Study On The Individualized Medication Of Paroxetine Based On Population Pharmacokinetics And Therapeutic Drug Monitoring

ObjectiveAccording to WHO data in 2022,322 million people suffer from depression worldwide,and the onset of depression tends to be younger.The treatment of depression is usually based on psychological plus pharmacotherapy,and paroxetine is currently the first-line antidepressant.Studies have found that after repeated administration,the pharmacokinetics of paroxetine shows nonlinearity with the increase in dose and duration of treatment.The domestic studies have found that nearly 80.41%of the patients’serum concentration of paroxetine in of the population is outside the predicted concentration range(20-65 ng·m L^-1),suggesting that the reference range is not consistent with the actual clinical application.The combination of therapeutic drug monitoring and the population pharmacokinetic（PPK）are helpful to clarify the reasons for the high interindividual variation of paroxetine,and to provide a reference for dose adjustment and administration regimen optimization.MethodsThe serum concentrations,demographic data,medication information,clinical biochemical indexes,and CYP2D6 genotype of patients with mental disorders treated with paroxetine in the psychiatric hospital from 2019 to 2021 were retrospectively collected.SPSS 25.0 software was used for the statistical analysis of therapeutic drug monitoring data.Firstly,the variables were investigated under the overall dose by single factor analysis,then the variables with differences were screened.Then multiple linear regression analysis was performed on the variables with difference.Population pharmacokinetics（PPK）analysis of paroxetine serum concentration data was performed based on NONMEM.The effect of covariates such as age,weight,height,BMI,daily dose,liver function index（aspartate aminotransferase,alanine aminotransferase,aspartate aminotransferase isoenzyme,total bilirubin,direct bilirubin,total protein,serum albumin）,renal function index（serum creatinine,blood urea nitrogen）,glucolipid metabolism（total cholesterol,and prolactin）was investigated by forward inclusion and backward elimination methods.The final model was obtained.The objective function value（OFV）,bootstrap method,normalized prediction distribution error（NPDE）,and goodness-of-fit（GOF）plot were used to evaluate the PK parameters of the final model and verify the stability and prediction performance of the model.Based on the estimated parameters of the final model and the commonly used clinical dosing regimens,the serum concentration of paroxetine under different dosing scenarios were simulated to provide insights into individualized dosing regimens.ResultsThe steady-state serum concentration data of 173 samples with paroxetine were included in the analysis of dose correction concentration（C/D）influencing factors.The potential influencing factors of serum concentration and C/D of paroxetine were analyzed from the perspective of overall dose and individual dose.At the overall dose,the daily dose and the serum concentration were positively correlated.Among173 TDM samples,41.2%of the serum samples were within the therapeutic reference range(20-65 ng·m L^-1),and 21.39%exceeded the laboratory warning concentration level(120 ng·m L^-1).The average C/D of paroxetine was 2.14 d·L^-1,82.08%of the patients were above the upper limit of the reference range(0.37-0.83d·L^-1),and only 16.18%of the patients fell within the predicted reference range.Single factor analysis found that gender,formulations,total bilirubin（TBIL）,and direct bilirubin（DBIL）had significant differences on the concentration of paroxetine（p<0.05）.The results of multiple linear regression showed that the serum concentration of paroxetine was positively correlated with the log of the daily dose of paroxetine（X1）,gender（X2）,and formulations（X3）,and negatively correlated with TBIL（X4）.The regression equation obtained was y=-388.631+287.931X1+32.885X2+44.532X3-1.819X4.The results of individual doses showed that there was a significant difference between sex at 40 mg and 50mg（p<0.05）,and TBIL at 20 mg（p<0.05）.Then a total of 372 serum drug concentration data from 184 patients with mental disorders were included for PPK analysis.A one-compartment model and first-order elimination could fit the PKs characteristics of paroxetine.The population typical values of apparent volume of distribution（V/F）and apparent clearance（CL/F）of the final model were determined to be 8850 L and 21.2 L·h^-1,respectively.Dose,formulations,and gender had a significant effect on paroxetine PK parameters.According to the final model,for every 10 mg increase in the dose,CL/F decreased1-2 times.Compared with sustained-release tablets,the V/F of the tablets was reduced by 66.6%.Gender affected bioavailability（F）,which was 47.5%lower in males than in females.The Bootstrap plot,NPDE plot,and GOF plot indicated that the prediction performance and stability of the model were good.The simulation results showed that:（1）When taking daily doses of paroxetine 20 to 60 mg,the serum concentrations ranged from 7.87 to 230.74 ng·m L^-1 in male patients and 10.22to 336.4 ng·m L^-1 in female patients.When taking daily doses of 20 to 40 mg·d^-1,the mean serum concentrations of paroxetine were within the laboratory alert level(120ng·m L^-1).（2）The observed variation at higher doses was significant between formulations,with the steady-state trough concentration of paroxetine tablets being much higher than that of paroxetine sustained-release tablets.（3）The results showed that the steady-state serum concentration of Quaque Die（QD）administration was lower than that of Quaque Nocte administration and Bis In Die administration.（4）The drug withdrawal scenario was simulated.The results showed that for the low dose of 20 mg,the blood drug concentration was relatively flat when the dose was reduced by one-quarter of the fixed-dose at a fixed interval of 3 or 7 days.For the middle dose of 40 mg,the dose was tapered at a fixed quarter at fixed intervals of 7or 14 days,and the serum concentration fluctuated more gently.Fluctuations in the serum concentration were more gradual for the high dose of 60 mg,followed by tapering at a fixed dose of 10 mg or less at fixed intervals of 14 days or more.ConclusionsThe study found that the dose correction concentration effect factor analysis did not reflect the actual situation of the expression of paroxetine concentration differences.The population pharmacokinetic model based on paroxetine TDM data characteristically describes the nonlinear increase of paroxetine concentration and quantifies the degree to which significant influencing factors（dose,formulations,and gender）affect the PKs of paroxetine among the evaluated variables.Model simulation is used to explain the high variability affecting patients and provide insights into individualized patient treatment regimens.Therefore,the followings are recommended for individualized paroxetine dosing regimens:（1）Within the AGNP therapeutic reference range,taking the conventional doses,male patients require higher doses than females to achieve similar drug concentrations.（2）To maintain good therapeutic efficacy,it is recommended that paroxetine tablets be administered at no greater than 30 mg·d^-1 in female patients and 40 mg·d^-1 in male patients.（3）If the dose of paroxetine is more than 40 mg per day,the blood concentration of paroxetine should be closely monitored in combination with clinical efficacy.（4）To avoid potential adverse effects from higher doses,formulation switching may be considered.For example,when women taking a 50mg sustained-release tablet are switched to the lower dose of 40 mg tablet,similar concentration levels were achieved.（5）A QD dosing regimen is recommended for paroxetine tablets and extended-release tablets and suitable for TDM.（6）Longer dose reduction intervals and lower doses（microdose）should be chosen as a stopping regimen as the dose increases.