Topical treatments are especially attractive for uncomplicated CL cases, offering significant advantages over systemic therapy, including easier administration, fewer adverse effects and cost-effectiveness. Despite being a very attractive route of administration, the topical route represents a challenge for many drugs. The outermost layer, the epidermis, is composed of two main layers: stratum corneum (SC) and viable epidermis. The SC, composed of corneocytes embedded in a lipid matrix, represents the main physical barrier of the skin, protecting the inner layers from the external environment. The viable epidermis is composed mainly of keratinocytes, melanocytes, Merkel cells and Langerhans cells. Adjacent to the epidermis is the dermis, which performs important functions of nutrition and support. The innermost layer of the skin, the hypodermis, is a fat layer providing mechanical support and thermal insulation [82,83,84].

After topical application, drugs can permeate through the skin by three different pathways: (i) appendageal or transfollicular, allowing the direct transport of substances via hair follicles and glandular ducts; (ii) intercellular or paracellular pathway, in which the drug diffuses between the cells, passing through the lipid matrix; and (iii) the intracellular or transcellular pathway, in which the drug passes inside the skin cells and through the lipid matrix [84,85,86]. It is assumed that a combination of these three pathways can contribute to the skin penetration of all substances, but the preferred route depends on their physicochemical characteristics [87].

In CL treatment, the goal of a topical formulation would be to target the infected macrophages located in the dermis [88]. An important aspect of CL is that the patient’s skin is not always intact, as, with the evolution of the disease, the SC is usually damaged. In CL, a papule initially forms at the inoculation site, which usually evolves into ulcerated lesions. In this process, there is a loss of the epidermis and part of the dermis as a result of the local inflammatory response [89]. Although this loss of SC initially facilitates the entry of drugs through the skin, re-epithelialization and wound healing during treatment, along with the production of collagen and metalloproteases of the extracellular matrix, may represent an additional challenge for topical treatment [90]. Thus, it is desired that the formulation works in all possible situations: intact, partially or completely damaged skin [91].

To ensure efficient penetration of a substance through the intact skin, it has been proposed that it must have some characteristics, such as a melting point of less than 200 °C, low molecular weight (less than 500 Da) and log p value between 1–3 [86]. The ability of the molecule to form hydrogen bonds and its degrees of ionization also need to be considered [83]. Other factors related to the individual, as well as the environment, can also impact skin permeation, such as age, hormonal balance, sebum production, skin hydration and pH gradient [84].

AmB has unfavorable physicochemical characteristics for topical administration, such as high molecular weight (~924 Da), amphoteric nature, low aqueous solubility at physiological pH and tendency to self-aggregation [88,90,93,94]. Its poor permeability across biological barriers severely limits its effectiveness, as reported by López and colleagues [94] in a clinical trial (NCT01845727) using a cream containing 3% AmB (Anfoleish^®). Although safe, this formulation showed low efficacy in patients with CL, which was attributed to the low transdermal permeation confirmed by the absence of AmB in patients’ plasma [94].

Thus, alternative formulations capable of promoting the topical delivery of AmB are needed. To improve the dermal penetration of drugs and topical therapy, different strategies can be used, such as passive methods (chemical permeation enhancers, for example), physical methods and nanocarriers. The nanocarriers offer a gentler alternative to facilitate drug permeation, being the least damaging one and capable of increasing the drug residence time in the skin [87]. In this sense, recent studies have investigated the use of different types of nanosystems to improve AmB skin permeation, including liposomes, lipid nanoparticles and polymeric carriers [95,96,97,98,99,100,101].

Liposomes were the first lipid nanocarriers investigated and marketed to enhance drug penetration into the skin for dermatological and cosmetic applications. The skin delivery of active substances by liposomes is highly affected by their lipid constituents, particle size, surface charge and lamellarity [93]. Over the past three decades, significant progress has been achieved in the design of more deformable vesicles, in particular niosomes, transfersomes and ethosomes, allowing delivery of drugs deeper into/through the skin [82,87].

Jaafari et al. [101] have investigated liposomes loaded with AmB at different concentrations: 0.1, 0.2 and 0.4% (Lip-AmB). An in vitro permeation study using intact mice skin showed that increasing the AmB concentration in the formulation resulted in a greater amount of AmB permeating the skin. In vivo studies on BALB/c mice infected by L. major showed that the efficacy of Lip-AmB 0.4% was greater compared to the other groups (Lip-AmB 0.1 and 0.2%, empty liposomes or PBS). According to the authors, the presence of skin permeation enhancers in liposomes could contribute to these positive results: significant reduction in lesion size and almost complete elimination of parasites in the skin and spleen [101,102]. The results led to development of topical Lip-AmB (0.4%) (Sina Ampholeish^®) [103].

Other interesting studies in AmB topical delivery have explored the potential of ultra-deformable liposomes (AmB-UDL), using Tween 80^®, sodium cholate or sodium deoxycholate as an edge activator. Perez et al. [104] noticed that the insertion of AmB reduced vesicle deformability. This finding is in line with other reports in the literature and can be explained by the interaction of AmB with the lipids and edge activators, reducing their mobility [88,104,105,106]. As shown by the authors, the increase in AmB content, in addition to reducing the deformability, modified the absorption spectrum, suggesting AmB self-association in liposome bilayers. An interesting observation was that increasing the surfactant concentration could circumvent this event, keeping AmB in the monomeric form [104]. This probably explains the improved in vitro skin penetration of AmB from this formulation, in comparison to the AmBisome^®. Carvalheiro et al. [105] conducted studies evaluating liposomes of similar composition, confirming that ultra-deformable liposomes promoted increased drug penetration into the skin. In addition, Peralta et al. [88] also showed that this type of liposome provided better drug penetration into/through human skin than conventional ones. The studies presented above showed improvement in AmB’s topical delivery. However, in vivo proof of concept was not performed. For a more complete view, studies on experimental models of CL are described below.

Fernández-García et al. [97] developed another AmB formulation in ultradeformable vesicles (AmB-TF) and evaluated in vitro the drug permeation across intact mice skin. Although there was no significant difference in permeation between the AmB-TF and AmB-DMSO solutions, the permeation flux from AmB-TF was about five times higher than that described previously for other liposomal formulations, including transferosomes loaded with AmB. The in vivo skin pharmacokinetic of AmB-TF was also assessed after topical administration in mice and showed permeation and accumulation of AmB in the dermis at therapeutic concentrations relevant for the treatment of leishmaniasis. In line with these findings, the topical application of AmB-TF in mice experimentally infected by L. amazonensis over 10 days resulted in almost complete elimination of the parasite burden in the lesion, which was similar to that observed after intralesionally administered Glucantime^®. Regarding the effect on the lesion size, the efficacy of intralesional Glucantime^® was greater than that of AmB-TF. However, the overall data suggested that increasing treatment time or twice-daily application of topical AmB-TF could lead to complete lesion healing.

In addition to using ultradeformable vesicles, some authors used another strategy: the drug combination. Mostafavi et al. [95] evaluated niosomes co-encapsulating AmB and Glucantime^® (AmB-Glucantime^® niosomes), composed by Span 40 and Tween 40, whereas Dar et al. [96] investigated ultradeformable liposomes co-loaded with AmB and miltefosine (AmB-MTF liposomes), composed by PC and Tween 80. Despite the large particle size of the niosomes co-encapsulating AmB and Glucantime^® (9.5 μm), the topical treatment of BALB/c mice infected with L. major (twice daily for 30 days) promoted reduction in the lesion in comparison to placebo and intramuscular Glucantime^® [96]. On the contrary, the topical treatment with AmB-MTF liposomes developed by Dar et al. [96] resulted in complete lesion resolution in mice infected with L. mexicana after twice daily treatment for 4 weeks. In agreement, the lesion parasite burden had a significant reduction for AmB-MTF liposomes when compared to the other groups—untreated, treated with plain AmB-gel or treated with ultradeformable liposome gel containing only AmB [96]. These results confirmed the benefit of drug combination due to a possible synergistic effect in CL treatment.

Although there are few clinical trials investigating AmB topical formulations, Khamesipour et al. [107] investigated the activity of liposomal AmB (0.4%) (SinaAmpholeish^®) developed by Jaafari et al. [101], which had already shown safety in a Phase I clinical trial [108]. The pilot study compared three treatment groups in patients with CL caused by L. major: (i) topical liposomal AmB (0.4%) alone twice daily for 28 days; (ii) topical liposomal AmB (0.4%) in combination with daily intramuscular Glucantime^®; (iii) weekly intralesional Glucantime^® plus biweekly cryotherapy (the standard treatment in the Islamic Republic of Iran). Complete cure was 92%, 95% and 48.5% of patients who received combination treatment (liposomal AmB 0.4% plus Glucantime^®), topical liposomal AmB only and standard treatment alone (Glucantime^® plus cryotherapy), respectively.

In turn, Horev et al. [109] performed a randomized, double-blind, placebo-controlled trial to investigate the efficacy of liposomal AmB 0.4% gel in L. major-infected patients treated twice daily for 56 days. Different parameters were evaluated, such as lesion diameter, ulceration and healing. At the end of treatment, the results were similar between the liposomal AmB gel-treated and control groups. The authors suggested that a longer treatment duration may be necessary to improve efficacy because clinical improvement, including negative PCR test, was more clearly observed after 56 days rather than earlier.

In the literature there are many reports about the ideal characteristics of the liposome carrier for topical application. However, the effects of the AmB aggregation state on the skin drug penetration and the formulation efficacy are still poorly explored. AmB insertion in lipid vesicles is a complex process because it can adopt different aggregation forms, depending on the AmB concentration, vesicle composition and preparation method [72]. Additionally, the development of new skin models, providing more realistic conditions, is an important point to increase the chance of bringing topical formulations from the bench to the market [84]. In this sense, it is worth noting that all studies presented here performed skin permeation tests on intact skin. This is an important limitation because, under pathological conditions like CL, considerable skin damage usually occurs, altering skin architecture and permeability [91].

The oral route is usually preferred for drug administration. Oral treatments often result in lower drug toxicity in comparison to the parenteral ones and improved patient compliance. This is especially important for neglected tropical diseases, such as leishmaniasis, which affect mainly poor people, who live in remote areas and have limited access to health centers.

However, AmB is a class IV drug, according to the BCS classification system, exhibiting low solubility in neutral pH and low membrane permeability, with expected low oral bioavailability.

Indeed, several physicochemical factors contribute to the low oral bioavailability of AmB from the existing commercial formulations, including AmBisome^®. These factors comprise the high molecular weight of the AmB molecule, its low solubility in both aqueous and lipidic environments and tendency to self-associate, and its instability at the low pH found in the stomach [110].

This context has stimulated the search for strategies to improve the oral delivery of AmB, with few successful cases [40,110,111,112]. The following drug carriers have shown improvement of AmB bioavailability or efficacy by the oral route: polymeric nanoparticles [113,114], polymer lipid hybrid nanoparticles [115], solid lipid nanoparticles [116], chitosan-coated nanostructured lipid carriers [117], cubosomes [118], emulsions [119,120], cochleates [121,122,123,124] and liposomes [77].

In a recent review, Wasan et al. [111] reported currently investigated AmB formulations for the treatment of parasitic infections, with an emphasis on two oral lipid formulations that have reached clinical trials. First, a self-emulsifying lipid-based formulation (iCo-019), consisting of a mixture of mono- and di-glycerides, in addition to D-alpha-tocopheryl poly(ethylene glycol) succinate, which completed two human Phase I trials. Second, an encochleated AmB deoxycholate formulation under Phase II trials to determine its efficacy for cryptococcal meningitis. Interestingly, a low aggregation state of AmB was claimed for both types of formulations [111,122]. Moreover, the safety, tolerability and pharmacokinetics data of iCo-19 following single doses to healthy humans supported long-lasting systemic drug exposure, with no evidence of gastrointestinal, hepatic or renal toxicities associated with AmB [125]. A similar safety profile has been reported in humans for the encochleated AmB deoxycholate formulation [124]. This first set of clinical data highlights the great potential of these lipid AmB formulations for the oral treatment of leishmaniasis.

Ramos et al. [77] reported for the first time an orally active liposomal AmB formulation. The nanoformulation contained the same lipids as AmBisome^®, but also included 4.7 mol% DSPE-PEG2000. Characterization of the drug aggregation state by CD suggested lower aggregation of AmB in the PEGylated formulation, when compared to AmBisome^®. This feature is likely critical, as the liposomal AmB formulation seems to share the low drug aggregation state with oral AmB formulations under clinical trials. The new liposomal AmB formulation exhibited much faster drug release than AmBisome^®, in agreement with the lower extent of drug aggregation. The PEGylated formulation also showed greater stability in simulated gastric fluid, when compared to the non-PEGylated formulation, regarding particle size distribution and AmB aggregation state. Importantly, the PEGylated liposomal AmB formulation exhibited therapeutic efficacy by the oral route in the murine model of CL, promoting significant inhibition in the lesion size growth and reduction in the parasite load in the lesion, when compared to the saline-treated control. This effect was achieved at a relatively low dose of AmB (5 mg/kg) given on alternate days. The reduced renal toxicity of oral treatment with PEGylated liposomal formulation was also supported by the absence of change in the plasma level of urea, in contrast to AmBisome^® given parenterally at the same dosage. Considering the low aggregation state of AmB in the oral liposomal formulation and the significant drug release, one can expect an effective intestinal absorption of AmB under the free form. In this context, a sustained drug release from the liposomal formulation in the intestine may result in a long-lasting drug plasma level and may explain the reduced toxicity, as proposed previously for iCo-019 [125].

The benefit of liposome PEGylation for oral drug delivery is consistent with previous reports in the literature for other drugs, including peptides, proteins and lipophilic substances [58]. The oral efficacy of PEGylated liposomal AmB formulation may be attributed to several factors, including: (i) the prevention of liposome aggregation in an acidic environment, (ii) the protection of liposomes from the action of bile, (iii) the protection of AmB from acidic degradation, and (iv) the low state of AmB aggregation, leading to more effective intestinal drug absorption. Further studies are needed to identify the factors that most contribute to improved oral drug efficacy and further optimize liposomal formulations for the delivery of AmB by the oral route.