Transdermal delivery is fashionable again, but enthusiasm should never outrun physiology. Sildenafil citrate—a mainstay for erectile dysfunction and a sometimes-blunt tool for premature ejaculation—works well by mouth, yet oral dosing is tethered to systemic peaks, interactions, and the occasional cardiology consult. A credible topical route promises faster local action with fewer whole-body consequences, provided we stop guessing about skin penetration and start measuring it with models that reflect human biology. A recent open-access study does exactly that: it builds a full-thickness 3D human skin equivalent derived from foreskin, challenges it with several sildenafil vehicles, and compares the outcomes to a well-established ex-vivo benchmark—porcine ear skin in Franz cells. The result is a practical map for formulation teams and clinicians who want local efficacy without systemic drama.
This article distills the study’s methods and findings into actionable guidance. We will translate rheograms and flux values into bedside implications, explain when a 3D human model can stand in for animal tissue (and when it cannot), and show, without hand-waving, how vehicle choice controls dermal uptake more than most other variables. Yes, there will be numbers. No, there will not be jargon for its own sake. A dose of friendly irony is included at no extra charge.
Why Consider Topical Sildenafil at All?
Oral sildenafil is efficient pharmacology: inhibit PDE-5, preserve cyclic GMP, amplify nitric oxide signaling, and let smooth muscle relax in penile tissue during sexual stimulation. The performance is reliable; the side-effects can be inconvenient. Peak plasma levels, interactions (notably nitrates), and cardiovascular concerns in vulnerable patients are the predictable baggage of a systemically delivered vasodilator. A local, skin-based route holds obvious appeal: reduce systemic exposure while delivering enough drug where it is actually needed. That logic has spurred a decade of transdermal concepts—microemulsions, transfersomes, nanolipid carriers, hydrogels—with enthusiastic in-vitro readouts and a modest clinical signal. What has been missing is a human-relevant, reproducible barrier to test across vehicles before anyone reaches for a prescription pad.
The study at hand attacks that gap with a full-thickness 3D human skin equivalent (HSE) grown from primary keratinocytes and fibroblasts isolated from foreskins. It is not a synthetic membrane; it is an engineered tissue that stratifies, forms a recognizable epidermis over a dermal compartment, and expresses proteins that underpin barrier function. In other words, it is a deliberate attempt to capture the biology of the intended application site, not just “skin in general.” For a drug aimed at penile tissue, that anatomical specificity is more than a nicety.
The model’s purpose is not to replace clinical trials; it is to sort formulations by penetration performance, flag obvious dead-ends, and reduce dependence on scarce human skin or ethically loaded animal models. As we will see, it does that job well—if you respect where its predictive power begins and ends.
Inside the Model: Building a Human Barrier You Can Trust
The investigators created the HSE by first seeding fibroblasts into Alvetex® polystyrene scaffolds to recreate a dermal matrix, then layering keratinocytes on top and lifting the construct to an air–liquid interface to drive epidermal differentiation. The resulting tissue measures roughly 0.6 mm thick, has a usable diffusion area of 1.12 cm², and—most importantly—stratifies in a way pathologists would recognize. Hematoxylin and eosin sections reveal basal, spinous, granular, and cornified layers in the right order. That is the morphology; the barrier markers tell the rest of the story.
Two proteins matter here. Involucrin, a scaffold for the cornified envelope, appears in upper spinous/granular layers, where it belongs during terminal differentiation. Claudin-1, a tight junction component, localizes to the granular layer, consistent with an epidermis that can keep ions and solutes at healthy distances. The model expresses both in the expected anatomical pattern. That does not make it identical to native human skin—no model is—but it does make it a credible substrate for permeation experiments that respect biology instead of imitating a coffee filter.
A practical advantage of 3D HSEs is reproducibility. Inter- and intra-individual variability confounds ex-vivo human and animal skin; engineered tissue narrows those error bars. Regulators have already leaned on such models for irritation and corrosion assessments, and permeability work is following the same path. The authors sensibly complement the HSE with porcine ear skin in Franz cells, the current workhorse surrogate for human tissue, to check concordance rather than demand faith. That two-track design is what lifts this paper from “interesting” to “useful.”
Vehicle First, Everything Else Later: What the Formulations Did
Sildenafil citrate was dispersed into three commercial transdermal vehicles as test beds, alongside a simple aqueous suspension:
- Formulation A: a liposomal cream (IPM, IPP, lecithin, fatty alcohols, carbomer, etc.).
- Formulation B: a more lipophilic liposomal organogel with higher viscosity.
- Formulation C: a classic oil-in-water (o/w) emulsion enriched with glycerin, C12–15 alkyl benzoate, and olive-derived lipids.
Rheology was not decoration; it explained behavior. All three showed shear-thinning (good for spreadability), but Formulation B was viscous and structurally tighter across temperature sweeps, the kind of matrix that clings to drug rather than letting it ride. The consequence was exactly what you would predict: lower permeation from B despite equal drug load, while A and C—less viscous, with enhancer-friendly architectures—let sildenafil move. Texture is not a cosmetic decision; it is a kinetic one.
In the 3D human model over 4 hours, aqueous suspension unsurprisingly delivered the highest receptor phase amount (~426 µg/cm²), confirming that sildenafil citrate can traverse the engineered barrier when available in solution. Among vehicles, Formulation A led (~217 µg/cm²), Formulation C followed (~165 µg/cm²), and Formulation B lagged (~54 µg/cm²). Steady-state fluxes and apparent permeability coefficients mirrored that order: A and C clustered around ~2.0 × 10⁻⁴ µg cm⁻² s⁻¹ and ~5 × 10⁻⁹ cm s⁻¹, while B trailed at roughly a third of that flux with Papp ~1.8 × 10⁻⁹ cm s⁻¹. Each of these differences reached statistical significance. Translation: vehicle architecture and viscosity controlled delivery far more than any esoteric variable.
Why this rank order? Two reasons. First, liposomal and o/w systems can enhance partitioning into the stratum corneum and sustain release toward viable layers; Formulation A pairs phospholipids with classic penetration enhancers, while Formulation C’s aqueous continuous phase and humectants hydrate the stratum corneum, easing diffusion. Second, viscosity throttles mobility: Formulation B’s gel network simply did not let sildenafil escape quickly enough. In transdermal design, “feels rich” to the hand often means “holds drug hostage” to the tissue.
The Reality Check: 3D Human Skin vs. Porcine Ear Skin
A model earns trust by surviving comparison. Using Franz diffusion cells, the investigators challenged porcine ear skin with the two best performers (aqueous suspension and Formulation A). As expected, absolute amounts in receptor fluid were lower in porcine skin than in the 3D model over the same 4-hour window (e.g., Formulation A: ~42 µg/cm² in porcine vs ~217 µg/cm² in HSE), and fluxes were roughly two-fold higher in HSE than in porcine tissue. That “HSE lets more through” finding is common; engineered skin usually has a slightly weaker barrier than native skin, but—crucially—the rank order is preserved.
Two diffusion parameters sharpen the comparison. First, for the aqueous suspension, diffusion coefficients calculated from lag time were not significantly different between HSE and porcine skin—evidence that the model captures passive diffusion of the dissolved drug reasonably well. Second, for Formulation A, diffusion coefficients diverged significantly between models, reflecting the extra complexity of semisolid vehicles interacting with each barrier’s microstructure. In plain English: the HSE is a good analogue for simple solutions; once you add complex rheology and excipients, you should still confirm results in a second system. That is not a failure; it is an honest boundary condition.
A practical footnote: lag time (t_lag) could be defined for porcine skin in some conditions (e.g., ~93 min for Formulation A), but was too short to quantify in the HSE for the fastest conditions, again echoing a more permissive barrier. For screening, that speed is an asset. For final go/no-go decisions, it is the reason you triangulate—HSE for throughput and ranking, porcine (or human ex-vivo) for calibration.
What These Data Mean for Clinical Translation
First, topical sildenafil is not fantasy. In this human-derived model, sildenafil citrate traversed the barrier in meaningful amounts within hours, and vehicles with appropriate architecture moved it better. That supports the clinical rationale for local delivery in erectile dysfunction (and plausibly in premature ejaculation, where desensitization strategies already exploit local pharmacology). A formulation that hydrates the stratum corneum, promotes partitioning, and does not imprison the drug in gel viscosity is your friend.
Second, local delivery does not erase systemic rules. Penile skin is vascular; even “local” delivery can reach the circulation. Patients on nitrates, with hypotension, or with unstable cardiovascular disease do not suddenly become ideal candidates simply because the drug starts on the surface. The upside is dose-sparing and smoother local kinetics; the downside is complacency. The study’s ex-vivo and in-vitro scope cannot quantify clinical systemic exposure; first-in-human PK over application sites and doses is the necessary next step.
Third, site matters. The model is built from foreskin—chosen deliberately because it matches the intended application area. That anatomical alignment counters a common translational failure: testing on dorsal forearm skin and pretending the result maps to genital tissue. If you plan a penile application, a foreskin-derived HSE is closer to the truth than generic epidermal models, and this study shows it behaves sensibly against a porcine benchmark.
Designing a Topical Sildenafil That Behaves: Practical Guardrails
- Let the vehicle do the heavy lifting. Favor liposomal creams or well-hydrating o/w emulsions with proven penetration enhancers (e.g., IPM, IPP, fatty acids). Avoid over-structured organogels that feel elegant but trap drug and blunt early flux. In the HSE, a liposomal cream (Formulation A) and an o/w emulsion (Formulation C) clearly outperformed a more viscous lipophilic organogel (Formulation B).
- Match rheology to the task. Shear-thinning is welcome; excess viscosity is not. The study’s rheology data foreshadowed performance: the highest-viscosity vehicle delivered the lowest sildenafil flux. If your formulation resists spreading, it will probably resist releasing.
The paradox of topical design is that “cosmetic feel” and “pharmacokinetic release” are the same problem seen from different chairs. The best-performing vehicles in this work were not exotic; they were well-balanced emulsions and liposomal creams that hydrated, partitioned, and flowed. That is good news for translation—these platforms are manufacturable at scale and familiar to regulators.
A Quick, Sensible Workflow From Bench to Body
Begin with HSE screening to rank vehicles by flux and Papp over the first few hours, using the penile-relevant foreskin model. Promote the top two to porcine skin in Franz cells to confirm rank order and quantify lag time and retention in stratum corneum versus viable epidermis/dermis. The present study provides the template: it preserved rank order and highlighted where diffusion coefficients match (simple solutions) and where they diverge (complex semisolids). Then, and only then, is it time for pilot human PK at the application site.
While you are at it, measure what matters: early-phase receptor amounts (for onset), steady-state flux, Papp, and skin layer retention after tape-stripping. The authors’ protocol extracts drug from stratum corneum and combined epidermis-dermis after 4 hours—exactly the data you need to ensure the drug is not just on the skin but also in the target layers. Pair this with rheology (flow curves, temperature sweeps) so you can explain performance mechanistically rather than guessing.
Finally, do not forget OECD-style skin absorption guidance and model validation logic. The study explicitly anchors its porcine Franz cell work to established recommendations, and it treats the HSE as a complement, not a replacement, which is precisely how regulators expect to see it used. The payoff is efficiency without shortcuts.
Limits You Should Respect (and How to Work Around Them)
Engineered skin equivalents, even the good ones, often have higher permeability than native human skin. The present HSE is no exception; its higher fluxes reflect a barrier that is biologically credible but not impenetrable. That is acceptable for ranking and mechanistic comparison, less so for predicting absolute clinical exposure. The authors acknowledge this and use porcine tissue as a calibration anchor—a wise habit to copy.
Vehicle-barrier interactions are idiosyncratic. In this study, diffusion coefficients for a simple aqueous solution matched between HSE and porcine skin, but diverged for a semisolid (Formulation A). The lesson is not “models are bad,” but “complex formulations need multi-model confirmation.” If your early HSE screen says a gel is brilliant, prove it in a second barrier before you commission labels.
Finally, scope matters. This is permeation science, not an efficacy trial. The tissue expresses involucrin and claudin-1 and looks like skin under a microscope; it does not measure erectile response, ejaculatory latency, or partner satisfaction. Those belong in carefully staged clinical studies that build on the biopharmaceutic feasibility proven here. Good models answer “can it get there?”; only clinics answer “does it help?”.
Putting It All Together: The Conservative, Effective Path Forward
Topical sildenafil has a plausible mechanism, a supportive human-derived barrier model, and clear formulation rules: hydrate, partition, and avoid over-viscous traps. In foreskin-based HSE, sildenafil citrate crosses readily when presented in solution, and it crosses well from a liposomal cream or o/w emulsion that respects both skin biology and rheology. Against porcine skin, the rank order holds, which is exactly what we want from a model tasked with smart screening rather than perfect prophecy.
If you are a formulator, the message is to choose vehicles that move—liposomal creams and balanced o/w emulsions—then verify in a second barrier. If you are a clinician, the message is conditional optimism: a topical path is credible, but it must be developed with the same discipline we demand of any pharmacotherapy that affects the vasculature. When the science honors the skin, the clinic benefits; when it shortcuts the barrier, everyone notices later.
The wry conclusion? Sometimes the smartest innovation is not a new nanomaterial but picking the right cream and testing it on the right skin. Here, the right skin was human-derived foreskin, and the right cream was the one that let the drug leave. That modest sentence contains most of the art and science of transdermal sildenafil.
FAQ
1) Can a human 3D foreskin model really replace animal skin for permeability testing?
For simple solutions, the study found no significant difference in diffusion coefficients between the 3D human foreskin model and porcine ear skin, suggesting the HSE is a valid alternative for early screening. For complex semisolids, differences emerged, so a confirmatory porcine (or human ex-vivo) study remains wise before clinical translation. Use the HSE to rank and optimize; use porcine/human tissue to calibrate.
2) Which topical vehicle is most promising for sildenafil citrate?
In head-to-head tests on the human 3D model, a liposomal cream and a well-balanced o/w emulsion delivered significantly higher flux and permeability than a more viscous lipophilic organogel. The aqueous suspension permeated most of all (proof of feasibility), but is not a practical product on its own. Favor vehicles that hydrate the stratum corneum, promote partitioning, and avoid excessive viscosity that traps drug.
3) Does topical delivery eliminate systemic risks like nitrate interactions?
No. Topical delivery can reduce systemic peaks but does not negate vasodilatory pharmacology. Patients on nitrates or with unstable blood pressure still require caution. The present work is permeation science; systemic exposure and safety must be addressed in first-in-human PK studies at penile application sites before routine use.
