Development and validation of a bioanalytical method to quantify povorcitinib in human skin with clinical application

ABSTRACT

Aim

Quantification of drug concentrations in human skin is essential for understanding mechanisms of action in dermatologic therapeutics but remains analytically challenging. The study aimed to develop and validate a standardized tissue homogenization and liquid chromatography – tandem mass spectrometry workflow for quantifying povorcitinib in human skin biopsies.

Materials & methods

Key skin homogenization parameters, including solvent composition, processing volume, and homogenate density, were systematically evaluated. Wet and dry homogenization approaches were compared. Following protein precipitation, samples were analyzed using liquid chromatography – tandem mass spectrometry.

Results

Eighteen solvent systems and various processing volumes were evaluated, with three solvent mixtures producing homogeneous preparations. Limiting homogenization volumes to less than half tube capacity and using 20 mg/mL improved sample consistency. Povorcitinib stability under acidic and thermal conditions was confirmed. Comparable performance was observed between wet and dry homogenization methods, and a dry workflow was validated.

Conclusions

This study provides the first systemic evaluation of human skin homogenization parameters and a direct comparison of wet and dry approaches. The validated dry homogenization method enabled accurate and reproducible quantification of povorcitinib in skin biopsies from a phase I clinical study and offers a robust framework for quantifying a broad range of analytes in human skin tissues.

Clinical trial registration: https://www.clinicaltrials.gov/study/NCT06505265 [clinicaltrials.gov] identifier is NCT06505265.

Plain Language Summary

Knowing how much a medicine taken by mouth reaches the skin helps scientists understand how skin treatments work. However, measuring this can be difficult. In this study, we figured out a dependable way to measure the amount of a medicine called povorcitinib in very small skin samples collected from people. We tested different ways to break down skin samples to make sure the medicine could be measured reliably. This included trying different types and amounts of liquid and comparing two ways of crushing the skin. In the wet method, skin samples were placed in liquid right away. In the dry method, the skin was first frozen and then crushed into very tiny pieces before liquid was added. We also checked that povorcitinib did not change during these steps. We found that using several types of liquids could produce consistent skin samples and that carefully controlling how much liquid to use helped with sample consistency. Both the wet and dry methods worked equally well, and the dry method was tested further to make sure it worked every time. Overall, this study provides a reliable way to measure how much povorcitinib reaches the skin in people taking part in a research study. This same approach can also be used to measure other medicines in skin samples and help scientists develop better treatments for skin diseases in the future.

1. Introduction

Povorcitinib (INCB054707), an orally available small-molecule inhibitor of the Janus kinase (JAK) family of protein tyrosine kinase with selectivity for JAK1, is currently being developed for the treatment of immune-mediated dermatologic disorders, such as hidradenitis suppurativa [1,2] and vitiligo [3]. Because the skin is both the primary site of disease pathology and a key therapeutic target for JAK1 inhibition, understanding whether and to what extent orally administered povorcitinib reaches skin tissue is essential for understanding its relationship to pharmacodynamic activity and clinical response, as well as for characterizing its pharmocokinectics. Direct quantification of povorcitinib concentrations in human skin provides critical insight into tissue exposure at the site of action, enabling evaluation of mechanistic hypotheses related to JAK1 inhibition in cutaneous immune pathways, assessment of the relationship between plasma and tissue concentrations, and support appropriate dose selection for dermatologic indications. Therefore, the development of a sensitive and robust bioanalytical method is a critical component of understanding the therapeutic functionality and mechanism of action of povorcitinib in skin.

Effective and robust quantification of small molecules in human skin presents unique challenges because of low expected concentrations, small sample volumes, and the complex nature of the skin structure [4,5]. Unlike liquid matrices such as plasma or urine, human skin is a heterogeneous and relatively tough soft tissue due to its multilayered fibrous and keratinized structure [6], dense extracellular matrix [7], and diverse lipid and protein content [8]. Homogenization is a crucial step for reliable and reproducible measurement of drug levels in human skin [4]. Effective homogenization ensures efficient and consistent extraction of analytes from skin tissue. Wet homogenization, typically involving mechanical or bead-based disruption in a solvent system, offers operational simplicity but poses the risks of analyte dilution and potential chemical degradation. Dry homogenization, often employing cryogenic grinding, minimizes these risks and preserves analyte integrity, but requires specialized equipment and handling precautions. Although there are works related to human skin bioanalysis [9–13], direct comparisons of the human skin homogenates generated from these approaches, particularly in the context of bioanalytical method performance for small molecules, are sparse in the literature. In addition, despite the growing scientific and clinical interest in measuring drug concentrations in skin for orally administered drugs targeting dermatologic diseases, rigorously validated bioanalytical methods that comply with current regulatory standards remain limited for accurate quantification of small-molecule drugs in this complex and heterogeneous tissue [4,14].

The present study addresses this unmet need by developing and validating a sensitive and robust liquid chromatography – tandem mass spectrometry (LC-MS/MS) method for the quantification of povorcitinib in human skin biopsies and demonstrating its application to clinical samples, thereby enabling reliable assessment of skin drug exposure in a clinical context. We systematically evaluated key parameters affecting method performance, including homogenization solvents, processing volumes and homogenate density, and analyte stability under thermal and acidic conditions. We also conducted a comparative evaluation of wet and dry homogenization, validated a dry homogenization procedure tailored for the efficient and reliable processing and quantification of povorcitinib in human skin tissues. Finally, we applied the validated method to a phase I clinical study to quantify povorcitinib concentrations in skin tissue, enabling the evaluation of local pharmacokinetics and exploration of the potential correlation between plasma and skin concentrations

2. Materials & methods

2.1. Chemicals, reagents, and matrix

Certified reference material for povorcitinib and internal standard INCB170045 (d6-povorcitinib) were from Incyte Corporation (Wilmington, DE, USA). LC-MS – grade acetonitrile, water, methanol, and formic acid were purchased from Thermo Fisher Scientific (Waltham, MA, USA). LC-MS – grade ammonium acetate was purchased from MilliporeSigma (Burlington, MA, USA). High-performance LC-grade acetonitrile, methanol, and water were purchased from MilliporeSigma. A large piece of full-thickness human skin tissue was obtained from BioIVT (Westbury, NY, USA) and stored at −70°C. This tissue served as the source material from which 4-mm skin punches were collected and subsequently homogenized for analysis.

2.2. Equipment and preparation of human skin biopsy samples and homogenates for method development

Bead Ruptor Elite Bead Mill Homogenizer from OMNI International (Kennesaw, GA, USA) was used for wet homogenization and Geno/Grinder^Ⓡ 2010 from Cole-Parmer (Vernon Hills, IL, USA) was used for dry homogenization of human skin biopsies. A 4-mm disposable biopsy punch with plunger was purchased from Integra LifeScience (Princeton, NJ, USA). The 2-mL tubes pre-filled with 2.8-mm ceramic beads were purchased from Thermo Fisher Scientific, and the 7-mL tubes pre-filled with 2.8-mm ceramic beads were obtained from OMNI International. The 3/8-inch stainless steel grinding balls were obtained from Cole-Parmer.

Skin biopsies were prepared from a frozen full-thickness human skin tissue using a 4-mm disposable biopsy punch with plunger, transferred into pre-weighted 2-mL Eppendorf tubes kept on dry ice, capped, and stored at −70°C for future use.

For wet homogenization, the frozen skin biopsies were weighed and placed into 2-mL or 7-mL bead mill tubes pre-filled with ceramic beads. Homogenization solvent was then added to achieve various final concentrations of skin homogenate. Homogenization was performed in a room at 4°C using a Bead Ruptor. For 2-mL tubes, the instrument was operated at 6.8 m/s for 30 seconds, with one round of two cycles and a 90-second dwell time. For 7-mL tubes, homogenization was carried out at 5.8 m/s for 45 seconds, with one round of three cycles and a 90-second dwell time. Following homogenization, samples were cooled on wet ice for ~5 min. Homogenates were either pooled and stored at −70°C for later use or used fresh as blank human skin homogenate matrix for sample preparation.

For dry homogenization, the frozen skin biopsies were placed into pre-weighed Spex^Ⓡ vials, reweighed, and a 3/8-inch stainless steel grinding ball was added to each vial. The vials were loaded onto a cryo-block and snap-frozen in liquid nitrogen for 5 minutes. The cryo-block was then transferred to a Geno/Grinder and run at 1500 strokes/min for 90 seconds. The resulting tissue powder was suspended in pre-chilled homogenization solvent while kept on ice. The vials were then shaken at high speed on a reciprocal shaker for 10 minutes. The resulting skin homogenates were transferred to clean tubes and stored at −70°C until use.

It is worth noting that the terms “wet” and “dry” refer to the homogenization strategy rather than the hydration state or water content of the skin samples. Wet homogenization denotes the direct homogenization of skin tissue in the presence of extraction solvent, allowing simultaneous tissue disruption and analyte extraction. In contrast, dry homogenization refers to mechanical grinding of skin tissue into a fine powder in the absence of solvent, followed by subsequent addition of extraction solvent.

2.3. Preparation of stocks, calibration standards, and quality control samples

Stock solutions of povorcitinib and internal standard INCB170045 were prepared separately in acetonitrile:water (50:50, volume/volume [v/v]) to obtain a concentration of 5 mM. Working solutions of povorcitinib were prepared by diluting appropriate amounts of stock or higher working solutions with acetonitrile:water (50:50, v/v). Duplicates of the povorcitinib stock solutions were prepared and verified to be equivalent (%difference calculated as: instrument response 1-instrument response 2/mean response x100 ≤ 5%) before use. This step aligns with ICH M10 guidance, which recommends confirming stock solution accuracy to enable the use of the same stock solution for preparing calibration standards and QC samples. An internal standard working solution (25 nM INCB170045) was prepared by first diluting the 5 mM internal standard stock to a 50 μM intermediate solution and then diluting the intermediate solution with 100% acetonitrile. The stock solutions and internal standard working solutions were stored at 4°C for further use. For the assessment of stock solution stability, aliquots of the 5 mM povorcitinib stock solution in acetonitrile:water (50:50, v/v) were stored at approximately 4°C for up to 740 days. Prior to LC – MS/MS analysis, two freshly prepared stock solutions and one stored stock solution were each diluted with acetonitrile:water (50:50, v/v), and an appropriate amount of internal standard working solution was added. Each diluted solution was analyzed in replicate (n = 6). The mean peak-area ratio of the stored solution was then compared with those of the two freshly prepared solutions, which were required to demonstrate equivalency (i.e., differences within ±5%). Stock solution stability was considered acceptable if the difference between the stored solution and the mean fresh solutions was within ±10%.

Quantitative analysis of povorcitinib was performed using a calibration-curve – based LC-MS/MS method incorporating a stable-isotope labeled internal standard. The isotope-labeled internal standard was added to all calibration standards, quality control samples, and study samples prior to extraction. Quantification was based on the analyte-to-internal-standard peak area ratio, which corrects for variability in extraction recovery, matrix effects, and instrument response. A multi-point calibration curve was generated in matrix and used to determine sample concentrations, consistent with current bioanalytical best practices. This approach aligns with the quantitative framework described in the ICH M10 guideline for bioanalytical method validation and provides accurate, precise, and robust quantification for complex biological matrices such as human skin. Specifically, calibration standards for povorcitinib were prepared in human skin homogenate by spiking appropriate working solutions to obtain concentrations of 1, 2, 10, 50, 200, 1000, 1800, and 2000 nM. Quality control (QC) povorcitinib solutions at lower limit of quantification (LLOQ), low, middle, and high concentrations of 1, 3, 800, and 1600 nM, respectively, were prepared in the same manner as the calibration standards. Calibration standards and QCs were freshly prepared on the day of analysis; previously frozen samples could be used if sufficient stability had been established, which is fully aligned with the ICH M10 guidance and provides flexibility to use frozen standards and QCs that remain within their defined stability period.

2.4. Sample preparation

Human skin homogenate was used as the blank matrix for preparation of calibration standards and QC samples and was stored at −70°C until use. For povorcitinib sample extraction during method development, 50 μL of skin homogenate was transferred into tubes arranged in a 96-well rack. Proteins were precipitated by adding 150 μL of acetonitrile containing 25 nM INCB170045 internal standard. The capped plate was vortex-mixed and centrifuged, after which 150 μL of the supernatant was transferred to a clean 96-well collection plate and evaporated to dryness under nitrogen. Each sample was reconstituted with 150 μL of methanol/water/formic acid (50:50:1, volume/volume/volume [v/v/v]), vortexed for ~1 minute, and placed in the autosampler tray for LC-MS/MS analysis. For povorcitinib sample extraction during method validation, 25 μL of skin homogenate was transferred into tubes arranged in a 96-well rack followed by the addition of 25 μL of internal standard working solution (250 nM). Proteins were precipitated by adding 450 μL of acetonitrile:formic acid (99:1, v/v). The plate was capped, vortex-mixed, and centrifuged. Then 100 μL of the supernatant was transferred to a clean 96-well collection plate and evaporated to dryness under nitrogen. Each sample was reconstituted with 200 μL of acetonitrile:water:formic acid at 35:65:0.1 (v/v/v), vortexed for ~2 minutes, and placed in the autosampler tray for LC-MS/MS analysis.

2.5. Assessment of acid and heat stability of povorcitinib

To specifically assess acid and heat stability under homogenization conditions, povorcitinib was spiked into the acetonitrile:water:formic acid (50:50:5, v/v/v) homogenization solvent at a low QC concentration of 3 nM in the presence of human skin tissue and subjected to bead homogenization. The peak area ratio was compared with that of freshly prepared povorcitinib in acetonitrile:water (50:50, v/v) neat solution. The difference between the homogenized sample and the neat solution must be within ±10%.

2.6. LC-MS/MS instrumentation and conditions

LC-MS/MS analysis was performed using an AB SCIEX Triple Quad 6500+ or an AB SCIEX QTRAP 6500 MS (Applied Biosystems; Waltham, MA, USA) coupled with Shimadzu Nexera X2 binary UHPLC pumps, a degasser, an autosampler and a column oven (Shimadzu; Columbia, MD, USA). During method development, chromatographic separation of povorcitinib was achieved on an XSelect HSS T3 column (3.5 µm, 50 × 2.1 mm; Waters Corp, Milford, MA, USA) under isocratic conditions with 47% mobile phase B at a flow rate of 0.5 mL/min. The mobile phases consisted of 2 mM ammonium acetate (A) and 100% acetonitrile (B). For method validation, chromatographic separation for povorcitinib was achieved on an Acquity UPLC CSH flouro-phenyl column (1.7 µm, 50 × 2.1 mm; Waters Corp) using an liquid chromatography gradient at a flow rate of 0.5 mL/min. The mobile phases were water with 0.1% formic acid (A) and 100% acetonitrile (B). The liquid chromatography gradient was as follows: 0–0.3 min, 35%B; 0.3–1.3 min, 35%–75%B; 1.3–1.4 min, 75%–95%B; 1.4–1.6 min, 95%B; 1.6–1.8 min, 95%–35%B; 2.2 min, stop. The column temperature was maintained at 40°C and the autosampler was set to 15°C during the LC-MS/MS analysis. It is worth noting that two chromatographic columns were used because method development and method validation were conducted in different laboratories. Internal method development employed the column historically applied in our validated LC – MS methods for povorcitinib, while the external contract research organization performing method validation selected the column from their previously validated povorcitinib human plasma assay, from which the LC – MS conditions were adapted for the skin analysis. As the LC columns are not critical to assay performance when fully validated, the use of two columns does not impact the reliability or suitability of the method.

The mass spectrometer was operated in positive electrospray ionization mode, and the resolution setting used was “unit” for both the first quadrupole and the third quadrupole. The multiple reaction monitoring transition was m/z 508.1→395.0 for povorcitinib and m/z 514.1→401.1 for the internal standard INCB170045. The optimized parameters of the mass spectrometer were curtain gas, 40 psig; ion spray voltage, 5500 V; source temperature, 600°C; nebulizer gas, 60 psig; turbo gas, 60 psig; and collision-activated dissociation gas, 9 psig. The optimized collision energy for both povorcitinib and INCB170045 was 29 V.

2.7. Method validation, calibration & data evaluation

Method validation parameters included selectivity, accuracy, precision, carryover, matrix effect, recovery, dilution integrity, and stability.

2.7.1. Selectivity

The selectivity of the method was evaluated by extracting and analyzing six individual lots of blank human skin homogenate without povorcitinib and the internal standard. For selectivity to be considered acceptable, none of the six individual lots of blank human skin homogenate could show an interference peak area at the retention time of the peak for povorcitinib that was > 20.0% of the mean analyte peak area of the lowest calibration standards and none of the six individual lots could show an interference peak area at the retention time of the internal standard that was > 5.0% of the mean internal standard peak area in the lowest calibration standards.

2.7.2. Accuracy and precision

Accuracy was determined by replicate analysis of samples containing known amounts of povorcitinib and expressed as percent bias (%bias). The acceptable maximum %bias for a valid assay was ±20.0% of the nominal concentration for the LLOQ QC sample and ±15.0% for all other concentrations. Precision was determined by the coefficient of variation (%CV). The acceptable maximum %CV for a valid assay was 20.0% at the LLOQ QC sample and 15.0% for all other QC concentrations.

2.7.3. Matrix effect

The matrix effect was determined in six individual lots of human skin homogenate at two concentrations (3 and 1600 nM, n = 3) for povorcitinib. Matrix effect was evaluated by analyzing three replicates of low and high QCs, each prepared using matrix from six individual lots. For each individual matrix lot evaluated, the accuracy should be within ±15% of the nominal concentration and the precision (%CV) should not be greater than 15%.

2.7.4. Recovery

Recovery was assessed at three concentration levels (3, 800, and 1600 nM, n = 5) for povorcitinib and at one concentration (250 nM, n = 5) for the internal standard. Recovery of both the analyte and internal standard was evaluated by comparing the mean peak areas of samples spiked prior to extraction (extracted sample) with those of samples spiked after extraction (post-extraction spiked samples). Percent recovery was calculated as mean peak areas of extracted samples/mean peak areas of post-extraction spiked samples x100. For the analyte, the variation in recovery (%CV) across the three concentrations levels was required to be ≤ 20.0%. For the internal standard, the %CV across the analyte concentration range in which it was evaluated must not exceed 20.0%.

2.7.5. Dilution integrity

Dilution integrity was assessed by preparing a human skin homogenate at a concentration of 4000 nM and diluted with pooled blank human skin homogenate at a dilution factor of 20 in five replicates. The mean of the determined concentrations of the diluted samples must be within 100 ± 15.0% of the nominal value before dilution, and the %CV of the determined concentrations of the diluted samples did not exceed 15.0%.

2.7.6. Stability

Analyte stability, including benchtop, freeze-thaw, and long-term storage stability was assessed by using QC samples prepared in human skin homogenate at concentrations of 3 and 1600 nM. The preparation accuracy of these stability samples was verified in six replicates. QC samples were aliquoted and independently subjected to each stability condition. For each condition, three individual aliquots were analyzed, with two replicate measurements obtained from each aliquot at each concentration, temperature, and time point. To meet acceptance criteria, the mean measured concentration at each condition was required to be within 100 ± 15.0% of the nominal value, and the variability (%CV) of replicate measurements was required to be ≤ 15.0%.

2.7.7. Carryover

Carryover was assessed by injecting an extracted blank immediately following the highest calibration standard in both the first and second calibration sets. The peak areas of the analyte and internal standard in these carryover blanks were required to be ≤ 20.0% and ≤5.0%, respectively, of the corresponding peak areas measured in the first and second replicates of the LLOQ.

2.7.8. Calibration and data evaluation

Concentrations were calculated using eight concentration levels for calibration curves, ranging from 1 to 2000 nM, with 1/x² weighted linear regression for povorcitinib. Peak area integrations and regressions were performed using Analyst software (v1.7.2; Applied Biosystems). Excel (Microsoft Corp., Redmond, WA, USA) was used for calculation.

2.8. Skin sample collection & analysis

Clinical study INCB054707-112 was an open-label skin and plasma pharmacokinetic study evaluating multiple oral doses of povorcitinib in healthy participants. The primary endpoints were dermal povorcitinib concentrations at steady state and pharmacokinetics for plasma povorcitinib. The study was conducted in accordance with the International Conference on Harmonization (ICH) Guidelines for Good Clinical Practice, including the archiving of essential documents, the principles of the Declaration of Helsinki and other applicable local ethical and legal requirements. A total of 18 participants were enrolled in this study. Informed consent was obtained from each participant before protocol-specific screening assessments were performed. The Advarra institutional review board approval was obtained for this study (reference number Pro00080431). For skin collection, a maximum of four 4-mm biopsies were obtained from the lower back from each participant at various timepoints using a standard skin biopsy collection technique according to the study protocol. Fat was trimmed if needed. The skin biopsy samples were placed in a storage tube in an ice bath or crushed ice before being frozen. Within one hour of the skin biopsy sample collection, the sample was stored in a − 80°C freezer until shipment. Overall, 72 skin biopsies from 18 participants were received and analyzed.

3. Results and discussion

3.1. Evaluation of human skin homogenization solvents

Human skin is a heterogeneous and relatively tough soft tissue, making homogenization both a critical and challenging step for accurate and reproducible quantification of small molecules [4,5]. A range of solvent systems comprising methanol or acetonitrile, with or without water and varying concentrations of formic acid, were evaluated for initial bead homogenization of human skin tissue (Table 1). The 18 solvent compositions tested included pure organic solvents (methanol or acetonitrile), binary mixtures (eg, methanol:water and acetonitrile:water at 90:10 or 50:50, v/v), and ternary mixtures containing formic acid at 1% or 5% (v/v/v). Among these, only three conditions yielded homogeneous skin homogenates: methanol:water:formic acid (90:10:5, v/v/v), methanol:water:formic acid (50:50:5, v/v/v), and acetonitrile:water:formic acid (50:50:5, v/v/v). The remaining 15 solvents failed to generate uniform homogenates under our laboratory conditions, as illustrated by representative examples (Figure 1). Interestingly, methanol:water:formic acid (90:10:5, v/v/v) produced a homogeneous human skin homogenate, whereas acetonitrile:water:formic acid (90:10:5, v/v/v) did not, leaving visible tissue fragments. Increasing the water content improved performance, as acetonitrile:water:formic acid (50:50:5, v/v/v) yielded uniform homogenates comparable to methanol:water:formic acid (50:50:5, v/v/v), highlighting the importance of the aqueous component in effective homogenization. Furthermore, acid concentration was found to be critical because both methanol:water (50:50, v/v) and methanol:water:formic acid (50:50:1, v/v/v) failed to produce uniform homogenates (Table 1). The homogenization differences can be attributed to the combined effects of solvent polarity, water content, and acid concentration on tissue breakdown. Pure organic solvents lack sufficient aqueous content to hydrate and soften the extracellular matrix, and therefore limit mechanical disruption during bead beating. Introducing water improves tissue pliability and enhances bead – tissue interaction, which explains why homogenization efficiency increased when the water content was raised from 10% to 50%. In addition, higher formic acid concentrations facilitated protein denaturation and matrix loosening, thereby promoting more complete tissue breakdown; this explains the poor performance of binary mixtures and ternary mixtures containing only 1% formic acid. Based on these observations, acetonitrile:water:formic acid (50:50:5, v/v/v) was selected as the homogenization solvent in this work.

Table 1.

Summary of solvents evaluated for homogenization of human skin biopsy samples.

Homogenization solvents evaluated	Homogeneous homogenate obtained, Yes/No
Methanol 100%	No
Acetonitrile 100%	No
Methanol: Formic Acid, 100:1, v/v	No
Acetonitrile: Formic Acid, 100:1, v/v	No
Methanol: Formic Acid, 100:5, v/v	No
Acetonitrile: Formic Acid, 100:5, v/v	No
Methanol:Water, 90:10, v/v	No
Acetonitrile:Water, 90:10, v/v	No
Methanol:Water:Formic Acid, 90:10:1, v/v/v	No
Acetonitrile:Water:Formic Acid, 90:10:1, v/v/v	No
Methanol:Water:Formic Acid, 90:10:5, v/v/v	Yes
Acetonitrile:Water:Formic Acid, 90:10:5, v/v/v	No
Methanol:Water, 50:50, v/v/v	No
Acetonitrile:Water, 50:50, v/v/v	No
Methanol:Water:Formic Acid, 50:50:1, v/v/v	No
Acetonitrile:Water:Formic Acid, 50:50:1, v/v/v	No
Methanol:Water:Formic Acid, 50:50:5, v/v/v	Yes
Acetonitrile:Water:Formic Acid, 50:50:5, v/v/v	Yes

Abbreviations: v/v, volume:volume; v/v/v, volume:volume:volume.

Figure 1.

Representative examples of human skin homogenate. Representative example of inhomogeneous (left) and homogeneous (right) human skin homogenate obtained with homogenization solvents shown in Table 1.

3.2. Evaluation of homogenization solvent volumes on skin homogeneity

The volume of homogenization solvent can influence both the homogeneity and consistency of human skin homogenates. To evaluate this effect, bead-based homogenization was assessed in 2-mL and 7-mL tubes at different volumes with a fixed density of 40 mg/mL. The two tube sizes served different purposes: 2-mL tubes were used for clinical study sample homogenization, whereas 7-mL tubes were used to generate sufficient amount of blank human skin matrix for the preparation of calibration standards and QCs.

Across both tube sizes, we observed that filling the tubes with solvent volumes greater than approximately half of their total capacity (i.e., > 1.0 mL in a 2-mL tube and > 3.5 mL in a 7-mL tube) resulted in suboptimal homogenization, with visible tissue fragments remaining (Figure 2). This reduced homogenization efficiency is likely due to diminished shear force in higher fill volumes. For 2-mL tubes to deliver consistent homogenization for clinical study samples, an optimal solvent volume range of 0.5–1.0 mL was identified. This range provided sufficient fluidity for efficient bead motion while maintaining effective tissue disruption. For 7-mL tubes, however, the goal was to maximize the total volume of blank matrix rather than to homogenize individual study samples. Therefore, a solvent volume close to 3.5 mL is recommended to generate the larger quantities needed for the preparation of calibration standards and QCs.

Figure 2.

Effect of homogenization solvent volumes to homogeneity in 2-mL tubes (A) and 7-mL tubes (B).

3.3. Evaluation of homogenization density on skin sample homogeneity

Homogenization density plays a critical role in determining the consistency, reproducibility, and applicability of human skin homogenate for concentration determination. This study evaluated the effect of homogenization densities at 20, 40, 80, and 160 mg/mL on the resulting homogeneity of human skin tissue samples. It was found that homogenates prepared at 20 and 40 mg/mL with optimal homogenization volumes were consistent and workable, whereas those at 80 and 160 mg/mL suspensions were too viscous to pipette (Figure 3). The 40-mg/mL human skin homogenate was slightly viscous and posed some handling challenges. Therefore, a density of 20 mg/mL was selected for the final tissue process, balancing pipetting consistency, cost-effectiveness, and the need to generate sufficient matrix for calibration standards and QC preparation.

Figure 3.

Effect of homogenization density on homogeneity.

3.4. Evaluation of povorcitinib stability under acidic and thermal conditions

It is important to evaluate the stability of povorcitinib during the bead homogenization process, particularly when considering that homogenization solvents contain 5% formic acid and the bead homogenization produces excessive heat. It has been established that 5 mM povorcitinib stock solution in acetonitrile:water, 50:50, v/v is stable for up to 740 days stored at 4°C based on long-term LC-MS/MS monitoring (Supplementary Table S1). The acid and heat stability of povorcitinib was assessed per Section 2.5, and the difference between the homogenized sample and the neat solution was 4.3% (Table 2), indicating that povorcitinib is stable during bead homogenization in the acidic solvent system.

Table 2.

Impact of acid and heat on povorcitinib stability.

	Conditions
	Acetonitrile:Water, 50:50, v/v, Neat Fresh	Acetonitrile:Water:Formic Acid, 50:50:5, v/v/v, Bead Homogenization
PAR-LQC	0.0326077	0.0347883
	0.0324743	0.0340243
	0.0327519	0.0332607
Overall mean	0.0326113	0.0340244
SD	0.000138835	0.000763800
%CV	0.4	2.2
% Difference	0	4.3
n	3	3

Abbreviations: CV, coefficient of variation; LQC, low quality control; PAR, peak area ratio; SD, standard deviation; v/v, volume:volume; v/v/v, volume:volume:volume.

3.5. Evaluation of human skin homogenates generated from a wet and dry process

Various methodologies are available for the homogenization of human skin tissues, including mechanical rotor-stator homogenization [11], pulverization and grinding [15], bead homogenization [16], and chemical and enzymatic solubilization [10]. In this study, we evaluated human skin homogenates prepared using two distinct methodologies: a wet homogenization process involving homogenization solvent and mechanical disruption (ie, bead homogenization) and a dry homogenization process using lyophilized tissue followed by mechanical pulverization and dissolution in homogenization solvent (ie, pulverization and grinding; Figure 4). The objective was to assess if the human skin homogenates generated by the two distinct methods were equivalent and interchangeable for the concentration determination of povorcitinib in human skin tissues.

Figure 4.

Schematic diagrams illustrating human skin homogenates obtained through wet (A) and dry (B) homogenization processes.

The matrix equivalency for the quantification of povorcitinib in human skin samples was assessed using back-calculated calibration standards and QCs prepared in skin homogenates generated from the two different processes. Calibration standards from both wet and dry homogenization methods demonstrated good accuracy, with percent bias (%bias) ranging from −4.0 to 4.0 for the wet process and −10.0 to 5.0 for the dry process (Table 3). The performance of low, mid, and high QCs was assessed against calibration standards prepared using both homogenization methods. When calibration standards were generated using the dry process, QCs prepared from wet process homogenates exhibited %bias values between −6.6 and −1.2 and percent coefficient of variation (%CV) values of 1.6 to 6.8, while QCs from dry-process homogenates showed comparable performance with %bias of −9.8 to 0.6 and %CV of 1.0 to 1.5 (Supplementary Table S2). Similarly, when calibration standards were prepared using the wet process, QCs from wet process homogenates yielded %bias values of −4.4 to −2.0 and %CV of 1.5 to 6.5, whereas QCs from dry process homogenates exhibited %bias of −7.6 to −1.9 and %CV of 0.8 to 1.5 (Supplementary Table S3). Together, these results demonstrate the equivalency of human skin homogenates generated by the wet and dry homogenization processes for the quantification of povorcitinib.

Table 3.

Back-calculated povorcitinib concentrations of calibration standards in human skin homogenate prepared by wet and dry homogenization.

	Concentration, nM
	1.00	2.00	10.0	50.0	200	1000	1800	2000
Wet homogenization	0.966	1.88	10.2	48.5	207	1030	1840	1980
	1.05	1.96	10.3	48.4	209	1050	1800	2030
Mean	1.01	1.92	10.3	48.5	208	1040	1820	2010
%bias	1.0	−4.0	3.0	−3.0	4.0	4.0	1.1	0.5
n	2	2	2	2	2	2	2	2
Dry homogenization	0.968	1.98	9.91	53.0	199	992	1620	2020
	1.06	1.98	9.86	52.0	201	979	1620	2060
Mean	1.01	1.98	9.89	52.5	200	986	1620	2040
%bias	1.0	−1.0	−1.1	5.0	0.0	−1.4	−10.0	2.0
n	2	2	2	2	2	2	2	2

3.6. Method validation for quantifying povorcitinib in human skin via a dry homogenization process

An LC-MS/MS – based method validation of human skin homogenates prepared using a dry process was conducted to ensure the method’s suitability for bioanalytical applications in accordance with the ICH M10 guidance [17]. Key validation parameters included selectivity, precision, accuracy, carryover, matrix effect, recovery, dilution integrity, and stability (Table 4). Because of the limited availability of human skin containing known amounts of povorcitinib, stability assessments (benchtop, freeze – thaw, long-term storage) and extraction recovery were evaluated using human skin homogenate spiked with known concentrations of povorcitinib. Povorcitinib showed a linear response over the concentration range of 1 to 2000 nM in human skin homogenate with a correlation coefficient ≥0.9942. The assay performance results are summarized in Table 4. The elements of the calibration curve included a blank sample (no analyte, no internal standard), a zero sample (blank plus internal standard), and eight calibrators at 1, 2, 10, 50, 200, 1000, 1800, and 2000 nM, respectively. The acceptance criteria for the calibration curve were that the non-zero calibrators should be ± 15% of nominal concentrations except at the LLOQ, where it should be ± 20% of the nominal concentration, and at least 75% and a minimum of six non-zero calibrators should meet the aforementioned criteria. For the back-calculated non-zero calibration standards obtained during accuracy and precision runs, the cumulative accuracy (%bias) from LLOQ to the upper limit of quantification (ULOQ) was −5.5% to 3.0% and the cumulative precision (%CV) was ≤7.1%. For accuracy and precision, intra-assay accuracy (%bias) ranged from −9.8% to 9.0% and precision (%CV) ≤8.6%, and inter-assay accuracy (%bias) ranged from −5.3% to 3.5% and precision (%CV) ≤11.3%. The matrix effect for povorcitinib evaluated with homogenate from 6 individual lots showed %bias of −14.7 to 6.3 and %CV ≤ 11.4%, indicating negligible matrix effect. Selectivity of the method met the acceptance criteria specified in Section 2.7.1. Typical blank human skin homogenate and LLOQ sample chromatograms with an injection volume of 2 μL are presented in Figure 5. The dilution integrity was assessed using 20-fold dilutions with five replicates. The accuracy (%bias) and precision (%CV) of the dilution QC sample was −9.5% and 9.2%, respectively. In addition, povorcitinib was found to be stable in human skin homogenate after 24 hours at ambient temperature (benchtop stability), through four freeze/thaw cycles. The long-term frozen sample storage stability was verified to be at least 29 days after homogenates were stored frozen at a nominal temperature of −70°C. It was also verified that povorcitinib in the injection solution was stable for at least 119 hours when kept at 4°C (processed stability). The validation results demonstrated that the bioanalytical method meets the regulatory criteria for bioanalytical method performance, supporting its use in the quantitative analysis of povorcitinib in human skin tissue matrices.

Table 4.

Summary of validation results.

Parameter	Results
Calibration standards cumulative accuracy from LLOQ to ULOQ,a %bias	−5.5 to 3.0
Calibration standards cumulative precision from LLOQ to ULOQ,a %CV	≤7.1
LLOQ to ULOQ QC Intra-run accuracy range, %bias	−9.8 to 9.0
LLOQ to ULOQ QC Intra-run precision range, %CV	≤8.6
LLOQ to ULOQ QC Inter-run accuracy, %bias	−5.3 to 3.5
LLOQ to ULOQ QC inter-run precision, %CV	≤11.3
Stock solution stability in acetonitrile:water at 50:50 (v/v)	740 day at 2°C − 8°C;25 hours at ambient temperature
Stability during tissue homogenization process	Met acceptance criteria
Processed sample stability	119 hours at 4°C
Benchtop stability in skin homogenate	24 hours at ambient temperature
Freeze/thaw stability in skin homogenate	4 cycles at −70°C (thawed at ambient temperature)
Long-term storage stability in skin homogenate	29 days at −70°C
Sensitivity	Analyte response (LLOQ/zero calibrator [ie, blank + IS]) > 5
Mean recovery of analyte, %	87.9
Mean recovery of internal standard, %	87.4
Matrix effect from 6 lots of matrix-accuracy range, %bias	−14.7 to 6.3
Matrix effect from 6 lots of matrix-precision range, %bias	≤11.4
Dilution	4000 nM diluted 20-fold; %bias, −9.5%; %CV, 9.2%
Batch size test	141 injections
Carryover	< 20% LLOQ (povorcitinib) in all runs except Run 1; < 5% carryover IS (povorcitinib-d6)

Abbreviations: CV, coefficient of variation; IS, internal standard; LLOQ, lower limit of quantification; QC, quality control; ULOQ, upper limit of quantification; v/v, volume:volume.

^aFrom 3 independent accuracy and precision runs.

Figure 5.

Multiple reaction monitoring chromatograms of povorcitinib and D6-povorcitinib in human skin homogenate. Povorcitinib in blank human skin homogenate (top), povorcitinib in human skin homogenate spiked at LLOQ (middle), and D6- povorcitinib (bottom) in an LLOQ sample.

Abbreviation: LLOQ, lower limit of quantification.

3.7. Application to clinical studies

Following the successful validation of the dry homogenization method, its applicability was demonstrated using human skin biopsy samples collected from subjects enrolled in a phase I INCB054707-112 study evaluating dermal povorcitinib concentrations at steady state after oral administration in healthy participants. Skin biopsies were collected at predetermined time points post-application of povorcitinib and processed using the dry homogenization protocol. The resulting homogenates exhibited consistent concentration data, as demonstrated by the incurred sample reanalysis (ISR) results (Table 5). ISR is a necessary component of bioanalytical method validation and is conducted to critically support the precision and accuracy measurements established with the QCs during method validation. If the total number of study samples is less than or equal to 1000, at least 10% of analyzed samples should be reanalyzed and the difference between the initial concentration and the repeated concentration should be within ±20% for at least 2/3 (ie, 66.7%) of the repeats for ISR to be considered acceptable [17]. For the INCB054707-112 study, 72 skin samples were collected, received, and analyzed. Of those, 22 samples were selected for the ISR, and 90.9% (20/22) of the repeated samples met ISR acceptance criteria, demonstrating the reliability of the reported concentration data.

Table 5.

Incurred sample reanalysis results.

Sample ID	Original concentration, nM	Reassay concentration, nM	%Differencea
101 A Skin Homogenate-1, 0 h predose	23.0	19.0	−19.0
101 A Skin Homogenate-1, 24 h postdose	19.4	17.9	−8.0
102 A Skin Homogenate-1, 12 h postdose	30.8	29.9	−3.0
103 A Skin Homogenate-1, 72 h postdose	7.48	7.17	−4.2
104 A Skin Homogenate-1, 48 h postdose	13.3	12.9	−3.1
105 A Skin Homogenate-1, 24 h postdose	13.6	12.1	−11.7
106 A Skin Homogenate-1, 48 h postdose	12.1	9.24	−26.8
107 A Skin Homogenate-1, 24 h postdose	23.9	20.5	−15.3
108 A Skin Homogenate-1, 12 h postdose	31.0	28.0	−10.2
109 A Skin Homogenate-1, 6 h postdose	39.7	39.3	−1.0
110 A Skin Homogenate-1, 48 h postdose	21.8	21.7	−0.5
111 A Skin Homogenate-1, 0 h predose	14.3	12.1	−16.7
111 A Skin Homogenate-1, 72 h postdose	2.79	2.24	−21.9
112 A Skin Homogenate-1, 12 h postdose	20.3	20.7	2.0
113 A Skin Homogenate-1, 24 h postdose	12.8	11.0	−15.1
113 A Skin Homogenate-1, 72 h postdose	3.68	3.7	0.5
114 A Skin Homogenate-1, 12 h postdose	16.6	15.1	−9.5
115 A Skin Homogenate-1, 24 h postdose	15.7	15.0	−4.6
116 A Skin Homogenate-1, 0 h predose	22.4	21.8	−2.7
117 A Skin Homogenate-1, 6 h postdose	39.5	38.8	−1.8
118 A Skin Homogenate-1, 0 h predose	19.6	18.5	−5.8
118 A Skin Homogenate-1, 48 h postdose	8.01	7.73	−3.6

^a%Difference = ([Reassay value – Original value]/Mean of reassay and original values) × 100.

4. Conclusions

This study establishes a robust and standardized approach for human skin tissue processing by identifying optimal homogenization solvents, processing volumes, and homogenate densities, and by characterizing analyte stability under heat and acidic conditions. Through a direct comparison of wet and dry homogenization techniques, we demonstrated that both approaches generate equivalently homogeneous matrices suitable for quantification of povorcitinib in human skin. We further validated a dry homogenization workflow optimized specifically for small and heterogeneous human skin samples. This work represents the first systematic evaluation and optimization of key homogenization parameters tailored specifically for human skin tissues, an area where standardized and validated procedures have been notably lacking. By refining homogenization conditions and developing both wet and dry processing workflows, this study provides a flexible platform capable of supporting a broad range of analyte quantifications, enhancing reproducibility, and enabling cross-study comparability in dermatologic research and therapeutic development. The successful application of the validated method to a Phase I clinical study further demonstrates its suitability and robustness for real-world clinical research. Together, these advances lay the groundwork for standardized human skin tissue processing protocols that can be widely adopted across preclinical and clinical settings. Future work may extend this methodology to skin tissues from additional clinical indications to further assess its compatibility with varied tissue architectures and biomolecular compositions.

Funding Statement

This work was supported by Incyte Corporation (DE, USA).

Article highlights

• Developed and validated optimized methods for human skin homogenization to quantify small-molecule drugs (eg, povorcitinib).

• Identified effective solvent mixtures, optimal solvent volumes (less than half of tube capacity), and a practical homogenate density (20 mg/mL).

• Demonstrated povorcitinib stability under acidic and heat conditions.

• Demonstrated wet and dry homogenization methods were equivalent, with a validated dry process for sample analysis.

• Established standardized protocols adaptable for a broad range of analytes and diseased skin tissues for therapeutic evaluation in dermatologic research and development.

Author contributions

Zhiyin Xun: study conception, experimental design, problem solving, data analysis, drafting of the original manuscript, and editing; Lin Zhang: experimental execution, manuscript reviewing and editing; Hongjin Wen: experimental execution, manuscript reviewing and editing; Ryan McGee: manuscript reviewing and editing; Phillip Wang: manuscript reviewing and editing. All authors reviewed and approved the final version of the manuscript.

Disclosure statement

Zhiyin Xun, Lin Zhang, Ryan McGee, and Phillip Wang are employees and shareholders of Incyte Corporation. Hongjin Wen works as a Professional Service Resource at Incyte. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript.

No writing assistance was used in the production of this manuscript apart from those disclosed.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Ethical declaration

The authors state that they have obtained Advarra Institutional Review Board approval (reference number Pro00080431) or have followed the principles outlined in the Declaration of Helsinki for all human or animal experimental investigations. In addition, for investigations involving human subjects, informed consent has been obtained from the participants involved.

Data availability statement

Data are available from the corresponding author upon request.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/17576180.2025.2612493