Gene Delivery to the Skin – How Far Have We Come?

Purpose: Gene therapies are powerful tools to prevent, treat, and cure human diseases, but their application to skin diseases has received little attention despite the easy accessibility of skin and urgent medical need. The main obstacle is the unique barrier properties of human skin, which severely limit absorption of biomacromolecules and efficient delivery of nucleic acid payloads. This review discusses current approaches, successes, and failures of cutaneous gene therapy and provides guidance for next-generation concepts, with delivery strategies as the major translational obstacle. ---

Hypothesis: No formal testable hypothesis is proposed; the central thesis is: if delivery strategies—viral, non-viral, and physical—are rationally engineered to overcome the skin barrier and target viable epidermal cells or skin stem cells, then gene therapies can be translated for skin diseases, especially rare genodermatoses such as epidermolysis bullosa (EB). ---

Aims: - Discuss current approaches, successes, and failures of cutaneous gene therapy. - Provide guidance toward next-generation concepts. - Focus on delivery strategies as the major obstacle preventing the full potential of gene therapies for skin disorders. - Review viral, non-viral (lipid-based, polymeric), and physical delivery methods. - Summarize the state of the art in clinical trials for skin diseases. - Highlight future directions, including human-relevant models and targeting skin stem cells. ---

Delivery system: Viral vectors: - Adeno-associated viral vectors (AAVs), adenoviral vectors, lentiviral vectors, retroviruses. - Payloads: DNA, cDNA, CRISPR components; limitations include cargo size (~4.7 kb for AAV, ~10 kb for lentivirus), immunogenicity, insertional mutagenesis, manufacturing cost. Non-viral lipid-based nanoparticles (LNPs): - Ionizable cationic lipids (e.g., DLin-MC3-DMA), lipoplexes, Lipofectamine, zwitterionic lipid cephalin. - Payloads: siRNA, mRNA, DNA, self-amplifying RNA, CRISPR/Cas9 RNP, sgRNA. - Advantages: clinically validated (ONPATTRO), large cargo capacity, reduced carrier toxicity, endosomal escape via pH-dependent protonation. Polymeric nanoparticles: - Polyethyleneimine (PEI) and derivatives, poly(β-amino ester) (PBAE), highly branched PBAE (HPBAE), PLGA, cyclodextrin-PEI, redox- and pH-sensitive polymers. - Payloads: siRNA, miRNA, minicircle DNA, plasmid DNA. - Polyplexes typically 50–200 nm. Physical methods: - Electroporation, iontophoresis, sonoporation, hollow and dissolvable microneedles (50–1500 µm). - Enhance penetration across stratum corneum and cell membranes. Targets and payloads: - Genodermatoses: EB (COL7A1, LAMB3), congenital ichthyosis (TGM1), Netherton syndrome (SPINK5). - Inflammatory skin diseases: psoriasis (IL-17A receptor). - Skin cancer: melanoma (STAT3 siRNA, miR155, AONs). - Target cells: keratinocytes, skin stem cells, fibroblasts, dendritic cells, tumor-associated macrophages. - Administration routes: ex vivo (cell isolation, gene correction, regrafting), in situ topical/intradermal, intratumoral, iontophoresis, microneedles; systemic delivery unlikely for skin. ---

Approach: Narrative review of published literature. No primary experimental groups. Model systems discussed include: - In vitro: human skin explants, 3D skin models, primary keratinocytes, fibroblasts, dendritic cells, melanoma cell lines. - In vivo: mouse models, xenografted rodent models, pig skin (suggested), human-based tissue grafted onto mice. - Clinical trials: 1052 gene therapy trials identified, only 23 focused on skin conditions; EB most studied, followed by melanoma, Netherton syndrome, and congenital ichthyosis (Table 2 lists trials including NCT04186650, NCT04227106, NCT02810951, NCT01263379, NCT01545323, NCT03536143, NCT04047732, NCT03605069). - Disease context: epidermolysis bullosa, congenital ichthyosis, Netherton syndrome, melanoma, psoriasis, pachyonychia congenita. ---

Key methods: Techniques highlighted across cited studies: - Luciferase reporter expression for LNP delivery in human skin explants. - Flow cytometry/fluorescence for transfection efficacy. - mRNA/protein knockdown (STAT3, IL-17A receptor). - Tumor weight and volume measurement. - Skin adhesion area and anchoring fibril formation. - Gene editing efficiency (indels, CRISPR correction). - Cytotoxicity and biocompatibility assays. - Clinical trial endpoints: safety, proof of mechanism, preliminary efficacy. ---

Key results: - Skin barrier: Human skin efficiently absorbs only small molecules (MW < ~800 Da) with moderate lipophilicity (logP 1–3), making biomacromolecule delivery challenging. - Clinical trial landscape: Of 1052 gene therapy trials, only 23 focus on skin conditions; ~70% of ongoing gene therapy trials are viral vector-based. - LNP optimization: Zwitterionic lipid cephalin enabled a 7-fold increase in luciferase expression compared with cationic DOTAP in human skin explants. - PEI efficacy: PEI transfection efficacy ranges between 1/10th and 1/1000th of an adenoviral vector depending on molecular weight; branched PEI complexes DNA 15-fold more efficiently than linear PEI. - Polymeric NPs: Cyclodextrin-PEI improved biocompatibility and increased transfection efficacy 4-fold. Redox/pH-sensitive NPs delivered miR155 to tumor-associated macrophages, inducing miR155 expression 100–400 times in vitro and after intratumoral injection in melanoma xenografts. - PBAE/HPBAE: Highly branched PBAE delivered minicircle DNA encoding COL7A1 to EB patient keratinocytes, restored collagen VII production, and showed lower cytotoxicity than PEI; licensed by Amynt Pharma. - Microneedles: STAT3 siRNA delivered intradermally reduced STAT3 mRNA by 60% and tumor weight/volume by 80% in a melanoma mouse model. - Iontophoresis: AON-loaded polyplexes increased skin delivery 3-fold vs free AONs and reduced tumor volumes to 45% vs 20% in mice. - LNP AON for psoriasis: Topical LNP-AON targeting IL-17A receptor reduced protein expression by 72–75% in psoriasis-like mouse models, 3D skin models, and normal human explants. - CRISPR in vivo: C7 gene function restored in EB mouse models; skin adhesion improved from 30% to 60%, but only 2% of stem cells were edited. - Ex vivo gene therapy success: A 7-year-old boy with life-threatening EB was treated with retrovirus expressing LAMB3 cDNA; transgenic skin grafts replaced 80% of his skin, and after 21 months his skin had a normal appearance without detachment. - AON EB studies: Restoration of C7 in 6% of patient cells initially; later ex vivo studies achieved 6–50% restoration, but in vivo intradermal/intravenous administration yielded only 10–14%. ---

Interpretation: The authors conclude that gene therapies have great capacity to treat all types of skin diseases, with EB patients likely to benefit first due to high medical need, orphan designations, and advanced clinical development. Restoration of only 10% of normal gene function is considered sufficient to alleviate skin conditions. Viral delivery remains the gold standard but has safety, capacity, and immunity limitations. Non-viral LNPs are the most advanced non-viral systems, but physical methods combined with non-viral delivery likely hold the greatest potential for in situ gene therapy. The field needs better delivery, human-relevant models, and targeting of skin stem cells for long-lasting effects. “We have the tools for gene therapy; we now need to optimize their delivery to the target site in the skin.” ---

Limitations: - Skin barrier: The stratum corneum and tight epidermal–dermal junction severely limit penetration of biomacromolecules; most topically applied nanoparticles do not overcome the stratum corneum, even in damaged skin. - Low transfection: Only a small number of cells get transfected; keratinocytes and skin stem cells are hard to transfect. - Viral vector limitations: Limited cargo size, immunogenicity, pre-existing immunity, insertional mutagenesis risk, complex and costly manufacturing. - Non-viral limitations: Endosomal escape, lower transfection efficacies than viral vectors, carrier-related toxicity (e.g., PEI), and limited clinical translation. - Physical methods: Electroporation causes cytotoxicity; sonoporation has low/uncontrolled transfection; microneedles have limited loading capacity and are not suitable for whole-body administration. - Animal model mismatch: Mouse skin differs significantly from human skin (1–3 epidermal layers in mice vs 15–20 in humans), limiting translational value. - Lack of suitable disease models: Need human skin explants, pig skin, tissue-engineered skin models, and human-based tissue grafted onto mice. - Long-term effects: Targeting keratinocytes yields effects limited to ~28 days due to continuous epidermal regeneration; curative approaches require skin stem cell targeting, but knowledge of skin stem cells is poor. - Systemic delivery unlikely: Lack of vasculature in viable epidermis and tight epidermal–dermal junction prevent systemic biomacromolecule delivery to target skin cells. - As a review: Not a systematic review or meta-analysis; no primary data.