Fractional CO2 laser ablation leads to enhanced permeation of a fluorescent dye in healthy and mycotic nails-An imaging investigation of laser-tissue effects and their impact on ungual drug delivery.

PURPOSE: Conventional oral antifungal therapies for onychomycosis (OM) often do not achieve complete cure and may be associated with adverse effects, medical interactions, and compliance issues restricting their use in a large group of patients. Topical treatment can bypass the systemic side effects but is limited by the physical barrier of the nail plate. Ablative fractional laser (AFL) treatment can be used to improve the penetration of topical drugs into the nail. This study visualized the effects of laser ablation of nail tissue and assessed their impact on the biodistribution of a fluorescent dye in healthy and fungal nail tissue.
METHODS: For the qualitative assessment of CO2 AFL effects on healthy nail tissue, scanning electron microscopy (SEM), coherent anti-Stokes Raman scattering microscopy (CARS-M), and widefield fluorescence microscopy (WFM) were used. To quantitate the effect of laser-pretreatment on the delivery of a fluorescent dye, ATTO-647N, into healthy and fungal nail tissue, ablation depth, nail plate thickness, and ATTO-647N fluorescence intensity in three nail plate layers were measured using WFM. A total of 30 nail clippings (healthy n = 18, fungal n = 12) were collected. An aqueous ATTO-647N solution was directly applied to the dorsal surface of 24 nail samples (healthy n = 12, fungal n = 12) and incubated for 4 hours, of which half (healthy n = 6, fungal n = 6) had been pretreated with AFL (30 mJ/mb, 15% density, 300 Hz, pulse duration <1 ms).
RESULTS: Imaging revealed a three-layered nail structure, an AFL-induced porous ablation crater, and changes in autofluorescence. While intact fungal samples showed a 106% higher ATTO-647N signal intensity than healthy controls, microporation led to a significantly increased fluorophore permeation in all samples (p < 0.0001). AFL processing of nail tissue enhanced topical delivery of ATTO-647N in all layers, (average increase: healthy +108%, fungal +33%), most pronounced in the top nail layer (healthy +122%, fungal +68%). While proportionally deeper ablation craters correlated moderately with higher fluorescence intensities in healthy nail tissue, fungal samples showed no significant relationship.
CONCLUSION: Fractional CO2 laser microporation is a simple way of enhancing the passive delivery of topically applied ATTO-647N. Although the impaired nail plate barrier in OM leads to greater diffusion of the aqueous solution, AFL can increase the permeability of both structurally deficient and intact nails.

INTRODUCTION

Onychomycosis (OM), the fungal infection of the nail unit, comprises more than half of all nail disorders and 30% of cutaneous fungal infections. With a rising prevalence of over 5.5%, a high recurrence rate of up to 20%, a reported decrease in quality of life, and potentially severe long‐term complications including acute bacterial cellulitis, OM constitutes a major public health concern. , , , , , ,

Despite the unfavorable side‐effect profile of systemic antifungal medication, OM often requires oral treatment owing to the low efficacy of topical treatment options and limited indication of laser devices (Table S1). Consequently, the concept of device‐assisted pretreatment for penetration enhancement is rapidly gaining interest as a way of increasing the efficacy of topical formulations. Ablative fractional lasers (AFL) can create an array of microscopic laser channels while keeping surrounding tissue intact. Recent studies have succeeded in using AFL as a method of pretreatment to enhance topical delivery into healthy nail tissue. While the majority of clinical laser‐assisted drug delivery studies have been conducted on skin conditions, there is a mounting body of evidence, summarized in Table 1, supporting the use of AFL as a penetration enhancer for topical treatment of nail disease. , In‐depth knowledge of the nail plate structure and laser–tissue interactions is crucial for understanding the challenges diseased nails pose to light‐based treatments and topical drug delivery. Clinical empirical evidence on the application and utility of AFL as pretreatment for enhanced efficacy of topical antifungals for treatment of OM has outpaced preclinical investigations. While laser‐assisted ungual delivery (LAUD) has been evaluated in healthy nail tissue, a comparative investigation and visualization of a topically applied model drug in ex vivo fungal and healthy nail tissue to determine the effect of AFL pretreatment for enhanced transungual delivery is lacking.

Table 1

Overview of studies on clinical and preclinical laser‐assisted drug delivery in healthy and diseased nail tissue

Year and author	Sample size	Condition	Compound	MW (g/mol)	Vehicle	Application	Laser type and device	Settings	Density
2013	2	Onychodystrophy	Desoximetasone 0.25%	376.4	Lotion	Once daily	CO2 (10,600 nm);	160 mJ	150 spots cm−2
							Mosaic eCO2, Lutronic	(3 passes)
EH Lim 14
2013	9	Onychomycosis	Amorolfine 5%	317.5	Lacquer	Once weekly	Er:YAG (2,940 nm); Etherea, Industra	50 mJ mtz−1	N.A.
								(number of passes dependent on nail thickness and patient tolerability)
OOc Morais 15 , a
2014	24	Onychomycosis	Amorolfine 5%	317.5	Cream	Once daily	CO2 (10,600 nm);	160 mJ	150 spots cm−2
							Mosaic eCO2, Lutronic	(2–3 passes)
EH Lim 16
2015	7	Onychomycosis	MAL 16%	181.6	Cream	Before PDT	CO2 (10,600 nm); –	160 mJ	600 µm spacing between mtz
								(3 stacks)
GB deOliveira 17
2015	75	Onychomycosis	Terbinafine 1%	291.4	Cream	Once daily	CO2 (10,600 nm);	99 mJ	410 spots cm−2
							2030CI, Wuhan Qi Zhi Laser Technology	(2–6 passes)
AK Bhatta 18
2016	60	Onychomycosis	Luliconazole 1%	354.3	Cream	Once daily	CO2 (10,600 nm);	10–15 mJ	10%
BR Zhou 19							AcuPulse, Lumenis	(1–3 stacks)
2016	9	Onychomycosis	Amorolfine 5%	317.5	Lacquer	Twice weekly	Er:YAG (2,940 nm);	35–62 J cm−2	120 spots cm−2
J Zhang 20							Sciton	(3 passes)
2017	31	Onychomycosis	Terbinafine 1%	291.4	Cream	Once daily	CO2 (10,600 nm);	15 mJ	15%
							AcuPulse; Lumenis	(multiple passes)
J Shi 21
2017	54	Onychomycosis	(a) ALA 20%	(a) 167.6	(a) Cream	(a) Before PDT	CO2 (10,600 nm);	150 mJ	3%
			(b) Amorolfine 5%	(b) 317.5	(b) Lacquer	(b) Once weekly	Lumenis Ultrapulse	(single pass)
A Koren 22
2018	102	Onychomycosis	(a) Tazarotene 0.1%	(a) 351.5	(a) Gel	Once daily	CO2 (10,600 nm);	Power 10 W	700 µm spacing between mtz
							Smartxide DOT, DEKA, Italy	(3 stacks)
			(b) Tioconazole 28%	(b) 387.7	(b) Solution
EB Abd El‐Aal 23
2019	30	Onychomycosis	Tioconazole 28%	387.7	Lacquer	Twice daily	CO2 (10,600 nm);	Power 11–15 W	500 µm spacing between mtz
RA El‐Tatawy 24								(1–3 stacks,
								1–3 passes)
							Smartxide DOT, DEKA, Italy
2020	21	Onychomycosis	Methylene Blue 2%	319.9	Solution	Before PDT	CO2 (10,600 nm);	180 mJ	500 µm spacing between mtz
M Abdallah 25								(3 stacks,
								2 passes)
							Fire‐xel scanner system, Bison
2020	160	Onychomycosis	Terbinafine 1%	291.4	Ointment	Once daily	CO2 (10,600 nm);	No fixed setting (dependent on heat sensation)	1000–1500 µm spacing between mtz
B Rajbanshi 26							Wuhan Miracle Laser Co., Ltd.
2020	120	Onychomycosis	Tioconazole 28%	387.7	Solution	Twice daily	CO2 (10,600 nm); SmartXide Square, DEKA, Florence, Italy	Power of 10–15 W (according to nail thickness), 3 stacks, pulse duration of 500 μs	700–800 μm spacing between mtz
AM Zaki 27
2021	78	Onychomycosis	Amorolfine 5%	317.5	Lacquer	Twice weekly	Er:YAG (2,940 nm);	35–60 J cm−2 (dependent on nail thickness and patient tolerability)	120 spots cm−2
							Sciton
J Zhang 28
2021	22	Nail psoriasis	Calcipotriol 50 μg/g + Betamethasone 0.5 mg/g	392.5 + 412.6	Gel	Once daily	CO2 (10,600 nm)b;	60–110 mJ (dependent on patient tolerability)	3%
							Ultrapulse Encore, Lumenis Ltd.
W Shehadeh 29
2021	27	Nail psoriasis	Tazarotene 0.1%	351.5	Gel	Once daily	CO2 (10,600 nm);	140 mJ	150 spots cm−2
N Essa Abd Elazim 30							Daeshin Enterprise Co., Ltd. Model: Multixel	2–3 passes
2014	1	Healthy	Saline	58.4	Solution	Once (<5 min)	CO2 (10,600 nm);	20–50 mJ	10%
							UltraPulse Encore Active FXTM; Lumenis
CH Yang 31
2016	3	Healthy	Sulconazole 1%	460.8	(a) Solution	Once (<5 min)	CO2 (10,600 nm);	20–50 mJ	N.A.
							UltraPulse Encore Active FXTM
					(b) Cream
MT Tsai 32
2017	3	Healthy	Caffeine 20 mg/mL	194.2	Solution	Once (72 h)	YDFL (532 nm)c;	31–250 mJ pore−1	500 spots cm− 2
							Fianium FP‐1060‐5‐FS
S Vanstone  13
2019	N.A.	Healthy	Methylene Blue 0.5%–3%	319.9	Gel	Before irradiation (dual stage LADD)	Er:YLF (2,810 nm);	3.9–4.1 mJ	N.A.
AV Belikov 33							–
2019	N.A.	Healthy	(a) Amorolfine 5%	(a) 317.5	Lacquer	Once (24 h)	CO2 (10,600 nm);	50–80 mJ	N.A.
			(b) Amorolfine 5% + Thioglycolic acid 5%	(b) +92.1			Candela CO2RE laser, Syneron Candela, Wayland, MA, USA
			(c) Amorolfine 5% + Urea 5%	(c) +60.6
I Sveikauskaite 34
2020	N.A.	Healthy	(a) Naftifine 5%	(a) 323.9	Lacquer	Once (24 h)	CO2 (10,600 nm);	50–80 mJ	N.A.
							Candela CO2RE laser, Syneron Candela, Wayland, MA, USA
			(b) Naftifine 5% + Thioglycolic acid 5%	(b) +92.1
			(c) Naftifine 5% + Salicylic acid 0.5%	(c) +138.1
I Sveikauskaite 35
2021	5	Healthy	Methylene Blue 0.001%	319.9	Solution	Before irradiation (dual stage LADD)	Er:YLF (2,810 nm);	1–4 mJ	N.A.
AV Belikov 36							–

Abbreviations: ALA, 5‐Aminolevulinic acid; LADD, laser‐assisted drug delivery; MAL, methyl aminolevulinate; mtz, microthermal zone.

Protocol only; study results published as thesis.

plus 595‐nm PDL (Perfecta, Syneron‐Candela) on nail folds.

Ytterbium‐doped fiber laser with frequency doubling lithium triborate crystal.

This study evaluated the distribution pattern and signal intensity of a fluorescent dye following laser‐pretreatment of both healthy and fungal toenails using widefield fluorescence microscopy (WFM). The objective of this study was to investigate the effects of AFL on drug delivery in both healthy and diseased nail tissue. As a preliminary step, the effects of fractional CO2 laser ablation of healthy nail tissue were investigated using scanning electron microscopy (SEM), coherent anti‐Stokes Raman scattering microscopy (CARS‐M), and WFM, and compared to previous investigations. , Following that, the signal intensity of a topically applied fluorescent tracer dye, ATTO‐647N, was measured on WFM images of healthy and fungal nail samples with and without AFL processing. By visualizing the laser–tissue effects and quantitating their impact on the diffusion of topically applied compounds in healthy and fungal nail tissue, this study tried to provide a deeper understanding of LAUD to further refine this technique.

MATERIALS AND METHODS

Study setup

To qualitatively demonstrate the effect of an ablative fractional CO2 laser (DeepFX, UltraPulse CO2, Lumenis; λ = 10,600 nm) on the micromorphology of healthy nail plate tissue, we used SEM, CARS‐M, and WFM. For the quantitative assessment of AFL poration and ATTO‐647N fluorescence intensity, WFM images of healthy and mycotic samples were captured.

Tissue specimens and processing

For the initial investigation of laser–tissue effects, fingernail clippings (n = 3) from two healthy, volunteers (Caucasian male, aged 26 and 51), were clipped, cleaned, and treated with a single pass of AFL (80 mJ/mb, 5% density). The image acquisition and image analysis were performed at the Danish Molecular Biomedical Imaging Center (DaMBIC, University of Southern Denmark).

For the transungual delivery investigation, nail samples from healthy participants (n = 18) and anonymous discard clippings from patients with microscopy‐verified fungal nail disease (n = 12) from the Department of Dermatology at the Copenhagen University Hospital in Denmark were collected. Direct microscopy was performed on anonymous samples incubated with an optical whitening agent, Blankophor, for 30 minutes to detect fluorescent fungal elements (hyphae, pseudohyphae, or spores) as part of routine diagnostics. The edges of the nail samples were trimmed to reduce their curvature and flatten the surface to facilitate a perpendicular beam impact before cleaning with deionized water. The nail clippings were patted dry before undergoing laser ablation and treated in one batch to avoid hydration‐dependent variation in laser–tissue interaction. Half of the fungal samples as well as nine healthy clippings (n = 15) were anchored at their lateral edge with adhesive tape to a styrofoam board and processed with the same AFL device (30 mJ/mb, 10% density, 300 Hz frequency, fixed pulse duration <1 ms). Twenty‐four samples were labeled with a tracer dye solution as described below. All thirty samples were embedded in optimal cutting temperature medium (Tissue TEK OCT) before sectioning on a microtome‐cryostat (Leica CM 3050S) at fixed temperatures (20°C in the chamber and −15°C for the specimen). To avoid tissue tears, samples were sectioned transversally in parallel to the fiber direction of the intermediate layer and the free edge of the nail plate at a thickness of 10 µm. For each nail sample, nine cryosections were prepared (n = 270).

Scanning electron microscopy

SEM enables high‐resolution examination of surfaces by scanning them with a focused beam of electrons and detecting the emission of secondary electrons. This electron microscopy method can be used to study the topography of solid tissues and ultrastructure of laser–tissue effects. To visualize and characterize the surface texture of nail tissue after AFL processing, SEM (FEI Quanta 200) imaging of unsectioned nail clippings was performed at an accelerating voltage of 10 kV in high vacuum mode. As nail tissue is not considered beam‐ or vacuum‐sensitive, gold sputter coating was omitted.

Coherent anti‐Stokes Raman scattering microscopy

CARS‐M is a label‐free chemically specific imaging technique capable of generating image data based on tissue‐intrinsic molecular vibrations, directly visualizing water and lipid content. While conventional microscopic imaging techniques fail to directly visualize molecules in the tissue, CARS‐M can visualize specific molecules such as water and lipids in tissue. To describe the influence of AFL on nail tissue composition, CARS‐M (Leica TCS SP8 microscope with a ×40 IRAPO objective) was performed with a laser tuned in on OH and CH‐group stretching vibrations to visualize water (3260 cm−1) and lipid (2850 cm−1) content, respectively. The laser used was a picoemerald (APE) with a Stokes beam at 1064 nm, and a pump beam at for lipids 816.4 nm and 790 nm for water. The CARS‐M signal was detected using an epi‐detector with a short pass filter (Multiphoton‐emitter HC 770/SP) to block the lasers and a bandpass filter at 661/11 BrightLine HC for the lipids, and at 620/14 BrightLine HC for the water. The filters were acquired from AHF Analysentechnik.

ATTO‐647N labeling

ATTO‐647N is a hydrophilic cationic fluorescent label (MW 746 g/mol) with a high quantum yield in the red spectral region (λex = 646 nm, λem = 664 nm) (Atto‐TEC GmbH). In addition, ATTO‐647N has been shown not to bind to keratin, that is, better odds of faster distribution. For this study, a 100 µM aqueous solution of the carboxy derivative of ATTO‐647N in dimethyl sulfoxide (DMSO), a penetration enhancer, and phosphate‐buffered saline (PBS) was prepared (500 µM stock solution in DMSO diluted in PBS to 100 µM). Following cleaning with deionized water, 24 samples (healthy: n = 12, fungal: n = 12; half of both groups AFL‐pretreated) were placed on PBS‐soaked gauze. A droplet of the ATTO‐647N solution was applied to the dorsal surface of intact and AFL‐pretreated nails, and incubated for 4 hours in the dark on top of PBS‐soaked gauze. The excess staining agent was removed by washing the nail samples for 2 minutes with PBS.

Widefield fluorescence microscopy

The cryo‐sections were placed on glass slides (Superfrost) and stored at −20°C until inspected under an inverted widefield fluorescence microscope (Olympus IX70. Fluorescence Microscope, Olympus Denmark A/S) equipped with a xenon lamp at 10‐fold magnification using a UV‐filter cube (λex: 375/28 nm, λem: beam splitter 415 nm λem 460/50 nm, exposure time of 150 ms) to capture morphological images and a custom‐made filter cube for ATTO‐647N fluorescence (λex:640/30 nm, λem: beam splitter 660 nm, λem: 690/50 nm, exposure time of 10 ms) in a room with dim green light to avoid photobleaching.

The stability of the excitation light was monitored before each microscopy session by an external power meter and fluorescence measurements on a fluorescent plastic standard slide (Chroma Technology, Corp.). The power of the light source proved stable during the sampling process (±10%). Standard measurements showed no significant variation in illumination.

Image analysis

All images were analyzed using FIJI (ImageJ 1.49, NIH ). SEM and CARS‐M images were assessed qualitatively. On UV‐WFM images, nail plate thickness and depth of AFL‐micropores were measured and evaluated for nail dystrophy signs. Nail plate thickness was measured as the shortest distance between dorsal and ventral layer. Micropores were defined as the absence of tissue‐corresponding signal. In cross‐sectional images, micropores were observed to be of triangular or semi‐elliptical shape. The degree of poration describes the relationship between the depth of an ablation crater and the thickness of the treated tissue, where complete poration (100%) represents ablation to a depth corresponding to the total thickness of the nail plate. The morphometric assessment of laser‐processed nail tissue has been described in detail in our previous publication. Fluorescence signal intensity in ATTO‐647N‐WFM images was quantitated in a standardized manner by drawing a line with a width of 100 pixels horizontally across the displayed nail sample in three regions of interest: the dorsal, the intermediate, and the ventral layer of the sample.

Statistics

Based on a significant Levene's test, nonparametric analysis using Kruskal–Wallis and paired Wilcoxon rank tests with Hochberg correction for multiple comparisons were conducted. By calculating the Pearson correlation coefficient (r), the linear association between fluorescence intensity and poration depth was explored. Data were visualized using grouped boxplots and scatterplots. Statistical evaluation and data visualization were performed using R and the ggplot2 package in R Studio.

RESULTS

Microscopy images were assessed qualitatively for the micromorphology of laser–tissue effects and nail plate ultrastructure. Quantitatively, morphometric dimensions of  AFL micropores and ATTO‐647N fluorescence signal intensities were measured.

Morphology and morphometry

SEM images of healthy nail tissue displayed a scale‐like dorsal nail surface with an elevated rim‐like structure surrounding the laser pore (Figure 1). A highly porous channel structure resembling a truncated cone connected the slightly larger dorsal aperture with the ventral surface opening. Nail tissue revealed nonconductive properties, resulting in electron accumulations on its surface, thereby creating charging artifacts that can mask the trilaminar structure.

Figure 1

Scanning electron microscopy image of healthy nail tissue exposed to 80 mJ/mb of ablative fractional CO2 laser (left: side view, right: top view). Images show complete poration of the nail plate and porous surface structure of the ablation zone

Using CARS‐M, however, the nail could be clearly subdivided into three distinct layers based on their water/lipid content, consistent with reported thickness relation (3:5:2) in literature (Figure 2). Rim‐like structures surrounding the laser channels, corresponding to coagulated keratin, did not exhibit any lipid signal, providing evidence for a potential break‐down of lipids following AFL‐processing. Strong green‐blue autofluorescent signal stemming from the laser‐channel surface and naturally occurring autofluorescent signals were only visible at the surface of the nail plate, both ventral and dorsal sides. After AFL‐processing, the pore surface and the ventral bottom layer also displayed a substantial increase in autofluorescence in WFM images.

Figure 2

Coherent anti‐Stokes Raman scattering microscopy image of a nail plate following 80‐mJ/mb AFL processing. While autofluorescence (top left, green pseudocolor) is seen mainly in the laser channel as well as the thermally affected ventral nail layer and the surface of the dorsal layer, the water (mid‐left, blue pseudocolor) and lipid (bottom left, red pseudocolor) content are restricted to the middle portion of the nail plate. The overlay (right) shows a trilaminar structure. AFL, ablative fractional laser

Fluorescence microscopy of untreated healthy controls and laser‐treated healthy controls without ATTO‐647N labeling resulted in equally low red background fluorescence. While we observed an increase in autofluorescence around the ablation crater, this peak in signal intensity was strictly limited to a thin rim of coagulated tissue. Morphometric measurements of the nail plate thickness and the ablation crater depth based off UV‐WFM images are summarized in Table 2. Average nail plate thickness was lower in healthy samples (372 µm; SD 75 µm) than in fungal nails (607 µm; SD 163 µm). The mean ablation depth (unlabeled healthy: 163 µm [SD 53 µm], ATTO‐647N‐labeled healthy: 199 µm [SD 56 µm], ATTO‐647N‐labeled fungal: 195 µm [SD 59 µm]) was not influenced by fungal infection but showed high variation resulting in correspondingly inconsistent degree of poration (SD > 15%). Signs of nail plate dystrophy (as illustrated in Figure 3), and autofluorescence intensities higher than those found in healthy tissue were visible in all fungal samples.

Table 2

Overview of morphometric microscopy measurements of nail tissue sections presented as means and standard deviations

Sample	Number of samples (sections)	Nail thickness (SD)	Ablation depth (SD)	Degree of poration (SD)
ATTO‐647N‐labeled fungal nail with AFL	6 (54)	607 µm (163 µm)	195 µm (59 µm)	35% (16%)
ATTO‐647N‐labeled healthy nail with AFL	6 (54)	372 µm (75 µm)	199 µm (56 µm)	56% (21%)
Unlabeled healthy nail with AFL	3 (27)	450 µm (80 µm)	163 µm (53 µm)	38% (15%)
ATTO‐647N‐labeled fungal nail	6 (54)	572 µm (266 µm)	‐	‐
ATTO‐647N‐labeled healthy nail	6 (54)	487 µm (98 µm)	‐	‐
Unlabeled healthy nail	3 (27)	476 µm (143 µm)	‐	‐

Abbreviations: AFL, ablative fractional laser; SD, standard deviation.

Figure 3

Overlay of UV (grayscale) and ATTO‐647N fluorescence (red pseudocolor) microscopy images of sections of intact and laser‐processed nail tissue. Healthy nail shows modest diffusion of ATTO‐647N seen primarily in the top layer (A), with a visible overall increase of signal intensity and wider distribution after fractional ablative laser treatment (B). In fungal nail tissue, ATTO‐647N diffuses more easily into deeper layers (C) with increased signal intensity in the top layer after laser processing (D)

ATTO‐647N delivery

AFL processing led to enhanced uptake and distribution of ATTO‐647N, with significantly higher signal intensity in laser‐treated healthy (mean 308.2 AU [SD 248.8 AU] and fungal nail samples (mean 428.1 AU [SD 421.8 AU] (p < 0.0001) compared to their untreated controls (healthy: mean 146.3 AU [SD 68.6 AU], fungal: mean 305.4 [SD 307.5 AU]) (Table 3). However, unprocessed fungal samples also exhibited stronger significantly ATTO‐647N fluorescence (p < 0.0001) than healthy controls (Table 4). There was no significant difference in signal intensity between AFL‐processed healthy and unprocessed fungal nail tissue (p = 0.09), resulting in similar increase in ATTO‐647N fluorescence compared with unprocessed healthy samples (+108% vs. 106%, respectively). While AFL processing resulted in improved penetration of ATTO‐647N in mycotic samples (top layer +69%, middle layer +16%, bottom layer +13%), the effects were more pronounced in healthy nail tissue (top layer +122%, middle layer +100%, bottom layer +102%).

Table 3

Overview of the average nail plate ATTO‐647N fluorescence signal intensity (values in arbitrary units)

Sample	Number of samples (sections)	Mean	SD	Median	MAD	Minimum	Maximum	Range	IQR
ATTO‐647N‐labeled fungal nail	6 (54)	305.4	307.5	167.7	92.0	102.1	1720.4	1618.3	228.0
ATTO‐647N‐labeled healthy nail	6 (54)	146.3	68.6	117.1	20.4	100.8	502.2	401.5	48.2
ATTO‐647N‐labeled fungal nail with AFL	6 (54)	428.1	421.8	273.8	195.2	102.2	2603.8	2501.5	324.9
ATTO‐647N‐labeled healthy nail with AFL	6 (54)	308.2	248.8	202.0	128.0	102.3	1564.1	1461.8	270.7
Unlabeled healthy nail	3 (27)	102.2	0.5	102.1	0.5	101.4	103.3	1.9	0.6
Unlabeled healthy nail with AFL	3 (27)	103.2	1.0	103.0	1.0	101.7	106.6	4.9	1.3

Abbreviations: IQR, interquartile range; MAD, median absolute deviation; SD, standard deviation.

Table 4

Pairwise comparison of ATTO‐647N fluorescence signal intensities using Wilcoxon rank‐sum test with continuity correction (Hochberg p value adjustment)

Samples	ATTO‐647N‐labeled fungal nail	ATTO‐647N‐labeled healthy nail	ATTO‐647N‐labeled fungal nail with AFL	ATTO‐647N‐labeled healthy nail with AFL
ATTO‐647N‐labeled fungal nail		<0.0001	<0.0001	<0.0001
ATTO‐647N‐labeled healthy nail	<0.0001		<0.0001	<0.0001
ATTO‐647N‐labeled fungal nail with AFL	<0.0001	<0.0001		<0.0001
ATTO‐647N‐labeled healthy nail with AFL	0.09	<0.0001	<0.0001

Delivery of the model drug followed a consistent pattern of marked accumulation on the dorsal surface and predominantly lateral diffusion in the proximity of ablation craters. Figure 4 shows ATTO‐647N fluorescence intensity to increase in both healthy and fungal samples after AFL pretreatment, with the most pronounced enhancement in the top layer (Tables 5 and 6). There was a moderate, statistically significant correlation between degree of poration and fluorescence intensities in healthy samples (dorsal layer: r = 0.34 p = 0.0116, intermediate layer: r = 0.36 p = 0.0075, ventral layer: r = 0.42, p = 0.0014). In fungal tissue, however, correlations were not statically significant.

Figure 4

Faceted boxplot showing ATTO‐647N fluorescence intensities measured in the dorsal, intermediate, and ventral portion of healthy and fungal nail samples with and without ablative fractional laser pretreatment

Table 5

Mean ATTO‐647N fluorescence signal intensities (values in arbitrary units) for each nail plate layer.

Nail plate layer	Unlabeled healthy nail	Unlabeled healthy nail with AFL	ATTO‐647N‐labeled healthy nail	ATTO‐647N‐labeled healthy nail with AFL	ATTO‐647N‐labeled fungal nail	ATTO‐647N‐labeled fungal nail with AFL
Top layer	102.2	103.2	198.7	442.0	448.3	758.9
Middle layer	102.2	103.3	126.2	252.0	254.8	294.9
Bottom layer	102.1	103.0	114.0	230.6	216.4	243.5

Table 6

Percentage change of ATTO‐647N fluorescence signal intensity in healthy and fungal ex vivo nail tissue

Nail plate layer	Increase in healthy nail after AFL	Increase in fungal nail after AFL	Increase in signal due to infection (healthy vs. fungal)
Top layer	+122%	+69%	+126%
Middle layer	+100%	+16%	+102%
Bottom layer	+102%	+13%	+90%
Average	+108%	+33%	+106%

DISCUSSION

By visualizing the microscopic effects of ablative fractional CO2 laser processing of nail tissue and quantitating ATTO‐647N fluorescence intensity in healthy and fungal nail tissue, our study demonstrated that AFL‐induced nail plate barrier disruptions can enhance the topical delivery of a small molecule (<900 g/mol). The results of this preclinical investigation support the growing body of evidence on the clinical application of AFL pretreatment for enhanced efficacy of topical treatment of OM as summarized in Table 1.

Laser–tissue effects in healthy nail plates have previously been investigated using electron microscopy to characterize the ablation crater. , Our SEM assessment of the effects of a CO2 laser, the most commonly employed device in LAUD studies, was consistent with previous reports of microscopic openings following AFL processing, describing the efficient poration of the nail plate and creation of highly porous ablation crater surfaces. CARS‐M visualized a thermal coagulation zone with strong autofluorescent signal associated with protein denaturation, which was also visible on WFM, as well as a trilaminar nail plate structure characterized by its high water and lipid concentration in the intermediate layer. The increase in porosity and chemical composition of the nail plate layers may be part of the explanation for the mechanism behind LAUD.

Current evidence suggests that the typical thickening of the nail plate in OM is a compensatory reaction to its pathologically increased porosity to normalize the trans‐onychial water loss. In contrast to the tightly‐packed nail layers of healthy nails with their brick‐and‐mortar‐like cellular arrangement, WFM images of fungal nail samples displayed a distorted structure with gaps and fissures corresponding to signs of clinical nail dystrophy. Since the concept of LAUD is to physically disrupt and thereby overcome barriers to topical drug delivery, our results show that AFL‐pretreatment of healthy nail tissue leads to a larger increase in permeation than in already impaired fungal nail samples. This finding is in line with the results from a study on laser‐assisted drug delivery in atopic dermatitis, demonstrating that barrier‐deficient skin benefits from microporation though to a lesser degree than intact skin. Other types of nail disorders, such as psoriatic nail disease, may present with similar clinical features but exhibit different micromorphologic changes and lower increase in porosity which could make them better targets for LAUD. , Since previous preclinical LAUD studies used healthy instead of diseased nail tissue, their results should be interpreted with caution as pretreatment effects for delivery enhancement may be overestimated.

Despite the paucity of preclinical studies of LAUD, several clinical trials have investigated the technique's potential for the treatment of OM and other nail disorders. Various AFL treatment regimens and parameters have been successfully combined with different formulations of low‐molecular‐weight compounds of <500 g/mol that can (Table 1). With a MW of 746 g/mol, ATTO‐647N is more than twice as large as conventionally used antifungal agents (MW of terbinafine: 291.4 g/mol) and exceeds the ‘500 Dalton rule’ for optimal cutaneous penetration, suggesting that low MW may not be critical to LAUD. An earlier study made by Vanstone et al. on AFL‐enhanced delivery of topically applied aqueous caffeine solution (MW = 194.2 g/mol) found the degree of poration to be an important factor in the delivery enhancement. In line with their results, our study found a correlation between the degree of poration and fluorescence signal intensity. However, this association was more pronounced in healthy tissue, suggesting that lower pulse energy may be sufficient to enhance drug distribution in diseased nail plates. Previous imaging investigations of the effects of AFL on nail tissue indicate that optical coherence tomography (OCT) is a reliable noninvasive method to detect variation in micropore dimensions and nail plate thickness. , By incorporating OCT, LAUD for nail disease could be optimized and adjusted to the clinical presentation by tailoring the pulse energy to achieve sufficient poration for optimal drug delivery enhancement depending on the extent of the fungal infection.

Limitations of this study are the lack of information on the specific fungal subtyping and the substantial variation in nail plate poration. Given the ex vivo design, our results cannot be used to draw clinical conclusions on the efficacy and safety of LAUD in patients with OM. While determining the fungal pathogen is critical for efficacious treatment, microstructural alterations of the nail plate and their impact on the barrier function are common for all types of fungal infections. Inconsistency in nail plate poration despite standardized device settings has previously been reported and is thought to be due to non‐perpendicular laser beam impact. While clippings can be trimmed to flatten the curvature of the nail, variation in microporation is to be expected when implementing LAUD in a clinical setting. Flattening the nail plate using a CO2‐laser transmissive optical window made of zinc selenide could present a convenient alternative that merits further investigation.

The current treatment paradigm for OM rests on pharmacological management with either topically or orally administered antifungal agents. While laser therapy of OM has only been cleared by the US Food and Drug Administration (FDA) for “temporary increase in clear nail,” in‐office laser treatments may prove beneficial for patients where medication adherence or systemic interactions are a concern. Given the low cure rate of approved treatment options (Table S1), off‐label treatment using LAUD of topical agents may represent an alternative for patients not tolerating or not responding to conventional approaches. Despite already promising meta‐analytic evidence on LAUD in clinical practice, identification of predictive factors for treatment response, such as dystrophic changes of the nail plate, is warranted.

In this study, we demonstrated that AFL‐processing causes micromorphologic changes that may explain the increased delivery of an aqueous solution after four hours of incubation in both healthy and damaged nail plates, supporting the results of clinical trials on LAUD for the treatment of nail disease. Nail dystrophy and its impact on the barrier function of the nail appear to be critical factors in determining the permeability and aptness of nail tissue for LAUD. More studies are needed to assess the kinetic relationship of LAUD in diseased nails, the benefit of repeated topical application of the compound, and the impact of vehicle formulations to evaluate their compatibility with this delivery technique.

CONCLUSION

Microporation of nail tissue using a fractional CO2 laser is a simple yet effective way of enhancing the delivery of a topically applied solution. Although dystrophic changes of the fungal nail plate structure lead to increased permeability, both structurally deficient and intact nails appear to benefit from laser pretreatment for ungual delivery.

AUTHOR CONTRIBUTIONS

Study conception and design: Vinzent Kevin Ortner, Jonathan R. Brewer, Merete Haedersdal, and Peter Alshede Philipsen. Acquisition of data: Vinzent Kevin Ortner, Nhi Nguyen, and Vita Solovyeva. Analysis and interpretation of data: Vinzent Kevin Ortner, Nhi Nguyen, and Peter Alshede Philipsen. Drafting of manuscript: Vinzent Kevin Ortner, Nhi Nguyen. Critical revision: Merete Haedersdal, Jonathan R. Brewer, Vita Solovyeva, and Peter Alshede Philipsen.

CONFLICTS OF INTEREST

The authors declare no conflicts of interest.

Supporting information.

Click here for additional data file.