표제지
목차
ABBREVIATIONS 7
제1장 서론 12
1. 배경 12
1.1. 약물동태학의 정의 및 발전 12
1.2. 대표적인 약물동태학 해석 방법 16
1.3. 계량약리학의 정의 20
1.4. 집단 약동학의 정의 및 분석방법 21
1.5. 생물학적 동등성 시험과 통계학적 의의 30
1.6. 고 변동성 약물의 생동성 범위 확장 32
1.7. R을 이용한 약동학 데이터 시뮬레이션 36
2. 연구 목적 40
제2장 재료 및 방법 42
1. 재료 42
1.1. 소프트웨어 (Software) 42
1.2. 가상 약물의 시뮬레이션 데이터 42
1.3. 실제 약물의 임상시험 데이터 46
2. 방법 57
2.1. 가상 약물에 대한 집단 약동학 분석 방법 57
2.2. 실제 약물에 대한 집단 약동학 분석 방법 66
제3장 결과 70
1. 가상 약물 70
1.1. 실험 1: 시나리오 1에 대한 집단 약동학 분석 수행 결과 70
1.2. 실험 2: 시나리오 2에 대한 집단 약동학 분석 수행 결과 74
2. 실제 약물 80
2.1. 에페리손 (Eperisone) 80
2.2. 심바스타틴 (Simvastatin) 84
2.3. 리스페리돈 (Risperidone) 88
2.4. 실제 약물 적용에 대한 종합 결과 92
제4장 결론 95
제5장 참고문헌 98
ABSTRACT 105
APPENDICES 108
Appendix 1. The example of NONMEM code for the virtual drug (IIV:30%, WV:30%) 108
Appendix 2. The example of NONMEM code for Mulex tablet as a reference drug of eperisone 109
Appendix 3. The example of NONMEM code for ZOCOR tablet as a reference drug of simvastatin 111
Appendix 4. The example of NONMEM code for RISPEDAL tablet as a reference drug of risperidone 113
Appendix 5. Goodness of fit plot of eperisone (Mulex tablet as a reference drug) 115
Appendix 6. Goodness of fit plot of simvastatin (ZOCOR tablet as a reference drug) 120
Appendix 7. Goodness of fit plot of risperidone (RISPEDAL tablet as a reference drug) 123
Appendix 8. VPC result of eperisone (Mulex tablet as a reference drug) 126
Appendix 9. VPC result of simvastatin (ZOCOR tablet as a reference drug) 129
Appendix 10. VPC result of risperidone (RISPEDAL tablet as a reference drug) 131
Appendix 11. Bootstrap result of eperisone (Mulex tablet as a reference drug) 133
Appendix 12. Bootstrap result of simvastatin (ZOCOR tablet as a reference drug) 136
Appendix 13. Bootstrap result of risperidone (RISPEDAL tablet as a reference drug) 138
PUBLICATION LIST 142
Table 1. The error models to explain inter-individual variability 27
Table 2. The error models to explain residual variability 29
Table 3. The Example of R-scaled approach as 3x3 BE study design 35
Table 4. Component for generation of simulation scenario 1 43
Table 5. Component for generation of simulation scenario 2 45
Table 6. BE study design (R-scaled approach) for two formulations of eperisone 47
Table 7. Result for BE study of between two eperisone tablets. Geometric menas ratio (GMR), 90% confidence intervals (CI), and within-subject variability for AUCt and Cmax using EMA method... 47
Table 8. BE study design (2 x 2 cross-over) for two formulations of simvastatin 52
Table 9. Statistical analysis of simvastatin pharmacokinetics parameter of ZOCOR tab from R scaled bioequivalence approach (cited from Clinical Pharmacology Biopharmaceutics Review of... 52
Table 10. Analysis of mean squared error, within-subject variability in risperidone bioequivalence studies s (cited from Polymorphisms in CYP2D6 have a greater effect on variability of risperidone... 55
Table 11. The example of R code to simulate population PK data 60
Table 12. Tabulated summary for results of first experiment (IIV=0, no change) 70
Table 13. Tabulated summary for results of second experiment (varied IIV combined with WV) 75
Table 14. Tabulated summary for population PK modeling results of eperisone 81
Table 15. Tabulated summary for population PK modeling results of simvastatin 85
Table 16. Tabulated summary for population PK modeling results of risperidone 89
Table 17. Tabulated summary for population PK analysis of 3 different drugs. 92
Figure 1. Typical pharmacokinetic structure after drug administration (The circle number ①, ②, ③, and ④ are meant absorption, metabolism, distribution and excretion process, respectively.) 13
Figure 2. The example of compartmental PK model. A: PK model structure and differential equations of 1-compartment model, B: PK model structure and differential equations of 2-... 19
Figure 3. The schematic diagram of the relationship among various variabilities of population PK analysis and clinical trial, IIV: Inter-individual variability, IOV: Inter-occasional variability, RV:... 22
Figure 4. Influence of the random and fixed effects on the observed plasma concentrations Cij from the population point of view. The open circle, lower left, is the population mean predicted clearance...[이미지참조] 26
Figure 5. Bioequivalence confidence interval with acceptance criteria ofseveral case 31
Figure 6. HVD's widening of bioequivalence limit based on variability of reference case (a. FDA, b. EMA). (cited from Bioequivalence of Highly Variable Drugs: A Comparison of the Newly... 34
Figure 7. The example of simulation work-flow for mrgsolve as an ODE based R package 38
Figure 8. The example of simulation using by mrgsolve R package 39
Figure 9. The components of variability in NONMEM and clinical trial (IIV: Inter-individual variability, IOV: Inter Occasional variability, RV: Residual variability, WV: Within-subject... 40
Figure 10. Individual PK profiles of group A after administration of reference drug for epersone (Upper: plasma concentration-time curve at first administration, Lower: plasma concentration-time... 48
Figure 11. Individual PK profiles of group B after administration of reference drug for epersone (Upper: plasma concentration-time curve at first administration, Lower: plasma concentration-time... 49
Figure 12. Individual PK profiles of group C after administration of reference drug for epersone (Upper: plasma concentration-time curve at first administration, Lower: plasma concentration-time... 50
Figure 13. Individual PK profiles of reference drug for simvastatin (Upper: plasma concentrationtime curve at period 1 of group 1, Lower: plasma concentration-time curve at period 2 of group 2) 53
Figure 14. Individual PK profiles of reference drug for risperidone (Upper: plasma concentrationtime curve at period 1 of group 1, Lower: plasma concentration-time curve at period 2 of group 2) 56
Figure 15. Overall experiment scheme for virtual drug(A : 1st Experiment; 5 different levels of WV(10%, 20%, 30%, 40%, and 50%) without IIV's change(0%) , B : 2nd Experiment; 5 different levels of WV(10%, 20%, 30%, 40%, and 50%) with IIV's change(10→50%)) CL: Clearance,... 58
Figure 16. The approximation of RV to WV as results of the first experiment for virtual drug (RV: Residual variability, WV: Within-subject variability as a true value) 72
Figure 17. The predictive success rate for the first experiment for virtual drug (Within-subject variability meant true value when generating simulation dataset using by R) 73
Figure 18. The approximation of RV for WV as a result of the second experiment for virtual drug. RV: Residual variability, WV: Within-subject variability, IIV: Inter individual variability 78
Figure 19. The predictive success rate for the second experiment for virtual drug 79
Figure 20. The Ka, CL and Vd as a fixed effect estimated by 1 compartment PK model of eperisone[이미지참조] 82
Figure 21. The sigma and omegas as a random effect estimated by 1 compartment PK model of eperisone 82
Figure 22. The Ka, CL and Vd as a fixed effect estimated by 1 compartment PK model of simvastatin[이미지참조] 86
Figure 23. The sigma and omegas as a random effect estimated by 1 compartment PK moddel of simvastatin 86
Figure 24. The Ka, CL and Vd as a fixed effect estimated by 1 compartment PK model of risperidone[이미지참조] 90
Figure 25. The sigma and omegas as a random effect estimated by 1 compartment PK model of risperidone 90
Figure 26. The relationship between estimated RV and CVwR for 3 different highly variable drugs (eperisone: open circle, simvastatin: closed square, risperidone: closed triangle, absolute match...[이미지참조] 94