Topical dutasteride for androgenic alopecia: current state and prospects

ABSTRACT

Androgenic alopecia has a high incidence, affecting 80% of men and 50% of women in their lifetimes. Although not a life-threatening disease, it can be a deep psychological burden to patients and still lacks an effective and safe treatment. Dutasteride is a5-alpha-reductase inhibitor approved to treat benign prostatic hyperplasia that is also commonly prescribed off-label to treat androgenic alopecia. However, oral dutasteride may cause several severe sexual and neurological sideeffects. Therefore, an effective, localized dutasteride treatment that can reduce the effects of systemic uptake is of great interest. Here, we review available therapies to treat androgenic alopecia focusing on topicalformulations developed thus far–including minoxidil, finasteride, and cosmetics–and on dutasteride-loaded nanocarriers targeting hair follicles.

TWEETABLE ABSTRACT

Dutasteride has gained popularity as a potent alternative to finasteride for treating alopecia. However, developing a safe and effective commercial product is challenging. We seek to identify a follicle-specific delivery vehicle that reduces dutasteride’s systemicabsorption.

1. Introduction

Androgenic alopecia is the most prevalent type of alopecia, affecting up to 80% of men and 50% of women in their lifetimes [1]. The condition is polygenetic, characterized by the miniaturization of hair follicles and shortening of the anagen phase, in part caused by elevated intrafollicular dihydrotestosterone (DHT) production and excessive response to androgens due to the high numbers of androgen receptors in the scalp [2,3]. Androgenetic alopecia in men typically presents a concentrated loss of hair around the temporal-frontal and crown regions of the scalp but can also be diffused throughout the entire scalp [4]. Meanwhile, androgenetic alopecia in women is almost always diffused throughout the whole scalp [5].

Alopecia causes a considerable psychological burden on patients, with over 40% of patients self-reporting increased anxiety [6] and decreased self-esteem [7]. Although the condition is both prevalent and impactful, treatment options are limited. Current FDA-approved treatments are restricted to topical minoxidil and oral finasteride. Both therapies have significant limitations as it will be further discussed, and an optimal alternative would be a topical treatment to treat androgenic alopecia. However, for optimal efficacy, such a topical treatment should accumulate the drug within the hair shaft milieu in a targeted manner, such that the drug is minimally absorbed by surrounding skin sites, like the viable epidermis. If intrafollicular targeting were achieved, the amount of drug that reaches local blood circulation and, consequently, systemic uptake could be limited. The challenge is obtaining such a formulation capable of targeting only the hair follicles.

Hence, we aim to review all topical treatments currently available to treat androgenic alopecia and the formulations developed and studied thus far, with a focus on how dutasteride has been explored in this context.

2. Hair follicle anatomical structure

Understanding the physioanatomy of the hair follicle is essential to design topical products targeted at this structure (Figure 1). The hair follicle is an appendage of the skin, originating in the dermis but extending upward through the epidermis and stratum corneum to the external world. The hair follicle can be divided into three sections: infundibulum, isthmus, and inferior region. The infundibulum is the outermost region, from the follicle opening of the skin down to the sebaceous duct. The isthmus is the middle region, from the sebaceous duct to the arrector pili muscle. The inferior region is the bottom region, from the arrector pili muscle to the base of the follicle, also referred to as the hair bulb. The hair bulb is the main region of interest for stimulating hair growth, as it includes the dermal papilla, connective tissue sheath, and hair matrix niches, which collectively comprise specialized fibroblasts, rapidly proliferating keratinocytes, blood capillaries, and nerve endings.

Figure 1.

Schematic illustration of the anatomical structure of a hair follicle. Own authorship. Created with adobe Illustrator®, version 27.7.

DHT, the sex hormone suppressed by 5-alpha reductase inhibitors, and testosterone, from which DHT is metabolized, are part of the androgen family. The dermal papilla is considered the main site of androgens’ binding to intracellular receptors and enzymes, causing alterations in the dermal papillas’ intracellular signaling and consequently altering its secreted factors that regulate its neighboring hair matrix keratinocytes, ultimately affecting the hair growth cycle. Hence, for a local effect following a topical application, the active drug must reach the bulb region, where the dermal papilla resides. Nonetheless, androgen receptors and 5-alpha-reductase isoenzymes, type I and type II, are present in other locations of the hair follicle. Literature has described androgen receptors and 5-alpha-reductase type I as uniformly spread throughout the hair follicle, while it presents conflicting data for 5-alpha-reductase type II. 5-alpha-reductase type II has been primarily found in the inner and outer root sheath of the epithelium, and, to a lesser extent, in the matrix cells of hair bulb in immunohistochemistry analyses of in-tact human scalp skin [8–10]. However, RNA-seq analysis of regions or single cells of microdissected human hair follicles illustrates 5-alpha-reductase type II as only present in the dermal papilla and surrounding connective tissue sheath [11], with the isoform upregulated only in the dermal papilla but normalized in the dermal sheath of miniaturizing follicles [12]. The discrepancy of 5-alpha-reductase read-outs throughout studies may be due to the length of time between tissue extraction and freezing or general tissue handling, as well as the accuracy of the antibodies and RNA-seq platforms used.

3. Current solutions

There exists extensive clinical reference to minoxidil’s effect on hair follicle growth [13–16]. First administered orally to treat hypertension in the 1970s, minoxidil’s discovery as an androgenic alopecia treatment was serendipitous [17]. The observed side effect of hypertrichosis led to the development of its first topical formulation in 1987. In vitro studies have identified several potential mechanisms of minoxidil’s pro-growth effects: potassium channel opening and resultant relief of mitochondrial calcium overload, hepocyte growth factor upregulation, vascular endothelial growth factor stimulation [18–21], enhanced prostaglandin synthesis [22,23], and nuclear ß-catenin accumulation [24], the combination of which results in stem cell proliferation [25–28] and promotes the growth of thicker and denser hair shafts. However, while minoxidil extends the anagen phase, it does not interrupt the balding process. Patients begin to lose hair again immediately after treatment cessation and return to baseline within 24 weeks of discontinuation [29]. Moreover, minoxidil’s effects require continuous treatment and outperform control groups in about 30% of patients only [29,30]. Hence, topical minoxidil treatment is not entirely effective and presents challenges for medical practices.

Several of topical minoxidil’s challenges can be attributed to its difficulty in achieving a targeted follicle delivery. Minoxidil has a water solubility of around 2.0 mg/mL, which requires vehicles with a high amount of ethanol or propylene glycol to achieve the prescribed concentrations of 2 or 5%. Such vehicles can be highly irritating to the skin following continuous application [30–32]. Moreover, when formulated in a volatile solvent, such as ethanol and propylene glycol, minoxidil molecules can crystallize on the skin’s surface [33], especially if applied on a dry scalp. In fact, minoxidil applied in a damp rat hair model resulted in five times higher hair follicle retention of the drug (p < 0.05) than in a dry rat hair model and was corroborated in the porcine skin model, with an 8-fold increase in minoxidil hair follicle accumulation in the damped model compared to the dried model [34]. Accordingly, the humidity condition of the scalp has been identified as a critical factor in maximizing minoxidil’s skin diffusion and its resultant efficacy for androgenetic alopecia patients.

Another potential way to circumvent minoxidil’s solubility problem is to administer minoxidil sulfate instead of the base form. Minoxidil sulfate is the active metabolite of minoxidil, 14 times more potent than minoxidil itself [35]. Despite its significantly higher water-solubility (20 mg/mL), minoxidil sulfate is unstable in both water and volatile solvents, compromising the drug’s bioavailability in topical administration [36]. Of note, low-dose oral minoxidil (<5 mg daily) has also achieved recent popularity in the treatment of several forms of alopecia, including androgenic alopecia. However, oral minoxidil is also accompanied by several severe, dose-dependent side effects, such as hypotension, heart palpitations, pericardial effusion, and edema [37–39]. Thus, topical minoxidil’s lack of efficacy and solubility, minoxidil sulfate’s compromised stability, and oral minoxidil’s risk of side effects each present challenges in medical administration, and the need for a more suitable topical treatment for androgenic alopecia remains.

Finasteride is more effective than minoxidil and acts by a different, known mechanism, inhibiting type II 5-alpha reductase, and consequently hampering the conversion of testosterone to DHT. Accordingly, finasteride lowers systemic DHT levels by 70% [29,40,41]. Dutasteride is a second-generation 5-alpha reductase inhibitor that suppresses both type I and II 5-alpha reductase isoenzymes, consequently suppressing systemic DHT levels by 98.4% [42,43]. Like finasteride, dutasteride was initially approved to treat benign prostatic hyperplasia. However, several observational studies have investigated dutasteride to treat alopecia with promising results [42–46], producing superior efficacy to finasteride in hair regrowth [42,47]. Both finasteride and dutasteride must be administered in perpetuity to maintain their effectiveness. Both drugs also present significant sexual side effects, such as a decrease in libido, ejaculatory and erectile dysfunction, sexual impotence, and gynecomastia, as well as significant neurological side effects, such as anxiety and depression, all of which can persist up to 4 years after treatment cessation [41,48–52]. The combination of continued treatment and risks of side effects presents challenges in medical practice for both drugs. Therefore, a localized topical 5-alpha reductase inhibiting therapy would be more suitable to minimize side effects while maintaining a maximally effective treatment. Indeed, a topical application could be convenient for patients, given a proper nonirritant formulation is used.

Although there exists no FDA-approved topical finasteride treatment today, a topical spray is approved in some European countries by the European Medicines Agency (EMA) to treat androgenic alopecia. It is worth mentioning finasteride and dutasteride medicines are now under review by EMA due to the reports of suicidal ideations after 5-alpha reductase inhibitors usage.

Regardless, compounding pharmacies have produced topical formulations containing finasteride in a plethora of concentrations and vehicles. Considering finasteride’s low water-solubility (0.00198 mg/mL), it has been explored in vehicles containing high amounts of alcohol and/or propylene glycol-like minoxidil. For example, a hydroalcoholic solution containing finasteride 0.1%, minoxidil 10%, biotin 0.2%, and caffeine citrate was applied to male patients for six months in a dose of 1 mL twice daily in a clinical trial. The trial included monitoring of patients for 180 days post-treatment cessation. The formulation provided moderate results, as evidenced by hair-thickening, increased scalp coverage, and improved general hair appearance [53]. Despite the vehicle containing not only ethanol but also propylene glycol, which, as mentioned, is not appropriate for prolonged use, no adverse effects were registered in the six months. Nonetheless, it was impossible to determine how much finasteride contributed to the outcome in this context, as no controls were used for comparison to the treated group. The synergic effect of topical finasteride and minoxidil has also been suggested with the use of a topical solution of 0.25% finasteride and 3% minoxidil, which was described to generate a 1.7-fold improvement in hair density and a 1.3-fold increase in hair shaft diameter, compared to a topical formulation of only 3% minoxidil in a 6-month clinical trial with 40 men [54]. Although the formulations’ compositions were not detailed, adverse events such as dry and flaky scalp, headache, and scalp pruritus in some patients suggest an inadequate product for sustained use [54].

Topical formulations containing only finasteride as active ingredient have also been assessed. A 1 mL hydroxypropyl chitosan solution with 0.25% finasteride applied twice daily was compared to a 1 mg oral tablet once daily in a 7-day observational study with 24 men. Even though the topical finasteride resulted in a lower plasma exposure 24 hours after its administration, both formulations produced the maximum effect after multiple doses: a reduction in DHT levels of around 70%, without significant differences [55]. Although no relevant adverse event was identified to be drug-related, the systemic decrease in DHT levels indicates the potential for systemic adverse effects. Accordingly, in a later observational study, the same 0.25% finasteride formulation was assessed in smaller doses of 100 to 400 µL to reduce the systemic effect on DHT levels while maintaining a similar local effect. Doses as small as 100 and 400 µL applied once daily lowered scalp DHT levels by 47 and 52%, respectively, while the decrease in systemic DHT levels was 24% for the first dose and 26% for the second one, although the study only lasted one week and thus could not take into account the slow accumulation of finasteride throughout the body over time [56]. A subsequent observational study of 50–200 µL applied once daily was conducted in a broader population of 446 patients and for a more extended period of 24 weeks, producing a 35% drop in DHT from baseline, although the observers did not state what percentage of patients used which dose range. A 50 µL dose would only equate to 0.14 mg of finasteride, misleading readers of the localized effect of the topical formulation used. Hence, the study has major limitations and methodological concerns [57]. While there may have been a significant improvement in reducing the impact on serum DHT levels, it is not possible to know if the study’s endpoints were due to only a decrease in time length or dose used. Even so, the risk of sexual adverse events remains with a 24–35% DHT systemic reduction.

In a different approach, finasteride has also been explored in dissolving and implantable microneedle patches. In this kind of system, biopolymers like hyaluronic acid are molded into microneedles shape, combining the features of a skin patch and subcutaneous injection, which allows for the drug to be delivered directly to the dermal layers of the skin [58]. Dissolving microneedles can not only stimulate the dermal papilla cells by themselves but a couple of studies demonstrated their capability to increase finasteride accumulation into hair follicles during in vitro skin penetration studies in porcine skin and in vivo tests in mice [59,60]. Despite the promising results, finasteride permeated through the skin in both cases, which keeps the risk of sexual adverse events high. Another limiting factor in the clinical use of microneedles is their reduced size to cover larger areas like the scalp of androgenic alopecia patients. In this sense, an implantable microneedle device was successfully developed for the systemic delivery of finasteride and showed a sustainable release for over 14 days [61]. As much as this can be considered a viable option for the administration of oral finasteride, it does not solve the problem of unwanted systemic side effects. So, there remains room for the development of better formulations with less systemic effects while improving local efficacy.

The cosmetic industry has produced uncountable over-the-counter cosmetics for hair loss, mainly containing botanical extracts, oils, minerals, and vitamins [2,62,63]. Unfortunately, the efficacy of these products is insufficiently proven. Examples include mainly rosemary and saw palmetto oils, which are both 5-alpha reductase inhibitors [64,65], and pumpkin seed oil, which has anti-inflammatory properties and increases microcapillary perfusion [66]. Despite these characteristics, and the rising interest in such oils [67–69], aside from not being FDA-regulated, there is a lack of data in the literature about their efficacy and safety for androgenic alopecia treatment.

Biotin (vitamin B7), the most used vitamin to improve hair conditions, is primarily taken orally. The vitamin was incorporated in a topical formulation previously discussed, although cannot be claimed as causal due to the other ingredients used [53]. Biotin was also present in another topical formulation with redensyl and saw palmetto, which, when combined with platelet-rich plasma (PRP) therapy, produced positive results in androgenic alopecia treatment [70]. However, no studies have been conducted to analyze the efficacy of topical biotin alone in treating androgenetic alopecia. These proposed alternative treatments marketed as over-the-counter cosmetics often show trichogenic effects in vitro at concentrations too high to be delivered in vivo, as it would require constant application throughout the day. Moreover, their low target specificity may cause potential side effects when used at higher dosages or for prolonged periods.

Aside from over-the-counter cosmetics, other physical therapies, such as fractionated laser therapy, low-level light treatment, and platelet-rich plasma, are employed as adjuvant therapy [71]. Platelet-rich plasma, especially, is becoming a more popular treatment option for androgenic alopecia. This therapy consists of an autologous blood product rich not only in platelets but also containing platelet-derived growth factors, which promote cell proliferation, angiogenesis, and cell differentiation, mechanisms involved in hair regrowth. Different clinical trials showed a similar or higher efficacy of monthly intradermic injections of platelet-rich plasma compared to a daily application of topical minoxidil solution [72–74]. The outcomes of such clinical trials are often measured by hair density and regrowth. Platelet-rich plasma has also showed to be beneficial during and after hair transplantation surgery [75,76]. Aside from being a painful therapy, there is no standard protocol a systematic review and meta-analysis considered the evidence of the benefits of platelet-rich plasma therapy for treatment of hair loss is still of low quality due to risk of partiality and inconsistencies [77].

4. Dutasteride for alopecia topical treatment

In the presented scenario, dutasteride is a promising yet minimally explored alternative for the topical treatment of androgenic alopecia. Dutasteride can suppress type I 5-alpha reductase by 100-fold and type II 5-alpha reductase by 3-fold when compared to finasteride [21,47,78]. Thus, dutasteride can be administered at a lower dose and frequency than finasteride, while remaining more potent. But one point of concern is dutasteride’s much longer half-life than finasteride: 5 to 6 weeks compared to 5 to 6 hours, which may increase the risk of side effects. Although this may hold to a typical oral therapeutic dose, a study showed dutasteride’s half-life behavior is dose-dependent, at lower concentrations, dutasteride’s elimination rate is non-linear, thus, dutasteride’s half-life can be shorter in smaller drug concentrations [79], minimizing side effect risks.

Dutasteride is a very lipophilic molecule (Figure 2), practically insoluble in water (log p = 5.09 and water solubility = 0.038 ng/mL). While the high lipophilicity of dutasteride improves penetration through the lipophilic stratum corneum, it worsens dutasteride penetration through the deeper layers of the skin. Dutasteride’s lipophilicity also hampers its incorporation into conventional topical drug delivery systems. As such, topical dutasteride solutions must use an oily or viscous formulation for penetration. However, such a formulation often results in a greasy hair appearance, lowering patient compliance [80]. There are evidences of mesotherapy, i.e., intradermic injection, of dutasteride being successfully employed in androgenic alopecia treatment, however, mesotherapy can be considered a painful treatment, also it presents a risk of drug absorption by adjacent tissues, which increases the potential to cause side effects [81–83]. So, alternative approaches involving drug nanoencapsulation are of utmost interest to circumvent these issues.

Figure 2.

Dutasteride chemical structure (MW: 528.53 g/mol).

4.1. Nanoparticles and dutasteride hair follicle targeting

In the skin’s structure (Figure 3), the stratum corneum acts as a barrier to the skin’s lower layers. Nanoparticles are “too big” to penetrate the stratum corneum but naturally accumulate within the hair follicle, rendering them exceptionally useful in targeting hair follicles when topically applied. As such, nanotechnology has been widely employed in the pharmaceutical field to produce more efficient drug delivery systems with fewer adverse effects for androgenetic alopecia treatment [84–91].

Figure 3.

Scheme showing the targeting of topically applied nanoparticles to the hair follicle. Own authorship. Created with adobe Illustrator®, version 27.7.

Hair follicles are permeable to nanoparticles of a wide range of sizes, from 50 nm to over 900 nm [92,93]. If properly designed, topically applied nanocarriers tend to accumulate in hair follicles. Furthermore, modulating the size of nanocarriers opens the possibility of further targeting specific regions in the follicle. A classic study in the delivery vehicle field studied the application of size-varying PLGA nanoparticles to porcine ear skin in vitro. While the smaller nanoparticles (122 nm − 230 nm) accumulated in the initial region of the terminal hair follicle, the larger nanoparticles (up to 643 nm) penetrated much more deeply (1400 µm in depth) [85]. However, above the 643 nm size, nanoparticle penetration depth significantly decreased.

Physical stimuli can also be an associated strategy to improve hair follicle targeting further. Studies show that a 3-minute massage is enough to enhance the follicular uptake of nanoparticles [80,84,94]. During the massage of a formulation onto the scalp, the structure of the hair shaft’s cuticles pushes the formulation’s particles deeper into the follicle in a phenomenon called the ‘ratchet effect.’ The ratchet effect is composed of radial and axial movements. The axial movement is responsible for enhancing the particles’ transport down the hair shaft, while the radial one is responsible for the particles’ ability to surround the hair shaft The radial and axial movements of the ratchet effect are most efficient at an ideal particle size, as particles smaller than 100 nm would not benefit from such effect [95].

Yet, for a targeted pharmacological action, the goal is to deliver the drug to its site of action and not the entire nanoparticle. Dutasteride must reach the 5-alpha reductase enzymes of the dermal papilla, even if the entire nanosystem does not. It can be expected that the interaction between the nanoparticle, hair shaft, and surrounding skin will cause the drug (dutasteride) to release from the nanoparticle and diffuse through the tissue, following a concentration gradient from the most concentrated to the less concentrated environment. It is important that the delivery vehicle can at least accumulate into the hair follicle shaft such that the drug, once released, can diffuse through the relevant hair follicle structures: first the hair bulge and outer root sheath, ultimately reaching the dermal papilla. By using a nanoparticle to target the hair follicle, instead of the rest of the skin, it is possible to avoid having the active pharmaceutical ingredient absorbed by the broader dermis and its attached local microcirculation, which feeds the drug into systemic circulation and could potentially lead to unwanted side effects.

To compare such an effect among topical formulations, skin permeation experiments, followed by differential extraction of the drug retained in the hair follicle and surrounding skin, must be run, the data from which allows us to calculate the targeting factor of the formulation. The hair follicle targeting factor is given by the following equation:

$$T_{\mathit{factor}}=\frac{\mathit{Drug}{\mathit{ }}_{\mathit{HF}}}{\sum \mathit{Drug}\mathit{penetrated}}$$

where “$\mathit{Drug}{ }_{\mathit{HF}}$” refers to the drug amount recovered from the hair follicles and “$\sum \mathit{Drug}\mathit{penetrated}$” refers to the sum of drug amount recovered from all skin layers: the stratum corneum, hair follicles, and viable skin, which comprehends epidermis and dermis.

Minoxidil and finasteride have been successfully loaded into nanocarriers to target hair follicles [28,96–98]. Nevertheless, concerning dutasteride-loaded nanocarriers, the scarcity of the literature indicates that it can be somewhat challenging to achieve such nanostructures (Table 1). Yet, the need for a more effective treatment for androgenic alopecia gives rise to increasing attempts. Researchers’ first choice to test topical dutasteride nanocarriers is usually one of several types of lipid-based nanoparticles, due to dutasteride’s lipophilicity. For example, dutasteride was successfully loaded in poly-(ε-caprolactone)-lipid core nanocapsules with and without chitosan-coating, with high encapsulation efficiency (>94%). In vitro skin penetration tests with porcine ear skin, post-6 hours after applying a massage as a mechanical stimulus and post-12 hours in passive conditions, showed the superiority of the nanoparticles in targeting hair follicles compared to the control group (p < 0.05), which was a dutasteride micellar water solution. The 12-hour skin penetration test showed bare nanocapsules enhanced dutasteride penetration in hair follicles compared to the chitosan-coated ones (dutasteride amount: 1.60 ± 0.35 µg/cm² vs. 0.70 ± 0.35 µg/cm²) and control, yet chitosan coating did not affect follicular targeting factor in either direction [80]. Chitosan is a biopolymer, often employed in nanoparticle coatings to enhance the contact between nanocarriers and biological tissue due to its mucoadhesive properties [99,100], which originate from its positive charge that interacts with surrounding tissue cell membranes’ negative charge, thus improving drug penetration [101]. However, chitosan may pose an additional barrier to drug release, which may explain why uncoated nanoparticles resulted in a higher dutasteride penetrated amount in this study.

Table 1.

Dutasteride-loaded nanostructured systems for topical treatment of androgenic alopecia.

Type of nanoparticles	Type of study	Main results	Reference
Poly-(ε-caprolactone) -lipid core nanocapsules with and without chitosan coating. ● Bare-nanocapsule:  – Size: 199.0 ± 0.5 nm  – Zeta potential: −13.6 ± 0.6 mV  – EE: 96.7 ± 1.8% ● Coated nanocapsule:  – Size: 224.9 ± 3.4 nm  – Zeta potential: 40.2 ± 0.8 mV  – EE%: 94.7 ± 3.0%	In vitro skin penetration tests with porcine ear skin. Drug delivery targeting to the hair follicles was assessed.	– After 6 h of skin treatment with a 3-min massage, both nanoparticles targeted and increased dutasteride penetration in hair follicles. – After 12 h, results showed bare nanocapsules benefited dutasteride penetration in hair follicles 2.2-fold compared to the coated ones and control. – Chitosan coating did not affect the follicular targeting factor.	[80]
Iron oxide nanocarriers. ● Dutasteride-loaded nanocarriers:  – Size: 117.7 ± 0.1 nm  – Zeta potential: −36.8 ± 0.45 mV  – EE%: 99.9 ± 0.1% ● Finasteride-loaded nanocarriers:  – Size: 116.4 ± 0.5 nm  – Zeta potential: −34.4 ± 0.3 mV  – EE%: 99.6 ± 0.1%	In vitro skin penetration tests with porcine ear skin. Drug delivery targeting to the hair follicles was assessed.	– Nanocarriers increased both drugs’ skin penetration compared to controls. – After 24 h, dutasteride- nanocarriers were twofold more effective than finasteride-nanocarriers in penetrating the skin and targeting the drug to the hair follicles. – While dutasteride-loaded nanocarriers provided a 3.4-fold increase in hair follicle penetration, finasteride-loaded nanocarriers increased by about 1.5-fold.	[103]
PLGA nanoparticles coated with hair follicle’s derma papilla cells encapsulated dutasteride and siRNA (silence RNA – Androgenic Receptor silencing agent). ● Uncoated-nanoparticles:  – Size: 153.2 ± 1.2 nm  – Zeta potential: −24.5 ± 0.8 mV  – Dutasteride EE%: 53.3 ± 0.6%  – siRNA EE%: 47.1 ± 0.3% ● Coated-nanoparticles:  – Size: 194.8 ± 3.2 nm  – Zeta potential: − 30.5 ± 1.0 mV  – Dutasteride EE%: 53.3 ± 0.6%  – siRNA EE%: 47.1 ± 0.3%	In vitro skin penetration tests in porcine ear skin for 6 h and 12 h. The therapeutic effect was assessed by in vivo tests with male C57BL/6 mice for 28 days. The positive control group was treated with 3% minoxidil 1× day, the model group received no treatment, and other than two groups that received coated and uncoated nanoparticles separately, uncoated nanoparticles containing one agent at a time, or both were also studied. Each group had a n = 4, and each treatment was applied once every other day.	– Hair follicle targeting factor and total drug accumulation was greater in the skin with coated nanoparticles (p < 0.05). – Dutasteride amount in hair follicles from coated particles was almost double after 6 h and 1.4 times higher after 12 h. – In vivo tests showed minoxidil group and the group treated with coated nanoparticles loaded with both agents had the best hair density, diameter (p < 0.05), and best hair growth compared to the other groups, even though the nanoparticles were applied at a lower frequency than minoxidil. – Coated particles also promoted faster onset of the anagen phase and faster hair development. – Dutasteride and siRNA had a synergic effect.	[102]

However, the challenge posed by dutasteride’s physicochemical characteristics has also prompted the development of more inventive-nanostructured carriers. For example, PLGA nanoparticles coated with dermal papilla cell membranes to increase both penetration and biocompatibility have been described. The nanostructure encapsulated both dutasteride and an androgen receptor-silencing agent (siRNA). The encapsulation efficiency was approximately 50% for both dermal papilla membrane-coated and uncoated particles, resulting in coated particles of approximately 200 nm and uncoated particles of approximately 150 nm in size. In vitro permeation assays with porcine ear skin were performed with both the coated and uncoated particles containing only dutasteride. The study demonstrated that coated particles promoted significantly greater dutasteride accumulation and higher hair follicle-targeting compared to uncoated particles. The therapeutic effect was assessed by in vivo studies with male C57BL/6 mice for 28 days, showing a synergistic effect between dutasteride and siRNA, represented by the higher degree of hair growth in the group treated uncoated particles loaded with both agents compared to the groups that received uncoated nanoparticles with only dutasteride or siRNA. Dermal papilla-coated particles with both dutasteride and siRNA had the best results, comparable to minoxidil solution, but with lower administration frequency [102]. Of note, murine skin is a poor model for human skin systems, leaving much to be desired for future research. On top of that, the increased complexity of the product may discourage further applications.

One unique approach was to load dutasteride in iron oxide nanocarriers. The performance of the nanocarrier was compared to finasteride loaded in the same carrier, both tested in porcine ear skin in vitro. Although drug controls contained Tween®80 and ethylene glycol – two permeation enhancers – the nanocarriers provided greater skin penetration for both drugs. After 24 hours, dutasteride-loaded nanocarriers were at least twice superior to finasteride-loaded ones targeting the hair follicles, the authors hypothesized finasteride’s faster release from the nanocarriers caused the system to disarrange, hence hindering the interaction between nanosystem and skin [103]. Further, the iron’s presence in the formulation gives a dark color to the formulation, which offers an instant optical effect of filling the scalp for those with naturally dark hair. However, the coloring effect can also limit the formulation’s application to only patients with darker hair. The main results of these studies are summarized in Table 1.

4.2. Studies limitations

As can be seen from the studies listed in Table 2, most experimental designs fail to verify the hair follicle targeting effect. For example, dutasteride was successfully loaded in lipid nanocarriers coated with a chitosan oligomer conjugated with lauric acid, which resulted in particles of approximately 190 nm. Skin penetration experiments were performed with porcine ear skin for 48 hours in vitro with chitosan-coated and uncoated lipid nanocarriers. Uncoated nanocarriers promoted a dutasteride accumulation almost 2-fold higher than the coated ones, although uncoated particles had nearly the same size (184 nm), which can be attributed to chitosan’s mucoadhesive interaction with the surrounding skin. However, the amount of penetrated dutasteride was only measured in the stratum corneum and viable skin, with no measure of the hair shaft. Also, the encapsulation efficiency of coated nanocarriers could not be determined due to methodology limitations, which makes for unreliable comparisons between past and future studies [104].

Table 2.

Studies main results and limitations of dutasteride-loaded nanostructured systems for topical treatment of androgenic alopecia.

Type of nanoparticles	Type of study	Main results	Reference	Type of nanoparticles
Nanostructured lipid carriers coated with lauric acid-chitosan oligomer. ● Uncoated-nanocarrier:  o Size: 184.2 ± 2.9 nm  o Zeta potential: −18.0 ± 2.3 mV  o EE%: 97.3 ± 1.2% ● Coated-nanocarriers:  o Size: 188.4 ± 2.2 nm  o Zeta potential: 24.8 ± 2.1 mV  o EE%: undetermined	In vitro skin penetration tests with porcine ear skin.	Uncoated nanocarriers enhanced almost twice the amount of dutasteride accumulated in the total skin (6.1 ± 1.1 µg/cm2 vs. 3.2 ± 0.3 µg/cm2, respectively).	EE% of coated nanocarriers was not determined due to method limitations. Hair follicle targeting was not assessed.	[104]
Nanoemulgel ● Dutasteride-loaded nanoemulgel:  o Size: 252.3 ± 8.6 nm  o PdI: 0.205 ± 0.60  o EE%: 91.3 ± 3.1	Ex vivo skin permeation tests for 24 h in male Swiss albino mice. In vivo study in male Swiss albino mice to assess hair shaft diameter and length for 20 days.	– A dutasteride solution and a dutasteride-loaded nanoemulsion were used as controls for the skin permeation test. – Nanoemulgel promoted the highest drug accumulation into the skin, followed by nanoemulsion and drug solution (37.4%, 21.6%, and 2.8%, respectively). – At the end of in vivo tests, nanoemulgel group animals had longer hair shafts (9.1 ± 0.8 mm) than the untreated group (8.2 ± 0.7 mm) and almost the same hair shaft diameter as the control group (38.9 ± 4.9 µm vs. 40.3 ± 5.7 µm, respectively).	Rodent skin was used in penetration tests. Hair follicle targeting was not assessed. In all cases, dutasteride permeated through the skin: drug solution (90.7%), nanoemulsion (40%), and nanoemulgel (13.7%).	[105]
Surface-modified liquid crystalline solid lipid nanoparticles loaded with finasteride and dutasteride. ● Uncoated liquid crystalline solid lipid nanoparticles:  o Size: 197.9 ± 2.5 nm  o Zeta potential: −20.1 ± 1.9 mV ● 1.25% Chitosan coated liquid crystalline nanoparticles:  o Size: 239.6 ± 3.3 nm  o Zeta potential: 19.8 ± 2.5 mV ● 7.5% Chitosan-coated liquid crystalline nanoparticles:  o Size: 259.8 ± 7.5 nm  o Zeta potential: 48.5 ± 1.2 mV ▪ EE% > 98 for all particles.	In vitro skin permeation tests in dermatomed abdominal porcine skin for 24 h. Solutions of dutasteride and finasteride were used as controls.	– Compared to controls, drugs’ encapsulation significantly increased the permeation (p < 0.01). – Chitosan-coating increased drugs’ permeation proportionally to the chitosan amount. – Coated nanoparticles increased dutasteride permeation 1.2-fold and 1.8-fold for 1.25 and 7.5% chitosan compared to uncoated particles.	The dermatomed skin model is not suitable for assessing hair follicle targeting. Hair follicle targeting was not assessed.	[106]
Nanoemulsion O/W ● A  o Size: 5230 ± 13.1 nm ● B  o Size: 4340 ± 15.2 nm ● C o Size: 32.5 ± 2.12 nm  o EE%: 45.1 ± 0.3 ● D  o Size: 29.5 ± 3.53 nm o EE%: 70.0 ± 0.5 ● E  o Size: 45.4 ± 2.4 nm  o EE%: 99.9 ± 0.1	Ex vivo skin permeation tests for 48 h in rat skin. Nanoemulsion “E” was selected to conduct the ‘permeation test.’	– Although the authors called the experiment “ex vivo study: permeation analysis,” what was really performed was a drug release test using rat skin as a membrane. – The system showed sustained release at 24 h and continued for up to 6 days, during which time 76% of DUT was released.	The skin model used in penetration tests is not ideal. Also, the skin was pre-shaved, which could interfere with the stratum corneum’s integrity.	[107]
Nanostructured lipid carriers coated with stearic acid-chitosan oligomer. ● Uncoated dutasteride-loaded nanocarriers:  o Size: 187.6 ± 7.0 nm  o Zeta potential: −18.3 ± 0.9 mV ● Dutasteride-loaded nanocarriers coated with 5% chitosan oligomer:  o Size: 208.4 ± 1.6 nm  o Zeta potential: 21.7 ± 2.1 mV ● Dutasteride-loaded nanocarriers coated with 5% stearic acid-chitosan oligomer:  o Size: 220.1 ± 11.9 nm  o Zeta potential: 26.0 ± 1.1 mV ● Dutasteride-loaded nanocarriers coated with 10% stearic acid-chitosan oligomer:  o Size: 230.1 ± 8.2 nm  o Zeta potential: 30.0 ± 1.2 mV	In vitro skin penetration tests with porcine ear skin for 48 h.	– Conjugation with stearic acid improved the stability of nanocarriers. – Control solution promoted the highest accumulation amount (12.62 ± 1.72 µg/cm2) compared to nanocarriers (p < 0.05). – Among nanocarriers, uncoated nanocarriers had the best penetration (6.09 ± 1.09 µg/cm2), significantly higher than 5% coating, 10% coating, and 5% coating without conjugation (2.82 ± 0.40 µg/cm2; 2.70 ± 0.35 µg/cm2; and 2.11 ± 0.64 µg/cm2, respectively).	Dutasteride solution used as a control was vehiculated in ethanol, a permeation enhancer. Hair follicle targeting was not assessed.	[108]
Lipid-based nanoparticles with and without oleic acid. ● Formulation A: with the smallest amount of oleic acid and the highest amount of surfactant.  o Size: 227.6 ± 7.4 nm  o EE%: approximately 80%. ● Formulation B: without oleic acid.  o Size: 181.8 ± 12.4 nm  o EE%: 96.7%	In vitro skin penetration tests with porcine ear skin for 48 h.	– Nanoparticles without oleic acid had higher dermis penetration (17.7 ± 0.7% against 11.5 ± 1.7%) compared to the one with oleic acid (p < 0.05). – After 48 h drug release from formulation A was 90% compared to formulation B, which presented 97.5% drug released in the same period.	Hair follicle targeting was not assessed.	[109]

A similar study was conducted by the same group, testing four dutasteride-loaded nanostructured lipid carriers: one uncoated, one coated with only a chitosan oligomer, and two coated with a chitosan oligomer but also conjugated with 5 or 10% stearic acid. The unconjugated, uncoated nanocarriers were approximately 190 nm, while the coated, conjugated nanocarriers were approximately 230 nm (nanocarriers coated with 10% stearic acid-chitosan oligomer). Dutasteride penetration was also assessed for 48 h in porcine skin in vitro. Here, a dutasteride ethanolic solution was used as control, which is not appropriate as ethanol is a permeation enhancer [110–112]. Not surprisingly, the control solution promoted the highest dutasteride accumulation in the skin, almost 13 µg/cm². Amongst nanocarriers, uncoated nanocarriers yet again promoted higher dutasteride accumulation in total skin compared to coated nanocarriers. At the same time, stearic acid conjugation seemed to increase the permeation of coated nanocarriers compared to those coated with chitosan oligomer alone (see Table 2). Unfortunately, dutasteride accumulation in hair follicles was not assessed, similar to their previous study [108].

A more recent study, also lacking follicular targeting measurements, tested lipid-based nanoparticles with and without oleic acid, which presents anti-androgenic properties. Lipid nanoparticles without oleic acid were approximately 180 nm, while the ones with oleic acid were approximately 230 nm in size. After in vitro skin penetration, nanoparticles without oleic acid promoted a 1.5-fold increase in dutasteride accumulation in total viable skin, compared to nanoparticles with oleic acid [109], which can be explained by dutasteride’s stronger affinity to oleic acid showed by the slower drug release for nanoparticles with oleic acid in the tested period. However, as noted, follicular targeting between the two formulations was not assessed.

A noticeable trend is the size range of the nanocarriers used in the studies reviewed here, varying from 30 nm to 260 nm. From these results, one can hypothesize that nanocarriers up to 230 nm would be more suitable for reaching the sebaceous glands, a target for skin conditions such as acne [87] and hidradenitis suppurativa [85]. In comparison, bigger nanocarriers in the range of 600 to 700 nm would be more suitable for reaching deeper portions of the hair follicles, such as the bulb region, our target for treating androgenic alopecia [28,80,84]. However, as previously explained, when evaluating hair follicle targeting, not only must particle accumulation be addressed but also the drug’s release and diffusion through the hair follicle structure and surrounding skin.

Another recurring limitation of the study is the use of rodent skin as a model for in vitro skin penetration tests. Rat and mouse skin is notably more permeable than human skin. This type of skin model, when aiming to determine the hair follicle targeting effect of a drug delivery system, may skew the results severely [113]. Two different studies used rodent skin to measure dutasteride permeation in vitro; the first study measured the permeation of a nanoemulgel in ex vivo tests in Swiss albino mice for 24 hours, while the second study measured the permeation of a nanoemulsion in rat skin for 48 hours. In both cases, dutasteride was detected in the receptor compartment [105,106], which indicates transdermal permeation and, hence, the possibility of a systemic effect, an undesirable feature in this case. None of the studies quantified the amount of dutasteride in the hair follicles nor any skin layer. Hence, the tests served more as a drug release experiment, with rat skin as a membrane.

Porcine skin model is considered the best surrogate for in vitro skin penetration tests [114], more suitable than human skin itself, not only for the limited access but once excised, human skin loses elasticity and contracts, causing a perpetual occlusion of hair follicles pores limiting up to 90% hair follicle penetration [115]. Notwithstanding, the skin source is not the only point to consider. A different example of an inadequate skin model to assess hair follicle targeting is the use of dermatomed skin. Defined by its namesake process to remove part of the skin dermis, leaving the uppermost dermal layer and the whole epidermis and consequently, the main barrier to drug penetration, the stratum corneum. Such a process is made to create uniform test samples for in vitro skin permeation studies but is completely inadequate in studies focusing on hair follicle accumulation, as the process may damage the hair follicle structures. Regardless, dermatomed abdominal porcine skin was used to verify the permeation of chitosan-coated and uncoated liquid crystalline solid lipid nanoparticles encapsulating both finasteride and dutasteride in in vitro experiments, which might explain contradictory results found. One of such studies, in contrast to the other studies mentioned prior, described chitosan coating increased the permeation of the drugs through the skin compared to uncoated nanoparticles. Although particle sizes were almost the same – uncoated nanoparticles were approximately 200 nm, while coated particles ranged between 239 and 260 nm [106] - the amount of finasteride permeated through the skin was 1.4 and 2.3 times higher than the amount of dutasteride for nanoparticles coated with 1.25% and 7.5% chitosan, respectively. The difference in drug penetration can be attributed to the higher lipophilicity of dutasteride, which can interact more with the skin’s natural lipids, consequently preventing further deeper penetration. However, this study did not quantify the presence of finasteride and dutasteride in the various layers of skin, instead, this study used total skin, so such an assumption cannot be confirmed.

5. Future perspectives

We remain far from an ideal drug delivery system to achieve a safe and effective treatment for androgenic alopecia. Unfortunately, most of the experiments performed to date do not assess hair follicle targeting, which is an essential consideration when developing a nanosystem to maximize hair growth while limiting systemic effects. Other important variables associated with hair follicle targeting are also ignored, such as the interaction of nanocarriers with the hair follicle’s sebum. There is growing evidence the hair follicle sebaceous content can completely modify nanoparticle targeting. We have demonstrated in a previous study that the sebum content hampered a clindamycin phosphate-loaded nanostructure lipid carrier deposition into hair follicles using an in vitro sebaceous skin model and comparing it with a conventional porcine skin model [116]. Thus, the physiological condition certainly must be considered when designing dutasteride drug delivery vehicles targeting hair follicles.

Another point is nanocarrier size may be a key factor for optimizing follicular targeting. The narrow range of dutasteride-loaded nanosystems developed to this point, from a few nanometers to under 300 nm, may not be ideal for reaching deeper portions of hair follicles. Looking forward, it is of interest to evaluate the follicular delivery of bigger particles, especially those around 600 nm, currently considered the ideal size to reach the hair follicle’s bulb region. Another point of improvement is the use of appropriate skin models, such as pig skin. Even though the hair density of rats (around 290 hair follicles per cm² [117]) may be closer to a human scalp density (between 150 to 292 hair follicles per cm² [118,119]), both rodent and dermatomed models are much more permeable than porcine and human skin. Thus, the use of such models overestimates dutasteride’s permeation, leading to erroneous conclusions about the potential of the drug delivery system.

As highlighted here, there is still quite a bit of work required to develop a suitable dutasteride-loaded nanosystem. With that in mind, future studies performed using better methodologies and obtaining more insightful read-outs may lead us to the development of an effective and safe treatment for androgenic alopecia.

Funding Statement

This paper was funded by HairDAO Association, a patient-led community based on Ethereum, dedicated to solving androgenetic alopecia.

Article highlights

• Dutasteride is a promising option for androgenic alopecia treatment.

• Oral dutasteride causes serious sexual side effects.

• Different nanocarriers are being explored for the follicular target delivery of dutasteride.

• Many studies fail to perform adequate controls during experiments.

• Nanocarriers with bigger sizes, around 600 nm, should be explored in the future.

• Efforts to obtain a safe and effective local therapy with dutasteride are in the initial stages.

Author contributions

J. F. M. Andrade: literature survey, writing original draft; A. Verbinnen: literature survey, review, and editing; A. Bakst: literature survey, review, and editing; M. Cunha-Filho: review and editing; G. M. Gelfuso: review, editing, and supervision; T. Gratieri: conceptualization, review, editing, and supervision.

Disclosure statement

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants, or royalties.