Method for the analysis of perhexiline and its hydroxy metabolite in plasma using high-performance liquid chromatography with precolumn derivatization

Determination of perhexiline and hydroxyperhexiline in plasma by liquid chromatography–mass spectrometry

A method for the quantitative determination of perhexiline and its main hydroxylated metabolites in human plasma, based on liquid chromatography-mass spectrometry (LC-MS), was developed. The method used protein precipitation with acetonitrile followed by dilution with water and subsequent direct injection of the extract into the LC-MS system. Hexadiline was used as internal standard and the intra-assay coefficients of variation were <or=5% for perhexiline and cis-hydroxyperhexiline over the target concentration range in patients. The lower limits of quantification were 0.005mg/l for perhexiline and 0.015mg/l for cis-hydroxyperhexiline, and the measuring ranges were from 0.05 to 3.0 and from 0.2 to 6.0mg/l, respectively. The method was compared with an established HPLC method with fluorescence detection and the correlation between the methods was close to 1 for both compounds. The predominant form of hydroxyperhexiline in 87% of the patient samples was found to be one of the diastereomeric pairs of cis-hydroxyperhexiline. In patients not forming this metabolite, trans-hydroxyperhexiline could be detected. We conclude that the present LC-MS method is suitable for use in a clinical routine laboratory.

Correlation of CYP2D6 genotype with perhexiline phenotypic metabolizer status

Perhexiline is metabolized by CYP2D6 and has concentration-related hepatoxicity and peripheral neuropathy. The risk of toxicity is reduced using therapeutic drug monitoring. CYP2D6 genotyping before therapy may allow earlier appropriate dosing. This study aimed to determine whether assessment of CYP2D6 genotype in patients on perhexiline could predict accurately metabolizer status as determined by the perhexiline metabolic ratio (MR). Blood samples from patients stabilized on perhexiline were analysed for CYP2D6 genotype and for concentrations of perhexiline and its hydroxy metabolite. The MR was determined. Of 74 patients, five were poor metabolizers (PM) defined by a MR<0.4, and the remainder were extensive metabolizers (EM). The genotypes were: *1/*1 (n=21), *1/*4 (n=18), *1/*2 (n=12), *1/*3 (n=2), *1/*5 (n=1), *1/*9 (n=2), *1/*10 (n=2), *2/*4 (n=4), *2/*2 (n=3), *4/*41 (n=3), *2/*41 (n=1), *41/*41 (n=1), *4/*9 (n=1), *4/*5 (n=1), *5/*6 (n=1) and *4/*6 (n=1). Allele frequencies were consistent with those reported in population studies. The 3 PMs with the lowest MR were predicted by genotype (*4/*5, *5/*6, *4/*6). The other 2 PMs had intermediate metabolizer genotypes and were on CYP2D6 inhibiting drugs. Amongst the EMs, the highest MR was associated with *1 and *2 allele combinations and the MR was progressively lower with the presence of alleles with intermediate function (*9, *10, *41) followed by alleles with no functional product (*3, *4, *5, *6). Thus, a gene-dose effect was observed. Genotype predicted PM phenotype and also intermediate metabolizers. Determination of CYP2D6 genotype before therapy with perhexiline may help predict perhexiline dose requirements and reduce the risk of perhexiline concentration-related toxicity.

Assessment of interlaboratory performance in the provision of perhexiline therapeutic drug monitoring services in Australia

Perhexiline is a prophylactic antianginal agent particularly useful in patients whose angina is poorly controlled or refractory to conventional drug regimens. Although perhexiline can cause serious hepatic and neurological toxicity, maintaining trough plasma concentrations between 0.15-0.60 mg/L minimizes the risk of toxicity while providing relief of angina symptoms in a majority of patients. All pathology laboratories are required to participate in interlaboratory proficiency testing (PT) programs. The authors therefore initiated a monthly PT program to assess the performance of Australian laboratories measuring perhexiline (n = 8). PT specimens included perhexiline-spiked drug-free human plasma and pooled plasma from patients administered perhexiline. The performance of 8 Australian laboratories participating in the program was examined over a 30-month period. The mean relative standard deviation of the group was 18.2%. All centers performed well with respect to accuracy, achieving mean percentage bias within +/-8% of target perhexiline concentrations. The usefulness of the PT program was highlighted by the identification of two laboratories with an unacceptable degree of variability (up to 30% of results varied more than +/-55% from the target concentration), and the identification of potential analytical problems with the use of perhexiline metabolite concentrations for determining patients' hydroxylator status. Continued and improved use of PT by pathology laboratories is essential to ensuring the safe and effective clinical use of perhexiline.