Administration of pilocarpine by microneedle patch as a novel method for cystic fibrosis sweat testing

Abstract The sweat test is the gold standard for the diagnosis of cystic fibrosis (CF). The test utilizes iontophoresis to administer pilocarpine to the skin to induce sweating for measurement of chloride concentration in sweat. However, the sweat test procedure needs to be conducted in an accredited lab with dedicated instrumentation, and it can lead to inadequate sweat samples being collected in newborn babies and young children due to variable sweat production with pilocarpine iontophoresis. We tested the feasibility of using microneedle (MN) patches as an alternative to iontophoresis to administer pilocarpine to induce sweating. Pilocarpine‐loaded MN patches were developed. Both MN patches and iontophoresis were applied on horses to induce sweating. The sweat was collected to compare the sweat volume and chloride concentration. The patches contained an array of 100 MNs measuring 600 μm long that were made of water‐soluble materials encapsulating pilocarpine nitrate. When manually pressed to the skin, the MN patches delivered >0.5 mg/cm2 pilocarpine, which was double that administered by iontophoresis. When administered to horses, MN patches generated the same volume of sweat when normalized to drug dose and more sweat when normalized to skin area compared to iontophoresis using a commercial device. Moreover, both MN patches and iontophoresis generated sweat with comparable chloride concentration. These results suggest that administration of pilocarpine by MN patches may provide a simpler and more‐accessible alternative to iontophoresis for performing a sweat test for the diagnosis of CF.

activity of an epithelial cell surface protein called Cystic Fibrosis Transmembrane Conductance Regulator that acts as an ion channel for the transport of chloride and bicarbonate. 2 Early diagnosis of CF has been made possible by universal screening for CF based on elevated serum concentration of immunoreactive trypsinogen. All newborns with a positive newborn screen for CF have been referred to undergo sweat testing in the United States since 2010, and early detection and treatment of CF has improved the long-term growth and respiratory outcomes of those affected. 3 The sweat test procedure in newborn infants involves collection of sweat from both forearms after stimulation of the sweat glands with pilocarpine nitrate (a cholinergic agonist) delivered into the skin via iontophoresis. 4 The delivery of pilocarpine into skin is intended to ensure a consistent rate of sweating and provide a small volume of sweat (generally 15-100 μl). In many instances, inadequate volumes of sweat are collected, necessitating repeat testing. This failure of adequate sweat collection is especially common when the sweat test is performed on infants less than 3 months of age, and may be related to several factors such as gestational age, 5 ethnicity, 6 and skin factors 7 that can result in inadequate sweat collection. This can lead to delays in diagnosis and treatment, and also causes significant anxiety 8 for the parents of the newborn who are waiting to find out if their child has CF. 9 The current technique for sweat test has remained unchanged since it was first standardized in the 1960s. 10 To improve the yield of sweat in newborn infants during the sweat test and to reduce diagnostic uncertainty, there is an urgent need to develop more accessible and simple-to-administer alternatives for inducing and collecting sweat. Such methodology will facilitate expedient and accurate diagnosis of CF in infants. The Cystic Fibrosis Foundation recommends that the rate of inadequate collection should be less than 10% of all sweat tests performed in children less than 3 months of age, and it should be less than 5% of all sweat tests done in children older than 3 months of age and adults. However, many clinical labs tend to struggle with meeting those targets, especially for the infants less than 3 months of age.
As an alternative to conventional iontophoresis, we propose pilocarpine administration by using a microneedle (MN) patch, which has been shown to deliver a variety of different drugs into the skin. 11-14 MN patches consist of an array of solid, conical needles measuring hundreds of microns in length that can be loaded with drugs. Upon application to skin, MNs penetrate across epidermis and into the dermis, where they dissolve and release their drug payload in the skin. 15,16 This procedure is minimally invasive, not painful and generally well tolerated. 17,18 In addition, application of MN patches is much easier than hypodermic injection or iontophoresis, and it can be performed after brief training. [19][20][21] patches have been studied in many clinical trials for administration of zolmitriptan for treatment migraine, 22 delivery of parathyroid hormone for treatment of osteoporosis, 23 administering influenza vaccine 20,24 and delivery of other compounds, 13 and are widely used for delivery of cosmetic products. 25 Cost of manufacturing MN patches is expected to be less than US$1.00. 26 Topical delivery of pilocarpine by MN patches has received limited prior attention. We previously conducted a study using metal MNs as a drug-free pretreatment of the skin to create micropores that enhance the efficiency of pilocarpine delivery by iontophoresis. 27 In another study, pilocarpine-coated metal MNs were used to administer pilocarpine to the eye for cholinergic effects. 28 However, the development and use of MNs to directly administer pilocarpine into the skin at levels suitable for sweat induction has not been described before.
The objectives of this study were to develop dissolvable MN patches for intradermal administration of pilocarpine and to determine their efficacy in inducing sweating using an animal (equine) model.
The horse is one of the few animals, other than primates, that sweat and was therefore an appropriate animal model for this study (see Supporting Information for additional information). [29][30][31] To the best of our knowledge, this work is the first study to develop dissolvable MN patches that load and deliver pilocarpine to skin for the purpose of sweat testing, and to assess the sweat induction of pilocarpine-loaded MN patches in comparison with conventional iontophoresis.

| Characterization of MN patches and iontophoretic Pilogel disks
MN patches were designed to encapsulate pilocarpine within a watersoluble matrix forming a conical shape with a sharp tip to facilitate penetration into skin ( Figure 1). Each MN patch consisted of a 10 × 10 array of MNs arranged within a square with approximately 7 mm sides (i.e.,~0.5 cm 2 ). In contrast, the iontophoretic Pilogel disks were circular with a diameter of 2.72 cm (i.e.,~5.8 cm 2 ), thereby contacting an area of skin more than 10 times larger than the MN patch.
For microscopic examination as shown in Figure 2a In the current formulation, pilocarpine comprises 40% of the solids content in the MNs. The total amount of pilocarpine loaded per MN patch was measured as 500 ± 48 μg (n = 4). After application to porcine skin ex vivo, the residual pilocarpine content per MN patch was 237 ± 73 μg (n = 6), indicating that the delivered dose was 263 μg (i.e.,~526 μg/cm 2 ) and the delivery efficiency was~53%.
This incomplete delivery is likely due to the slow dissolution of PVA polymer, which comprised 40% of the solids content in the MNs.
However, the high level of PVA used in the formulation is required to provide sufficient mechanical strength to the MNs to withstand the process of insertion into skin. Future studies could further optimize formulation to make MN dissolution faster and more complete, which would increase delivery efficiency.
The iontophoretic Pilogel disks were considerably larger and therefore had much higher pilocarpine loading amount, measured  (Table 1).

| Comparison of sweat volume induced by MN patches and iontophoresis
Sweat samples were collected from 25 sites on the neck of four horses after introducing pilocarpine by either MN patches or iontophoresis. No changes to physical exam parameters, cervical skin irritation, or other adverse effects were noted in any horses during or after the study. From all application sites, at least 10 μl of sweat was collected. The average total sweat volume from an iontophoresis site was 101 ± 49 μl over a pilocarpine application area of 5.8 cm 2 , corresponding to a sweat collection density of 17 ± 8 μl/cm 2 . The average total sweat collected from an MN patch site was 17 ± 8 μl over a pilocarpine application area of 0.5 cm 2 , corresponding to a sweat collection density of 34 ± 16 μl/cm 2 (Figure 3a,c). We believe that sweat density is the most appropriate basis for comparison between the two techniques because sweat production is expected to scale directly with area and because sweat collection is usually done over a standard area of skin using a sweat collection device like the Macroduct Sweat Collector used here.
While the total amount of sweat collected from the iontophoresis sites was greater than that collected from the MN patch sites (Figure 3a), when accounting for the different pilocarpine application areas, the sweat collection density from the MN patch sites was 2.0-fold greater than that collected from the iontophoresis sites ( Figure 3c). This ratio was relatively consistent on each of the four horses (2.3-, 1.6-, 2.1-, and 1.6-fold greater). This suggests that the difference between MN patches and iontophoresis on sweat induction was not determined by the individual differences between horses. Instead, the difference in sweat collection density appears to mainly reflect the different sweat-inducing abilities of the two pilocarpine delivery procedures.
To help explain why the sweat collection densities differed between the two methods, we calculated the sweat volume per unit of pilocarpine dose delivered to the skin. This analysis revealed no significant difference between iontophoresis (73 ± 36 μl/mg) and MN patches (66 ± 34 μl/mg) (Figure 3b). However, because the amount of pilocarpine delivered per unit area of skin was 2.2-fold greater when administered by MN patches (~526 μg/cm 2 ) compared to iontophoresis (~238 μg/cm 2 ), this likely accounts for the greater sweat collection density seen after pilocarpine delivery by MN patch. This further indicates that using MN patches to deliver pilocarpine has a comparable or possibly better sweat-inducing capability as the traditional iontophoresis.
It should be noted that although sweat collection density was greater using a MN patch, the MN patch induced less total sweat volume than iontophoresis. Because the MN patch delivered twice as much pilocarpine per unit area, we expect that a larger MN patch with the same area as the pilocarpine disk used for iontophoresis (i.e., 5.8 cm 2 ) would correspondingly deliver twice as much pilocarpine and thereby induce more total sweat volume compared to iontophoresis, because sweat production is known to increase with pilocarpine dose delivered. 32,33 Additional studies will be needed to develop larger MN patches and measure the resulting sweat production.  This simple and low-cost method of pilocarpine delivery by MN patch compared to iontophoresis could make sweat testing more widely available and no longer limited to use by specialty clinics. In addition, the larger pilocarpine dose per unit area could enable MN patch delivery to more consistently generate the amount of sweat required to perform a chloride measurement, thus potentially making the sweat test more reliable and avoiding the need for repeated measurement attempts.

| Fabrication of MN patches
The pilocarpine-loaded MN patches were fabricated by a two-step molding process using polydimethylsiloxane (PDMS) molds based on an established method. 39 The first casting solution was a mixture of  and allowed to eat grass hay as desired to ensure compliance. All horses used in this study were accustomed to handling and restraint and remained calm during sampling; sedation was specifically avoided due to potential neuroendocrine effects that could impact sweat responses.

| Animals and preparation for in vivo studies
Immediately prior to all testing, the entire cervical region was clipped bilaterally with electric clippers (#40 blades). On the right cervical region, three 3 × 3 cm areas were further shaved manually with a straight razor to permit good contact of the MN patches, iontophoretic pilocarpine disks and sweat collection pads to the skin. The skin was then washed with a mild detergent soap and dried thoroughly with a cotton towel to remove skin oils and facilitate uniform adherence of sweat induction and collection devices.
Horses' vital parameters (temperature, pulse, respiratory rate), behavior, and cervical skin were monitored for any adverse effects before testing, every 10 min during the pilocarpine administration and sweat collection period and then 6, 12, and 24 h after sample collection was completed.

| Application of MN patches and iontophoresis to induce sweating in vivo
As a preliminary experiment, we first confirmed pilocarpine-induced sweat production in horses and optimized small volume sweat collec-

| Sweat chloride quantification via chloridometer
All sweat samples were batch-analyzed at the end of the study period after thawing to room temperature. Because horse sweat chloride concentration is much higher than that of human sweat, all sweat samples from horses were diluted 4-fold with distilled water prior to measurement of chloride concentration. Chloride concentration of the collected sweat samples was measured using a chloridometer (ChloroChek®, Wescor). The measurement requires a minimum volume of 10 μl of sweat and is based on the principle of coulometric titration. Following the manufacturer's instructions, the sweat sample was added to an acid buffer/working solution into which silver electrodes were immersed. The results obtained for the horse sweat samples were then converted to final values by multiplying by 4 to account for the dilution performed prior to testing.

| Statistics
All data are presented as mean ± SD.

DATA AVAILABILITY STATEMENT
The animal data included in this article has not been uploaded to any public databases or archives. There is no human subject data in this article.

DISCLOSURE
The fabrication and use of microneedles made out of pilocarpine for the induction of sweat production during sweat testing has been submitted as a patent to the United States Patent and Trademark Office (Patent Pending).