Determination of perhexiline and its metabolite hydroxyperhexiline in human plasma by liquid chromatography/tandem mass spectrometry

Abstract

Perhexiline is a medication indicated for the management of moderate to severe angina pectoris when other treatments have proven ineffective. This drug exhibits a narrow therapeutic index, and its metabolism is saturable and influenced by genetic polymorphism, specifically involving the CYP2D6 enzyme. Monitoring the concentrations of both the parent drug and its primary metabolite is deemed essential to optimize therapeutic benefits and minimize the potential for hepatotoxicity and neuropathy.

A swift, straightforward, and sensitive liquid chromatography/tandem mass spectrometry assay was developed for quantifying perhexiline and its metabolite cis-hydroxyperhexiline in human plasma. Following protein precipitation with acetonitrile, perhexiline, its primary metabolite cis-hydroxyperhexiline, and nordoxepin, employed as the internal standard, were separated using a phenyl-hexyl column with a gradient elution of 0.05% formic acid and methanol. Detection of the three compounds was achieved through electrospray ionization in the positive mode.

Calibration curves demonstrated linearity across the concentration range of 10–2000 µg/L, with a correlation coefficient exceeding 0.999. The assay exhibited a bias within ±10%, intra- and inter-day coefficients of variation (imprecision) of ≤8.1%, and a limit of quantification of 10 µg/L for both perhexiline and hydroxyperhexiline. This assay is currently being implemented successfully in clinical practice to enhance the safety and efficacy of perhexiline usage.

Introduction

Perhexiline, an anti-anginal agent, was first introduced to the market in the 1970s. While it proves to be an effective treatment for severe angina, a notable occurrence of severe hepatotoxicity and neurotoxicity led to a rapid decline in its utilization within the initial years. Despite the withdrawal of perhexiline from the market in numerous countries globally during the 1980s, its prescription persisted in Australia and New Zealand for patients with intractable angina who exhibited resistance or intolerance to conventional therapies, or who were unsuitable candidates for coronary bypass surgery.

Given the drug’s low therapeutic index and the observation that significant toxicity was concentration-dependent, therapeutic drug monitoring became an integral component of treatment. This necessity was further substantiated by subsequent evidence indicating that perhexiline metabolism is subject to genetic polymorphism involving the CYP2D6 enzyme, and that individuals identified as slow metabolizers for this enzyme were at a higher risk of experiencing toxicity. Another contributing factor to dose-concentration-effect challenges was the presence of saturable, or non-linear, metabolism.

Perhexiline undergoes primary metabolism into a pair of geometric isomers of the mono-hydroxy metabolite, specifically cis-hydroxyperhexiline and trans-hydroxyperhexiline. The hydroxylation of perhexiline to cis-hydroxyperhexiline represents the major metabolic pathway governed by cytochrome P450 2D6 (CYP2D6). The ratio of cis-hydroxyperhexiline to perhexiline concentrations can serve as an indicator for identifying poor metabolizers early in perhexiline therapy, thereby facilitating appropriate dose adjustments to prevent unnecessary drug toxicity. Routine monitoring of plasma perhexiline and cis-hydroxyperhexiline levels has facilitated the safer utilization of this drug in an increasing number of patients within Australia and New Zealand, and holds the potential for broader application in other countries.

Various analytical methodologies have been developed for the quantification of perhexiline and cis-hydroxyperhexiline in human plasma, including high-performance liquid chromatography with fluorescence or ultraviolet detection, and gas chromatography. Services providing perhexiline therapeutic monitoring have predominantly employed high-performance liquid chromatography methods. Due to perhexiline’s lack of significant absorption within the ultraviolet spectrum, sample preparation for high-performance liquid chromatography methods necessitates derivatization for fluorescence or ultraviolet detection and multiple liquid-liquid extraction steps, which are both reagent-intensive and time-consuming.

In recent years, high-performance liquid chromatography coupled with tandem mass spectrometric detection has emerged as a powerful technique for the quantitative analysis of drugs and metabolites in biological matrices. This technique offers high selectivity and simplification of both sample extraction procedures and chromatography. A previously reported simple liquid chromatography/mass spectrometry method for determining perhexiline and hydroxyperhexiline utilized protein precipitation for sample preparation.

In comparison to liquid chromatography/mass spectrometry, liquid chromatography/tandem mass spectrometry can achieve enhanced specificity and sensitivity by employing collision-induced dissociation while monitoring unique precursor-to-product ion transitions. The objective of the present study was to develop and validate a rapid, simple, specific, sensitive, robust, and reliable liquid chromatography/tandem mass spectrometry method for the determination of perhexiline and cis-hydroxyperhexiline in human plasma, suitable for routine drug monitoring applications. An additional aim was to utilize a small plasma volume and a straightforward sample preparation procedure without compromising specificity and sensitivity.

Experimental

Materials

Perhexiline maleate and nordoxepin were obtained from Sigma Co. in Australia. Cis-hydroxyperhexiline was a kind donation from Dr Benedetta Sallustio in Adelaide, Australia. This reference material for cis-hydroxyperhexiline contained a small amount of trans-hydroxyperhexiline, specifically 2.2%. High-performance liquid chromatography grade acetonitrile, methanol, and formic acid were purchased from BDH in Poole, UK. Purified water was produced using a Milli-Q Reagent Water System from Millipore in MA, USA. Human plasma, used as the assay blank and for preparing standard solutions, was sourced from New Zealand Blood Services in Christchurch, New Zealand.

Instrumentation and analytical conditions

The liquid chromatography-tandem mass spectrometry system comprised a Shimadzu LC-20AD high-performance liquid chromatography system from Shimadzu Corporation in Kyoto, Japan, connected to a 3200 Q TRAP mass spectrometer from Applied Biosystems in Foster City, Canada, which was equipped with a TurboIonSpray source. Analyst software, also from Applied Biosystems in Foster City, Canada, was employed for controlling the equipment, coordinating the acquisition of data, and analyzing the resulting data. Perhexiline, cis-hydroxyperhexiline, and the internal standard nordoxepine were separated using gradient elution on a Luna Phenyl-Hexyl 3 µm, 50 mm × 2.0 mm internal diameter analytical column, which was fitted with a Phenyl 4.0 mm × 2.0 mm internal diameter guard column, both from Phenomenex in Torrance, CA, USA.

The mobile phase consisted of two components: solvent A, which was a 0.05% formic acid solution, and solvent B, which was methanol. The flow rate of the mobile phase was maintained at 0.3 mL per minute. The initial condition of the mobile phase was 80% solvent A and 20% solvent B. A linear gradient was then applied, where the proportion of mobile phase B was increased from 20% to 90% over a period of one minute. Following this, the mobile phase composition was returned to the initial condition and allowed to re-equilibrate for two minutes. The total duration of each analysis was five minutes.

The mass spectrometer was operated in the positive ion mode. The flow rates for the curtain gas, Gas 1, and Gas 2 were set at 20, 45, and 60 psi, respectively. The ion spray voltage was set to 5000 V, and the source temperature was maintained at 500 degrees Celsius. Data acquisition was performed using multiple reaction monitoring. For perhexiline, cis-hydroxyperhexiline, and the internal standard nordoxepin, the ions corresponding to their protonated molecular species were selected in the first mass spectrometer. These selected ions were then fragmented using nitrogen gas to produce specific product ions, which were subsequently monitored by the second mass spectrometer.

The optimized precursor-to-product ion transitions that were monitored were mass-to-charge ratio 278.3 to 95.2 for perhexiline, with a declustering potential of 56 V and a collision energy of 37 V; mass-to-charge ratio 294.3 to 95.2 for cis-hydroxyperhexiline, with a declustering potential of 56 V and a collision energy of 43 V; and mass-to-charge ratio 266.1 to 107.1 for nordoxepin, with a declustering potential of 36 V and a collision energy of 29 V.

Standards

A stock standard solution of perhexiline at a concentration of 1.0 mg/mL was prepared by dissolving 14.2 mg of perhexiline maleate in 10 mL of methanol. Similarly, a stock standard solution of cis-hydroxyperhexiline at a concentration of 1.0 mg/mL was prepared by dissolving 10 mg of cis-hydroxyperhexiline in 10 mL of methanol. Two identical sets of these stock standard solutions were prepared, one for the generation of plasma calibration curves and the other for the preparation of plasma quality control samples.

The plasma calibration curves for both perhexiline and cis-hydroxyperhexiline were generated by spiking drug-free human plasma with the standard solutions to achieve a calibration range of 10 to 2000 µg/L for both analytes. A stock internal standard solution of nordoxepin at a concentration of 1.0 mg/L was prepared by dissolving 10 mg of nordoxepin in 10 mL of methanol. A working solution of the internal standard at a concentration of 250 µg/L was prepared by diluting 2.5 µL of the stock solution to a final volume of 10 mL with water. Plasma quality control standards for perhexiline and cis-hydroxyperhexiline were prepared in single 5 mL aliquots at concentrations of 10, 40, 100, 500, and 2000 µg/L and stored at a temperature of -30 degrees Celsius until they were analyzed.

Sample preparation

To each 50 µL aliquot of blank plasma, standard plasma, quality control plasma, or patient plasma samples, 50 µL of the internal standard nordoxepin at a concentration of 250 µg/L was added. The resulting mixture was briefly vortexed, and then 200 µL of acetonitrile was added to precipitate the proteins present in the plasma. Following centrifugation at 15,000 × g for a duration of 5 minutes, a 50 µL aliquot of the clear supernatant was carefully transferred and mixed with 200 µL of a 0.05% formic acid solution. This final mixture was then transferred to the autosampler 96 well plate. A volume of 10 µL of this prepared sample was subsequently injected into the liquid chromatography-tandem mass spectrometry system for analysis.

Validation

The standard curves were constructed by plotting the ratios of the peak areas of the analytes, perhexiline and cis-hydroxyperhexiline, to the peak area of the internal standard against the corresponding concentrations of perhexiline and cis-hydroxyperhexiline. The linearity of these standard curves was evaluated using linear regression analysis with a weighting factor of 1/x. To assess the recoveries of the assay and any potential matrix effects, three sets of standard samples were prepared, following a modification of the method described by Matuszewski and colleagues. This was done for both perhexiline and cis-hydroxyperhexiline at four different concentrations: 40, 100, 500, and 2000 µg/L, and for nordoxepin at a concentration of 250 µg/L, which was the concentration used in the assay.

The first set of standards was prepared in plasma obtained from six different sources, with six replicate samples at each concentration. The second set was prepared by spiking the after-protein precipitation extracts of blank plasma, also from the same six different sources as the first set, with the analytes and the internal standard. The third set consisted of standard solutions prepared directly in the mobile phase.

The absolute recoveries at each concentration level were determined by comparing the peak areas of perhexiline, cis-hydroxyperhexiline, and the internal standard in the plasma standards to their respective peak areas in the spiked after-protein precipitation blank plasma extracts at the corresponding concentrations, with six replicates for each comparison. The absolute recovery was calculated as the ratio of the peak area of the analyte from the spiked plasma sample to the peak area of the analyte from the spiked after-protein precipitation blank plasma extract sample, multiplied by 100%.

Matrix effects were evaluated by comparing the peak areas of perhexiline, cis-hydroxyperhexiline, and the internal standard from the spiked after-protein precipitation blank plasma extracts with the responses obtained from standard solutions of the same concentrations prepared in the mobile phase, again with six replicates for each comparison. Quality control was assessed by analyzing six replicate samples at each quality control concentration on the same day to determine intra-day variability, and by analyzing one sample at each quality control concentration on six different days to determine inter-day variability.

Bias was calculated as the difference between the measured concentration and the actual concentration, expressed as a percentage of the actual concentration. Imprecision was quantified as the intra-day and inter-day coefficients of variation. The limit of quantification for this assay was defined as the lowest concentration of perhexiline and cis-hydroxyperhexiline that could be reliably detected with acceptable accuracy and precision, based on the analysis of six replicate samples. According to the guidelines provided by the US Food and Drug Administration for bioanalytical method validation, the mean value determined at the lowest quantifiable concentration should not deviate by more than 20% from the true value, and the precision at this concentration, expressed as the coefficient of variation, should not exceed 20%.

The stability of perhexiline and hydroxyperhexiline concentrations under freeze-thaw conditions was evaluated using quality control samples at concentrations of 10, 40, 100, 500, and 2000 µg/L, which were subjected to four freeze-thaw cycles prior to analysis. The stability of plasma quality control samples stored at a temperature of -30 degrees Celsius was assessed by analyzing their concentrations at weekly intervals over a period of ten months. The stability of the stock standard solutions of perhexiline and hydroxyperhexiline when stored at 4 degrees Celsius for ten months was evaluated by comparing their responses to those of freshly prepared standard solutions. The stability of the processed samples maintained at 4 degrees Celsius, which is the typical temperature of the autosampler, for a duration of three days was also evaluated by comparing the obtained results with the initial results. In all stability studies, perhexiline and hydroxyperhexiline were considered stable if the observed degradation was less than 10% of the initial concentration measured at day zero.

Results and discussion

Mass spectrometry and chromatography

The mass spectrometry and mass spectrometry parameters were optimized to achieve the highest possible signal responses for perhexiline, cis-hydroxyperhexiline, and the internal standard nordoxepin using electrospray ionization in the positive ion mode. The protonated molecular ions observed were mass-to-charge ratio 278.3 for perhexiline, mass-to-charge ratio 294.3 for cis-hydroxyperhexiline, and mass-to-charge ratio 266.1 for nordoxepin. The transitions that produced the most abundant product ions were mass-to-charge ratio 278.3 to 95.2 for perhexiline, mass-to-charge ratio 294.3 to 95.2 for cis-hydroxyperhexiline, and mass-to-charge ratio 266.1 to 107.1 for nordoxepin.

Perhexiline, cis-hydroxyperhexiline, and the internal standard were separated from other components in the plasma matrix using a Phenomenex Luna Phenyl-Hexyl column and a mobile phase composed of 0.05% formic acid and methanol. Due to the significant difference in polarity between perhexiline and its hydroxylated metabolite, separation was a lengthy process under isocratic elution conditions.

Therefore, gradient elution was chosen to improve resolution and enhance separation efficiency. Under the chromatographic conditions employed, the retention times observed were approximately 2.69 minutes for the internal standard, 2.81 minutes for cis-hydroxyperhexiline, and 2.96 minutes for perhexiline. Because the cis-hydroxyperhexiline reference material contained a small amount of trans-hydroxyperhexiline, specifically 2.2%, the reference material produced a minor peak corresponding to trans-hydroxyperhexiline. This peak appeared just before the cis-hydroxyperhexiline peak and exhibited identical precursor-to-product ion transitions to cis-hydroxyperhexiline.

Complete baseline separation of cis-hydroxyperhexiline and trans-hydroxyperhexiline was not achieved; however, this did not significantly affect the quantification of cis-hydroxyperhexiline. Blank plasma samples obtained from more than six different sources of the same matrix were tested for potential interferences. The peaks corresponding to perhexiline, cis-hydroxyperhexiline, and the internal standard were free from any interference from other peaks present in the blank plasma samples. Furthermore, no carry-over was observed when an extract of blank plasma was injected immediately after the highest calibration standard.

Sample preparation

The high sensitivity of the liquid chromatography-tandem mass spectrometry technique allowed for the use of a very small volume of plasma, specifically 50 µL, for the quantification of perhexiline and cis-hydroxyperhexiline in plasma. Protein precipitation is recognized as the simplest and most rapid method for preparing plasma samples for the measurement of drug concentrations. To identify the most effective precipitating agent for sample preparation, three commonly used precipitants, namely acetonitrile, methanol, and perchloric acid, were compared. Precipitation with acetonitrile was found to be the most suitable for sample clean-up, with a volume ratio of acetonitrile to plasma of 4:1 being optimal. Perhexiline, cis-hydroxyperhexiline, and the internal standard were free from interference from endogenous compounds present in the plasma. To ensure consistent long-term performance of the chromatographic system, the guard column cartridge was replaced every 200 to 300 injections. The analytical column showed no signs of performance degradation even after more than 2000 injections.

Method validation

Plasma standard curves for perhexiline and cis-hydroxyperhexiline exhibited linearity, with correlation coefficients greater than 0.999, over the concentration range of 10 to 2000 µg/L. This range adequately covers the therapeutic range of perhexiline, which is 150 to 600 µg/L in our laboratory. The y-intercept of the standard curves was not significantly different from zero. The typical standard curve equations were y = 0.948x + 2.85 × 10−6 with a correlation coefficient of 0.9995 for perhexiline, and y = 1.37x + 6.4 × 10−6 with a correlation coefficient of 0.9997 for cis-hydroxyperhexiline, where y represents the ratio of the analyte peak area to that of the internal standard, and x represents the plasma concentration of the analyte.

The lower limit of quantification for both perhexiline and cis-hydroxyperhexiline was approximately 10 µg/L in plasma. At this concentration, the mean measured values were within ±10% of the spiked values, and the intra-day and inter-day coefficients of variation were less than 8.5%. No consistent positive or negative bias was observed for the plasma quality control samples, and the mean measured values were within ±5% of the spiked values. Imprecision was low, as indicated by both intra-day and inter-day coefficients of variation of less than 5.0% at all concentrations ranging from 40 to 2000 µg/L.

The absolute recoveries of perhexiline and cis-hydroxyperhexiline at concentrations of 40, 100, 500, and 2000 µg/L were similar and consistent, with mean values exceeding 95%. The absolute recovery of the internal standard nordoxepin at the concentration used in the assay was 82%. Matrix effects were evaluated by comparing the response of perhexiline, cis-hydroxyperhexiline, and the internal standard from the spiked after-protein precipitation blank plasma extracts with the response of standard solutions at the same concentrations prepared in the mobile phase. A matrix effect value of 100% indicates that the responses in the mobile phase and in protein-precipitated plasma were identical, suggesting no absolute matrix effect. A value greater than 100% indicates ionization enhancement, while a value less than 100% indicates ionization suppression. The matrix effects, expressed as mean ± standard deviation percentage, determined at concentrations of 40, 100, 500, and 2000 µg/L for perhexiline were 99.6 ± 5.5%, 108 ± 5.9%, 98.2 ± 5.1%, and 106 ± 8.2%, respectively. For cis-hydroxyperhexiline, the matrix effects at the same concentrations were 106 ± 4.1%, 113 ± 4.3%, 105 ± 4.9%, and 106 ± 8.8%, respectively. The matrix effect observed for the internal standard nordoxepin was 94 ± 5.0%. These results indicated that there were no significant matrix effects observed in this assay. Perhexiline and hydroxyperhexiline were found to be stable in plasma for at least four freeze-thaw cycles when stored at a temperature of -30 degrees Celsius. The plasma quality control samples at concentrations of 10, 40, 100, 500, and 2000 µg/L remained stable for a period of at least 10 months when stored at -30 degrees Celsius. The stock standard solutions of perhexiline and hydroxyperhexiline were shown to be stable for at least 10 months when stored at 4 degrees Celsius. The processed samples were stable for a minimum of 3 days when stored at 4 degrees Celsius.

Application of the assays

The described method is currently employed in our laboratory service for the purpose of monitoring the plasma concentrations of perhexiline and its metabolite hydroxyperhexiline in patients undergoing perhexiline therapy. To ensure the accuracy and reproducibility of this method, our laboratory participates in a monthly interlaboratory proficiency testing program for perhexiline therapeutic monitoring services. This program is organized by the Department of Clinical Pharmacology at The Queen Elizabeth Hospital in Woodville, South Australia. The monthly reports received have consistently demonstrated that the performance of this method for the analysis of perhexiline and cis-hydroxyperhexiline in plasma has been acceptable.

Conclusions

In conclusion, a validated liquid chromatography-tandem mass spectrometry method for the determination of perhexiline and its metabolite cis-hydroxyperhexiline has been described. This method has proven to be rapid, sensitive, specific, accurate, and precise. It is currently being utilized in routine clinical service for monitoring the plasma concentrations of perhexiline and cis-hydroxyperhexiline in patients receiving perhexiline therapy.