that bimatoprost stimulates hair growth and lengthens eyelashes, without creating new hair follicles. It influences the eyelash growth cycle by increasing the proportion of follicles in the anagen (growth) phase, encouraging telogen (resting) follicles to reenter the growth phase, and keeping follicles in the anagen phase longer by preventing their transition to the catagen (regression) phase. Additionally, bimatoprost thickens eyelashes by increasing the diameter of the hair bulb [16].

The safety and efficacy of bimatoprost for the treatment of hypotrichosis of the eyelash were investigated in several clinical trials. In a Phase 3 randomized vehicle‐controlled clinical trial including 278 subjects with eyelash prominence, after 4 weeks of treatment with either 0.03% bimatoprost ophthalmic solution or vehicle, at 8‐week, 12‐week, and 16‐week, bimatoprost was significantly more effective than vehicle as measured by the GEA [9].

The current study evaluated the skin pharmacokinetics and local (scalp, trunk) dermal tolerability of 1% bimatoprost following 14 days of once daily topical administration in Formulations A and B in male patients with AGA. Formulation A at concentrations up to 3% has been previously established to be safe and well‐tolerated, based on the data from four completed clinical studies (one Phase 1 pharmacokinetic study) [17] and three 6‐month Phase 2a safety and efficacy studies [18, 19, 20]. Pharmacokinetic analyses in these earlier studies indicated low and dose proportional plasma concentrations after topical administration of Formulation A. Formulation B was developed to optimize delivery of bimatoprost to the hair follicle. The local pharmacokinetics of Formulation B was evaluated along with Formulation A. Biopsy samples from the lateral trunk area were also collected in this study because scalp biopsy sample collection for pharmacokinetics assessment is not routine practice and data may be confounded by other factors such as greater bleeding potential.

Pharmacokinetic analyses in the current study revealed no significant differences in bimatoprost concentrations between Formulations A and B in the epidermis/dermis layer of the scalp punch biopsy samples from the same study participants. However, in subcutaneous fat layers of the scalp, bimatoprost concentrations after administration of Formulation B were trending up to 2‐fold higher than after administration of Formulation A in the same study participants. Both formulations at 1% achieved at least a 10‐fold higher bimatoprost concentration in subcutaneous fat, where the hair papilla are typically situated (presumed drug target for bimatoprost), than effective in vitro concentration needed for maximal increased hair growth (42 ng/g) isolated hair follicles [21]. This finding suggests that with topical application of bimatoprost 1% Formulation A and Formulation B sufficient drug levels were achieved at the target site to engage the pharmacological target. However, the clinical relevance of the bimatoprost in vitro hair growth potency data still needs to be determined as the drug bioavailability may differ under physiological condition.

Overall, the horizontal skin sectioning method utilized in this study proved useful to evaluate pharmacokinetics and local drug distribution in skin following topical dermal administration. Although other methods to study skin pharmacokinetics, including non‐invasive methods such as confocal RAMAN spectroscopy [22] have been reported, they do not offer the combination of advantages of high sensitivity, spatial (depth) resolution of 0.5 mm and absolute quantification at skin depths relevant for hair growth products. The potential for mass spectrometry imaging methods such as MALDI‐MS imaging to study drug biodistribution in skin is also promising, but faces limitations such as difficulty in quantification and low sensitivity, as demonstrated by lack of signal for bimatoprost at relevant skin depths. In addition, Birngruber et al. suggest that dermal open flow Microperfusion (dOFM) method is a robust way to assess unbound levels of topically‐administered drugs directly in dermal interstitial fluid. These studies depict that this method can provide standardized, continuous sampling, and can reduce the need for large clinical endpoint studies [23]. Bodenlenz et al. compared dOFM and dermal biopsy methods to measure seven different topical drug concentrations, using a dOFM pig model. They concluded that directly measuring the drug concentrations in interstitial fluids is more reliable than dermal biopsies [24]. Taken together, these findings suggest that dOFM is a strong complementary method for evaluating skin drug delivery and can be especially useful alongside imaging approaches such as MALDI‐MS by providing quantitative concentration‐time data in the relevant dermal compartment.

Local (scalp, trunk) dermal tolerability assessments by the study participant and investigator did not reveal any clinically significant findings or clinically meaningful differences between Formulations A and B. Findings were primarily of mild severity and originated exclusively in the bimatoprost application areas of the scalp, with folliculitis as the most frequent finding. It did not appear that folliculitis was related to the introduction of the new formulation as it occurred at equal rates between Formulations A and B. Bimatoprost was found to be safe and well tolerated in this population of adult males with AGA. All TEAEs were mild and not considered treatment‐related with no report of deaths or other SAEs. There were no discontinuations from study or treatment due to adverse events.

In conclusion, following once‐daily administration of bimatoprost 1% Formulations A and B for 14 days in male participants with AGA, there were no significant differences observed in bimatoprost concentration between Formulation A and Formulation B in the epidermis/dermis layer of the scalp punch biopsy samples from the same study participants. Following administration of Formulation B, bimatoprost concentrations in the subcutaneous fat layers of the scalp were trending up to 2‐fold higher than after administration of Formulation A in the same study participants. Both formulations resulted in concentrations well above what would be needed for pharmacological effect at the depth of the target tissue (hair papilla) based on in vitro potency data. In most of the samples bimatoprost acid concentration was BLQ for both formulations, suggesting that this metabolite is unlikely to contribute to the pharmacology of topical bimatoprost. Local (scalp, trunk) dermal tolerability profiles did not differ between application areas receiving Formulation A versus B. Bimatoprost Formulations A and B were safe and tolerated based on the adverse event profile. This translational research provides insight into both the pharmacokinetics and local tolerability of different bimatoprost formulations which is highly valuable for informing product development. Overall, these findings help guide formulation selection and adjustment, reducing development risks, increasing the likelihood of product development success, and maximizing the efficacy and safety.

Author Contributions

S.K. wrote the manuscript; J.S., J.R., and M.A. designed the research; J.S., M.A., J.R., L.B., and W.C. performed the research; J.S., M.A., W.C., and S.K. analyzed the data.

Funding

Allergan, an AbbVie company, funded this research, analysis, data collection, interpretation of data, and the review and approval of the publication. All authors had access to relevant data and participated in the drafting, review, and approval of this publication. No honoraria or payments were made for authorship.

Conflicts of Interest

S. Khajeh pour, J. Rowe, L. Borbridge, and J. Shen are employees of AbbVie and may hold AbbVie stock and/or stock options. W. Chen and M. Attar were employees of AbbVie and may hold AbbVie stock. M. Attar is currently an employee of Bausch & Lomb. W. Chen was a former AbbVie employee, and has no conflicts of interest.

Supporting information

Data S1: cts70710‐sup‐0001‐Supinfo.pdf.