Gene therapies have not yet been approved to treat skin diseases but the progress in the field has been remarkable over the last couple of decades. For context, China was the first country to approve and commercialize a gene therapy product back in 2003 ^[1][32]. It was not until 2012 that Glybera became the first gene therapy approved in Europe by the European Medicines Agency (EMA) for the treatment of lipoprotein lipase deficiency, an ultra-rare inherited disorder ^[2][33]. Five years later, the United States Food and Drug Administration (FDA) gave approval to Luxturna, ushering in a new range of possibilities for disease treatment ^[3][34]. Today, there are close to a dozen gene therapy products approved by the different regulatory agencies, and this number is on the rise ^[4][5][4,35].

Progress on gene editing and gene delivery in dermatology is evidenced by a number of pre-clinical and clinical reports ^[6][36]. The biggest strides made in clinical studies by gene therapy in dermatology have been toward epidermolysis bullosa (EB), followed by melanoma and other rare diseases like Netherton syndrome and congenital ichthyosis ^[7][8][9][37,38,39]. EB is a family of rare genetically heterogenous disorders that cause fragility and blistering of the skin and mucous membranes. Its severity ranges from mild to fatal. Junctional EB (JEB) is the most severe subtype caused by the absence of anchoring proteins due to mutations in the laminin 322 genes (LAMA3, LAMB3, and LAMC2) ^[10][40]. The skin blistering characteristic of dystrophic EB (DEB) is ascribable to mutations in the gene that encodes for type VII collagen, COL7A1 ^[11][41]. Within the DEB spectrum, recessive DEB (RDEB) is the most severe form with blisters spread over the whole body also affecting mucous membranes in the gastrointestinal tract. Milder symptoms of this condition are noted for dominant DEB (DDEB), primarily localized on the hands, elbows, feet, and knees. Efforts have been made to deliver the 8833-nucleotide open reading frame of the COL7A1 gene, to DEB patients to restore the production of type VII collagen ^[12][42]. Nonetheless, the large size of the gene hinders its packaging into viral vectors and limits the transduction efficiency in addition to reducing the viral titer. Most of the gene therapy research has been ex vivo, focused on cell culture and transplantation of cultured skin grafts ^[13][14][43,44]. Many groups initially explored the delivery of COL7A1 ex vivo using a retrovirus to correct primary patient keratinocytes that were later xenografted onto immunodeficient mice ^[15][16][45,46]. Since then, the number of gene therapy approaches that have made it to the clinic has increased. Recorded clinical trials using gene therapy for EB and other skin conditions are compiled in Table 1.

Genetically engineered autologous skin equivalent grafts dominate the EB treatment landscape in clinical trials, as shown in Table 1. A notable example of gene editing in dermatology is the restoration of the skin by autologous, transgenic epidermal grafts on approximately 80% of the total body surface area of a 7-year-old patient suffering from JEB ^[19][49]. In this report, keratinocytes derived from the patient were transduced with a functional LAMB3 gene ex vivo using a retroviral vector to generate permanently edited keratinocytes. Epidermal grafts were created from the transgenic cells and used to transplant onto the patient, thereby restoring the skin of the patient, which resulted in a sustained and robust epidermis throughout the 21-month follow-up period ^[19][49]. The potential for long term benefits from grafted transgenic epidermis was demonstrated in another study using a similar approach on a smaller area of the body of a JEB patient. Here, the positive effect of treatment was still present 6.5 years later ^[20][50]. Castle Creek Biosciences’ D-Fi (debcoemagene autoficel), named FCX-007 prior to Fibrocell Science acquisition, is comprised of autologously-derived fibroblasts from RDEB patients genetically corrected using a lentiviral vector encapsulating the COL7A1 gene. These genetically modified fibroblasts are then intradermally injected to the patient at the wound site. Wound healing has been observed to last for up to a year using the D-Fi treatment ^[21][51]. D-Fi has received orphan drug designation by the FDA for DEB. Unlike D-Fi, EB-101 from Abeona Therapeutics consists of autologous RDEB keratinocytes—instead of fibroblasts—isolated from skin biopsies but transduced with a recombinant retrovirus—instead of a lentiviral vector—containing COL7A1 ^[22][23][52,53]. Abeona Therapeutics just released promising results from their phase I/IIa clinical study following treatment with EB-101 for RDEB. Wound healing of at least 50% was observed in approximately 70% of the wounds after three years and 80% at year six. Absence of pain was recorded for all treated wounds. EB-101 phase III results on large chronic wounds are due in mid-2022 ^[22][52]. Other studies, however, have suggested that, for successful anchoring fibril formation, type VII collagen needs to be expressed in both fibroblasts and keratinocytes ^[24][54].

Netherton syndrome, a rare skin disease caused by loss of function mutations in the SPINK5 gene, has also been the target of autologous skin grafts pre-clinically and on a phase I clinical trial ^[25][26][27][55,56,57]. The SPINK5 gene encodes for the lymphoepithelial Kazal-type-related inhibitor (LEKTI) responsible for the regulation of skin desquamation. In a clinical trial, a de-epidermized skin lesion was treated with lentiviral gene therapy to restore the function of LEKTI. Unfortunately, despite having used an integrative vector, LEKTI expression was transient and did not sustain past three months ^[27][57]. Krystal Biotech is investigating the potential of topically administering the SPINK5 gene via non-integrating herpes simplex virus type 1 (HSV1) vectors to treat Netherton’s syndrome ^[28][58].

Besides questions to the feasibility and practicality of grafting ex vivo cultured grafts onto substantial areas of the body of patients, the permanent delivery of a gene into the genome presents concerns over insertional mutagenesis stemming from integration of the gene into an undesirable location in the genome. The latter concerns may be addressed by editing the mutated gene using clustered regularly interspaced short palindromic repeats (CRISPR)-based strategies, these, however, fail to address the large practical challenge and cost of grafting. Furthermore, CRISPR-mediated editing would in many cases need to be tailored to individual patient mutations which, particularly in the context of rare diseases, makes it challenging to translate into viable clinical therapies.

As an alternative approach to permanent gene delivery, the potential to transiently deliver functional genes to the skin in vivo is also currently being pursued. Examples of this include Krystal Biotech’s phase I/II study where functional COL7A1 was delivered to the wound bed of RDEB patients using an attenuated, non-replicating HSV1 viral vector (B-Vec) ^[29][59]. Data from these initial studies showed wound closure in 90% of treated wounds. Subsequently, recruitment for an ongoing phase III study (NCT04491604) was completed in March 2021 ^[29][59]. A major concern with viral delivery is the possibility of immunological responses, particularly in the light of transient therapy that would need repeated dosing throughout the patient lifespan. An attractive, yet less explored, path thus remains transient, non-integrating gene delivery using non-viral delivery vectors. Preclinical work from Amryt Pharma shows that delivery of the COL7A1 gene formulated in highly branched poly β-amino ester–AP103 can achieve five times higher expression of type VII collagen in RDEB keratinocytes compared to healthy keratinocytes ^[30][60]. When applied topically to human RDEB skin grafted onto mice, type VII collagen was observed at the interface between the dermis and the epidermis up to 10 weeks after treatment. A phase I/II for AP103 is scheduled for 2022 ^[31][61].

In multiple ex vivo studies, CRISPR/Cas9 has been harnessed to correct the COL7A1 mutation in induced pluripotent stem cells derived from patients with RDEB with subsequent transplantation onto mice of the skin equivalents grown from corrected keratinocytes ^{[8][9][32][33][34]}[38,39,62,63,64]. CRISPR-based gene editing is also being explored for non-rare dermatological conditions with a recent study demonstrating in vivo delivery of ribonucleoprotein using microneedles in mouse models of atopic dermatitis and psoriasis ^[35][65]. For a more comprehensive review of gene editing and gene therapy in the context of dermatology, we refer the reader to these excellent recent reviews ^[36][37][66,67]. Furthermore, mesenchymal stromal/stem cells derived from skin or bone marrow are being evaluated in clinical trials in adults and children suffering from RDEB ^[38][39][40][68,69,70]. Intravenously administered recombinant type VII collagen protein replacement is also under investigation in a phase II clinical trial (NCT04599881) on RDEB patients ^[41][71]. The advances on the dermatological field using stem cell therapies are notable, enabling tissue regeneration or tissue damage correction at the genetic level ^[19][42][43][49,72,73]. While protein replacement and cell therapy approaches are beyond the scope of this review, multiple research articles describe the treatment options for skin diseases including chronic auto-inflammatory diseases, EB, and would healing ^{[43][44][45][46]}[73,74,75,76].

The progress on oligonucleotide chemistry in the 1960s set a landmark on the evolution of oligonucleotides, which were already tested in clinical trials in the 1990s. Fomivirsen, delivered via injection to the eye, was the first antisense oligonucleotide (ASO) approved by the FDA in 1998 for the treatment of cytomegalovirus-induced retinitis ^[47][77]. Twenty years later, in 2018, the same agency gave global approval to the first small interfering RNA (siRNA) therapy, Patisiran, for the treatment of TTR, as described earlier. It was encapsulated into lipid nanoparticles for hepatocyte delivery ^[48][49][78,79]. Today, over a dozen of oligonucleotide therapeutics have already been approved by the FDA to treat several indications caused by single gene mutations ^[50][80]. This has prompted a surge of research focused on oligonucleotides for the treatment of rare diseases. There are over 85 and 115 registered clinical trials in the U.S. with a focus on siRNA and ASOs, respectively. A compilation of the ongoing and completed clinical trials in the dermatological field, excluding skin melanoma and skin wounds, is exhibited in Table 2.

Pachyonychia congenita, an ultra-rare autosomal dominant disorder resulting from a mutation in one of the keratin genes (KRT6A, KRT6B, KRT6C, KRT16, or KRT17), was the first inherited skin disorder to be targeted using oligonucleotide therapeutics in humans. Specifically, the patient was treated with an siRNA targeting the N171K mutation characterized by the cytosine-to-adenine single nucleotide K6a mutation ^[51][81]. This selective mutation depletion seemed to be the trigger of the callus reduction observed in the trial, hence, the use of siRNA has the potential of correcting the molecular etiology of the disease.

Hypertrophic scarring is another dermatological condition highly targeted with oligonucleotides. It consists of pathological thickened and elevated scars resulting in a collagen imbalance at the wound site. No pharmaceuticals have yet been approved by the FDA or EMA to treat such a disease, but several RNAi regimens are being investigated at the bench and in the clinic ^[52][53][82,83]. Amongst these, several siRNA treatments stand out in clinical trials, as shown in Table 2. The first one to note is the asymmetric siRNA from Olix Pharmaceuticals with proven efficient gene regulation. OLX10010, Olix Pharmaceuticals siRNA candidate, is being tested in a phase II trial for the treatment of hypertrophic scarring by intradermal injection and Hugel has acquired its exclusive sales rights to treat this condition in Asian countries ^[54][84]. STP705, Sirnaomics’ lead candidate, is another siRNA that diminishes both inflammation and fibrotic activity. It is being investigated for both hypertrophic scar reduction and skin squamous cell carcinoma by intradermal or intralesional injection. This siRNA is delivered encapsulated into Sirnaomics’ proprietary polypeptide nanoparticles consisting of a branched histidine lysine polypeptide, which confer protection to the siRNA and enable delivery to the targeted body cells ^[55][85]. The potential of the self-delivering RNAi platform developed by RXi Pharmaceuticals, now Phio Pharma, is being evaluated for its ability to reduce dermal scarring, also known as fibrosis, by silencing the connective tissue growth factor. This hybrid oligonucleotide leverages the advantages of both antisense technologies and RNAi enabling target specificity, efficient cellular uptake, high potency, and serum stability while minimizing the immune response activation ^[56][57][86,87]. Pfizer’s ASO PF-06473871, however, was abandoned during clinical development because despite showing a successful inhibition of the scarring process, reducing scar severity, it did not significantly outperform the surgical approach ^[53][83].

Several preclinical studies have validated the essential role that GATA3 (GATA binding protein 3), a transcription factor, plays in inflammatory disorders. Specifically, decreased expression of GATA3 enhances skin inflammation in many chronic inflammatory diseases including psoriasis and atopic dermatitis ^[58][59][88,89]. GATA3 is responsible for the production of key inflammatory cytokines including interleukin-13 (IL-13), which mediate inflammation. The overproduction of IL-13 is one of the mainsprings of the pathogenesis of atopic dermatitis. Consequently, inhibiting the function of either IL-13 or GATA3 can be a strategy to treat inflammatory diseases. This aim was pursued by Sterna biologics, a pharmaceutical company that made GATA3 a druggable target using the active pharmaceutical ingredient hgd40. Hgd40 is a catalytic ASO, namely a DNAzyme, which specifically cleaves GATA3 mRNA averting immune system activation. Sterna’s DNAzyme asset completed phase IIa clinical trials with a topical formulation as a proof of concept for atopic dermatitis and is being pursued for other indications outside dermatology ^[60][90].

ASOs have been widely explored to restore gene functionality and, in turn, evoke disease correction ^[61][62][63][91,92,93]. Exon skipping induced by ASOs provides a way to modulate the pre-mRNA splicing process to eliminate a mutated exon from the mature mRNA. Exons 73, 80, and 105 in the COL7A1 gene are known to harbor recurrent mutations in RDEB. The dispensability of these pathogenic mutation-containing exons can restore the function of type VII collagen, shifting the severity of the disease phenotype. Pre-clinical studies in primary RDEB fibroblasts and keratinocytes with mutations in exons 73 and/or 80, transfected with 2’-O-methyl antisense oligoribonucleotides, have demonstrated the potential of exon skipping to restore type VII collagen expression and the formation of anchoring fibrils ^[63][93]. In the clinic, phase I/II results are pending for the QR-313 asset by Wings Therapeutics, an exon skipping ASO comprised of 21 oligonucleotides. It has been topically administered to subjects with DDEB or RDEB with one or multiple mutations in the COL7A1 gene ^[64][94]. QR-313 hybridizes to a specific sequence in the COL7A1 pre-mRNA to exclude exon 73 from the mRNA. In a murine model of atopic dermatitis, where IL-13 is overproduced, an IL-13 ASO administered topically using liposomes significantly suppressed the IL-13 production up to 70% compared to the control group. Furthermore, it also reduced the infiltration of inflammatory cells, reducing the skin thickness ^[65][95]. Exicure’s (formerly AuraSense Therapeutics) ASO-based assets AST-005 and XCUR17 are products of a nanoparticle-based nucleic acid delivery platform called spherical nucleic acids (SNA™). These are formulated as a topical gel and have been evaluated in phase I clinical trials for psoriasis ^[66][67][96,97]. AST-005, tested in collaboration with Purdue Pharma, knocks down a tumor necrosis factor (TNF), a pro-inflammatory cytokine demonstrated to be a key psoriasis mediator. On the other hand, XCUR17 targets the IL-17 receptor alpha, an essential protein in the initiation and maintenance of psoriasis. Despite positive safety and tolerability results of SNA and reductions in the major psoriatic inflammatory markers, efficacy did not meet the expected statistically significance. Nonetheless, Allergan has partnered with Exicure to explore the potential of XCUR17 for alopecia, and Purdue retains the rights to further develop anti-TNF drug candidates. XCUR17 could also be used for the treatment of Netherton syndrome, an application that is still available for out-licensing ^[66][68][96,98]. Not to forget, ASOs have also been investigated in the clinic to treat multiple melanomas including skin melanoma, but as previously stated, cancers fall outside the scope of this manuscript ^[69][99].

The main challenge associated with gene therapies is delivery. The obstacles associated with the delivery of nucleic acids into targeted tissues are attributed to their low serum stability, size, charge, and immune system stimulation ^[70][100]. Moreover, oligonucleotides must be able to escape from the endosomes and enter the cytosol after being taken up by the cell. Delivery of plasmids, furthermore, needs to overcome the challenge of accessing the cell nucleus. To circumvent these challenges and guarantee a therapeutic effect, it is imperative to select an appropriate carrier. Viral and non-viral vectors have been investigated for both oligo and gene transfer as described in several recent reviews ^[71][72][73][101,102,103]. Herpes simplex virus, lentivirus, adenovirus, and adeno-associated virus (AAVs) are some of the most common viral vector candidates in addition to retroviruses, which so far have been the most frequently employed viral vectors for cutaneous gene transfer in the clinic (Table 1). Viral vectors, specifically AAVs, have extensively been studied since their discovery in 1965 for their ability to deliver a cargo to the nucleus of the cell. In 1995, AAVs were utilized in the first human application, and currently, several treatments leverage their delivery potential in the market ^[73][103]. AAVs have also been harnessed in pre-clinical studies for dermatological applications. An example of this is a study aiming to correct a mutation in the COL7A1 gene in keratinocytes from a RDEB patient. In this study, AAVs and adenoviral vectors expressing donor template DNA and transcription activator-like effector nucleases (TALENs), respectively, were combined to correct gene function by restoring the reading frame through introduction of small insertions and deletions ^[74][104]. The limited packaging capacity of approximately 5 kb, the immune response evoked by viral backbones, and the high manufacturing costs, however, remain an obstacle for further expansion of gene therapy applications of viral vectors ^[75][76][105,106]. These are some of the reasons why non-viral delivery systems are gaining popularity for encapsulation and delivery. Lipid nanoparticles, polyplexes, nanospheres, dendrimers, and exosomes are some examples of non-viral vectors ^[77][78][79][2,107,108]. Nanoparticles are the most prevalent approach to circumvent the delivery problem. They are also being widely investigated for oral delivery with the goal to protect gene vectors from gastric acid degradation and to facilitate its transport across the intestinal epithelium ^[80][109]. Despite the fact that oral administration of gene therapies has not yet reached the clinic, multiple studies have validated its potential in preclinical studies ^{[80][81][82][83][84]}[109,110,111,112,113].

Regardless of the method used for encapsulation and cell targeting, the route of administration has a great impact on therapy adherence rates. All the approved therapies based on oligonucleotides are administered via different types of injections, either intravitreal, subretinal, intrathecal, intravenous, intramuscular, or subcutaneous [4]. Considering the pain associated with the skin conditions described in this review, injections and the topical application of a therapy can exacerbate the discomfort. As a case in point, a patient receiving siRNA intradermal injections for the treatment of pachyonychia congenita reported intense injection-related pain. As a consequence, the patient had to be pre-treated with pain killers and anesthetics to mitigate the pain of the injections ^[51][81]. While the siRNA therapy led to disease regression, the pain of administration precluded its clinical translation, evidencing the need for alternative drug delivery regimens.

Topical delivery of nucleic acids has also been investigated to achieve local targeting of the skin ^[85][86][87][114,115,116]. Microneedles, a transdermal patch used to breach the stratum corneum in a minimally invasive manner, and electroporation alone and combined have successfully delivered gene therapies ^[88][89][117,118]. One example is the use of biocompatible microneedles to encapsulate and deliver CRISPR-Cas9 to treat inflammatory skin disorders ^[35][65]. However, microneedle delivery can be challenging with altered skin properties including open wounds characteristic of RDEB where the delivery may be inconsistent or extremely painful, respectively. On the other hand, the lack of the stratum corneum barrier in dermatological diseases characterized by wounds and skin lesions could be harnessed for topical delivery without the need for physical methods to disrupt the skin.

The route of administration has a direct impact on the clinical and commercial success of a given therapy. Many dermatological diseases are treated with macromolecules that need to be injected including peptides, proteins, and monoclonal antibodies. This includes 11 biologics approved by the FDA for the treatment of moderate to severe psoriasis. Injections pose a burden to many patients, revealing an underlying need to identify alternative drug administration options such as enabling absorption across the intestinal and skin barriers ^[90][119]. The oral route remains an attractive alternative for delivery, but despite being the preferred route of administration, it is often overlooked for gene delivery due to the susceptibility of nucleic acids to the intestinal environment. Seminal advances are key to overcoming such challenges, attaining new solutions for diseases with unmet needs, and optimizing existing treatments. As a matter of fact, diabetes care and the means of the delivery of biologics has evolved dramatically in the past few years. While injections were the only means for the successful delivery of biologics in the 1920s, in the recent decades, various novel approaches for oral delivery have emerged. These include the utilization of permeation enhancers, formulation approaches, and the use of emerging drug-device combinations.