Development and validation of a fast ultra‐high‐performance liquid chromatography tandem mass spectrometry method for determining carbonic anhydrase inhibitors and their metabolites in urine and hair

Abstract A new, rapid, sensitive, and comprehensive ultra‐high‐performance liquid chromatography tandem mass spectrometry (UHPLC–MS/MS) method for quantifying diuretics (acetazolamide, brinzolamide, dorzolamide, and their metabolites) in human urine and hair was developed and fully validated. Twenty‐five milligrams of hair were incubated with 500‐μl M3® buffer reagent at 100°C for 1 h for complete digestion. After cooling, 1‐μl supernatant was injected onto chromatography system. Urine samples were simply diluted before injection. The chromatographic run time was short (8 min) through a column with a mobile phase gradient. The method was linear (determination coefficients always higher than 0.99) from limit of quantification (LOQ) to 500 ng/ml in urine and from LOQ to 10 ng/mg in hair. LOQs ranged from 0.07 to 1.16 ng/ml in urine and from 0.02 to 0.15 ng/mg in hair. No significant ion suppression due to matrix effect was observed, and process efficiency was always higher than 80%. Intra‐ and inter‐assay precision was lower than 15%. The suitability of the methods was tested with six urine and hair specimens from patients treated with acetazolamide, dorzolamide, or brinzolamide for ocular diseases or systemic hypertension. Average urine concentrations were 266.32 ng/ml for dorzolamide and 47.61 ng/ml for N‐deethyl‐dorzolamide (n = 3), 109.27 ng/ml for brinzolamide and 1.02 ng/ml for O‐desmethyl‐brinzolamide (n = 2), and finally, 12.63 ng/ml for acetazolamide. Average hair concentrations were 5.94 ng/mg for dorzolamide and 0.048 ng/mg for N‐deethyl‐dorzolamide (n = 3), 3.26 ng/mg for brinzolamide (n = 2), and 2.3 ng/mg for acetazolamide (n = 1). The developed method was simple and fast both in the extraction procedures making it eligible in high‐throughput analysis for clinical forensic and doping purposes.


| INTRODUCTION
Diuretics are pharmaceutical products commonly used to increase the renal function and stimulate the elimination of sodium. Therefore, they have long been used in the treatment of hypertension, edemas, electrolyte decompensation, and heart and renal failure. [1][2][3] They are classified by their chemical structure, efficacy, speed of onset, or site of action. The most common classes are (i) loop diuretics, (ii) thiazides and related diuretics, (iii) potassium-sparing diuretics, (iv) carbonic anhydrase inhibitors (CAIs), (v) osmotic diuretics, and (vi) miscellaneous diuretics. 4 Among CAIs, the most commonly prescribed drugs are dorzolamide, brinzolamide, and acetazolamide, which are mainly used in ocular pathology for the treatment of glaucoma by reducing the intraocular pressure. [4][5][6] CAIs inhibit the activity of carbonic anhydrase (enzymes that catalyze the interconversion between carbon dioxide and water and the dissociated ions of carbonic acid) in proximal convoluted tubules and prevent reabsorption of bicarbonates from renal tubule. 2 For their pharmacological action, CAIs are included in class S5 (Diuretics and Masking Agents) of the prohibited list of substances of the World Anti-Doping Agency (WADA), with the exception of drospirenone, pamabrom, CAIs in topical ophthalmic administration (e.g., dorzolamide and brinzolamide), and felypressin in local administration in dental anaesthesia, 7 and therefore are banned in competitive sports mainly because they can be used to tamper with urine drug test results through dilution means. Hence, the identification of CAIs in urine and other biological matrices to provide orthogonal data on exposure (e.g., hair) can be of interest to document misuse. 8 Presently, only few analytical methods [9][10][11][12][13][14][15][16][17][18] for the detection of CAIs in biological samples have been published in the international literature and none of them simultaneously determined the three main CAIs (dorzolamide, brinzolamide, and acetazolamide) and/or their metabolites.
Herein, a fast and simple method by ultra-high-performance liquid chromatography tandem mass spectrometry (UHPLC-MS/MS) was developed and fully validated to quantify dorzolamide, brinzolamide, acetazolamide, and the metabolites of the two first compounds, Nacetyl-dorzolamide, N-deethyl-dorzolamide, and O-desmethylbrinzolamide, in human urine and hair. The method was then applied to samples collected from six authentic cases.

| Human samples
Blank human urine and hair samples were obtained from the laboratory storehouse of blank biological samples. In detail, drug-free urine and hair were gently donated by university personnel involved in this study, signing an informed consent. All samples were prescreened by the below-reported method to assess absence of the analytes under investigations.
Urine and hair specimens were collected from six patients treated

| Sample treatment
Urine samples (100 μl) were fortified with 5-μl ISs working solution in conical glass tubes and vortexed. After adding 5 ml of belowdescribed mobile phase A:B 95:5 (v/v), tubes were capped, vortexed for 15 s, and centrifuged at 15,000 g for 3 min. Supernatants (100 μl) were transferred into autosampler glass vials, prior to injection onto the chromatographic system.
Hair samples were washed twice with dichloromethane and dried under nitrogen at 45 C. An aliquot of 25 mg was finely cut into pieces (<5 mm) in glass tubes and fortified with 5-μl ISs working solution.
After adding 100-μl M3 ® reagent, tubes were capped and incubated at 100 C for 1 h, for complete digestion. Then samples were cooled down at room temperature, and 1 ml was transferred into autosampler glass vial, prior to injection onto the chromatographic system. Initial conditions were 5% B, held for the first 0.25 min, increased to 20% B from 0.26 to 3 min, increased to 95% from 3.1 to 5 min, held for 0.5 min, and returned 5% B to 8.00 min for re-equilibration.

| UHPLC-MS/MS analysis
Autosampler and oven temperatures were 10 C and 50 C, respectively. The injection volume was 3 μl for urine and 1 μl for hair samples.
The detection was performed with a triple quadrupole mass spectrometer operating in scheduled multiple reaction monitoring (MRM) mode, with two transitions for each analyte and IS, as reported in Table 1

| Method validation
The method was fully validated in urine and hair according to most recent criteria for the validation of analytical toxicology. [19][20][21][22] Calibration points ranged from limit of quantification (LOQ) to 500 ng/ml in urine and from LOQ to 10 ng/mg in hair for each analyte. Least-squares linear regression with the reversed square of the concentration of the analyte (1/x 2 ) was employed to define the calibration curve using the ratios of the peak area of analytes and IS.
The effect of three cycles of freezing at À20 C and thawing on five different aliquots of QC samples (urine and hair) were verified.
The post-extraction stability of the compounds under investigation was assessed under various conditions. The stability of five different extracted aliquots of QC samples (urine and hair) was assessed for 0, 1, 2, 4, 24, and 48 h in the light and dark at room temperature and at À20 C for up to 1 month.  The method exhibited good linearity and showed coefficients of determination (r 2 ) always better than 0.99. The percentage of relative error for all the calibration points of the curve was less than 10%.

T A B L E 1 Mass spectrometry parameters for analytes and internal standard under investigation
LOD ranged from 0.02 to 0.17 ng/ml in urine and from 0.01 to 0.05 ng/mg in hair, and LOQ ranged from 0.07 to 1.16 ng/ml in urine and from 0.02 to 0.15 ng/mg in hair; accuracy and precision were within ±20% of the target at the LOQ. REs ranged from 81% to 98.1% (relative standard deviation [RSD] between 1.5% and 5.6%) in urine and from 96.5% to 99% (RSD between 4% and 12%) in hair. PE was always better than 73% and ion suppression due to ME within 12% (RSD between 3% and 7.6% in urine and between 6.5% and 10% in hair).
The effect of freezing at À20 C and thawing cycles on QC samples did not significantly alter the compounds concentration in biological matrices (differences from controls less than 10%). Furthermore, the post-extraction stability of the substances under investigation was proved at room temperature and at À20 C, with differences in concentrations less than 10% control samples.

| Real samples
Target analyte concentrations in authentic samples are reported in Note: Low, medium, and high quality control (QC) samples contained all standards at 31.5, 87, and 350 ng/ml, respectively. Analytical recovery, matrix effect, and process efficiency are displayed as mean value of low, medium, and high QC values. Abbreviations: CV, coefficient of variation; LOD, limit of detection; LOQ, limit of quantification.   Francesco Paolo Busardò https://orcid.org/0000-0002-3082-4532