As discussed in Section 8, the in vitro inhibition constant (Ki) is not the sole determinant of whether a drug will produce a clinically meaningful or even detectable in vivo effect on a cytochrome P450 (CYP) enzyme under clinically relevant dosing conditions. Instead, it is only one factor in the equation:

Equation 1:

For this reason, the K_i has relatively limited value without knowing or at least having a reasonable estimate of the concentration of the potential inhibitor at the site of action (eg, the CYP enzyme). Parenthetically, the location of the relevant CYP enzyme may be extrahepatic in some instances, such as in the gut wall in the case of the inhibition of the first pass metabolism of terfenadine.

Although some reviews have attempted to rank or list drugs as enzyme inhibitors based on primarily in vitro data or isolated case reports of possible interactions,² this approach does not convey to the clinician whether an interaction is likely to occur in clinical practice or to what extent. Answers to these questions require knowing the concentration of the drug that will occur under clinically relevant dosing conditions as well as the K_i.

This appendix is for the reader who would like more background on this issue from a mathematical perspective, but is not intended to be a rigorous mathematical treatise on the subject. This discussion is relevant to any drug or class of drugs capable of inhibiting a CYP enzyme; however, it will use the SSRIs as the example since they are the subject of this book. The reader who would like further discussion of these matters is referred to the work of Segel³ and von Moltke and colleagues.⁴

For this discussion, we must recall some basic principles of biochemistry, specifically enzymology. The relationship between the inhibition of an enzyme caused by a competitive inhibitor and the concentration of the inhibitor is expressed by the equation:

Equation 2:

i = [I] / ( [I] + K_i (1 + [S] / K_m) )

where “i” is the fractional inhibition of the enzyme, [I] is the concentration of the inhibitor, K_i is the inhibition constant of the inhibitor for the enzyme, [S] is the concentration of the substrate normally biotransformed by the enzyme, and K_m is the affinity constant of that substrate for that enzyme.

If [S] is less than K_m as is the case in vivo for drugs with linear pharmacokinetics,then the fractional inhibition is independent of the substrate concentration and the equation reduces to:

Equation 3:

i = [I] / ( [I] + K_i )

Hence, to the first approximation, the fractional inhibition of the enzyme is determined by the concentration of the inhibitor directly and by the K_i of the inhibitor reciprocally. The smaller K_i is to [I], the more “i” approaches unity:

Equation 4:

i = [I] / [I] (when K_i <<<< [I])

Equation 4 illustrates why both K_i and [I] are important when trying to determine whether a clinically meaningful degree of enzyme inhibition is likely to occur under clinically relevant dosing conditions. Parenthetically, the relative K_i’s in some studies (eg, Crewe et al, 1991) can be outliers relative to the ones derived from other in vitro studies (Table 8.7). If the concentration of the substrate used in the study not being much less than its K_m and not reflective of what would be expected in vivo.

Since all of the SSRIs except possibly fluvoxamine have “active” metabolites in terms of the inhibition of specific CYP enzymes such as CYP 2D6, one must have equation 3 for the parent drug and all of its relevant metabolites when doing projections from in vitro work to in vivo reality. Parenthetically, formal in vivo pharmacokinetic studies measure the summed effect of the parent drug and all of the relevant metabolites at the relative concentrations that they occur under clinically relevant dosing conditions, assuming that the drug is given as it would normally be given and given for a sufficient period of time to reach steady-state conditions.

This last condition has not been true for many of the fluoxetine studies reviewed in Section 8. Instead, loading dose strategies (eg, 60 mg/day × 8 days) have commonly been used due to the long period of time that the study will have to go to reach steady-state conditions of fluoxetine and norfluoxetine. The results of such loading dose strategies may underestimate the actual effect that will occur under more clinically relevant dosing conditions for several reasons. First, norfluoxetine is more potent than fluoxetine in terms of the inhibition of some CYP enzymes such as CYP 3A3/4. Secondly, the conversion of fluoxetine to norfluoxetine takes time. Third, this conversion is inhibited by high concentrations of fluoxetine and/or norfluoxetine. Hence, such loading dose approaches may underestimate the ratio of norfluoxetine to fluoxetine that will be expected under truly steady-state conditions and thus underestimate the concentration of the more portent CYP 3A3/4 inhibitor, norfluoxetine. For these reasons, the results from the loading studies with fluoxetine must be interpreted cautiously and do not reflect the effect that will occur under steady-state conditions at at 20 mg/day, much less 60 mg/day.

The next step in trying to relate in vitro results to in vivo reality is to determine what is the relevant concentration at the CYP enzyme. The concentration of the drug in plasma is the most readily measured concentration; however, the enzyme is not in the plasma, but rather in some tissue compartments. The hepatic tissue compartment is the one which is generally most relevant to predicting whether a pharmacokinetic drug-drug interaction is likely to occur due to the inhibition of a CYP enzyme. For this reason, we need to either measure or estimate the hepatic concentration of the drug which occurs in this compartment under clinically relevant dosing conditions.

Typically, the hepatic concentration is estimated rather than directly measured. It is estimated based on measuring the plasma drug concentration of the drug and its plasma:liver partition coefficient (Table 10.1). The plasma:liver partition coefficient is empirically determined in animals and then can be confirmed in man using autopsy or surgical material.

Table 10.1 illustrates how such information can be used to estimate the relative inhibition of a specific CYP enzyme using such information. Table 10.1 compares the estimated inhibition of CYP 2D6 by 4 different SSRIs. First, the average K_i for each SSRI was calculated using the data from the 5 in vitro studies reported in Table 8.7. Since all 5 of these studies measured the Ki for fluoxetine but not for all the other SSRIs, the average K_i for fluoxetine was determined and used to normalize the K_i values for all the other SSRIs in terms of their relative value compared to fluoxetine. The plasma levels of the different SSRIs which would be expected under comparable antidepressant treatment conditions was then determined. The hepatic concentration of each SSRI which would be expected under such conditions was then estimated based on each drug’s plasma:liver partition coefficient times their expected plasma drug concentration at steady-state under comparable antidepressant treatment conditions. These values and equation 2 above correctly predict the substantial differences in the degree of CYP2D6 inhibition which has been measured in the 11 formal in vivo pharmacokinetic studies reviewed in Section 8 on the relative effects of these SSRIs on CYP 2D6 function. As can be readily seen, a comparison using the K_i alone will lead one to erroneously conclude that there may be minimal differences between these different SSRIs with regard to the inhibition of this enzyme under clinically relevant conditions.

The next issue is how the fractional inhibition of the enzyme relates to the change in the concentration of a concomitantly administered drug which is dependent on the functional integrity of this enzyme for its clearance. In clinical practice, the term “concentration” usually refers to either a 12-hour post-dose concentration or a trough concentration (ie, the concentration immediately before the next dose of drug is given). In both instances, these concentrations are typically measured after steady-state has been attained or assumed to have been attained. A more rigorous measurement is to measure the area under the plasma concentration-time curve (AUC) under steady-state conditions. The latter approach is typically used in formal pharmacokinetic studies. Nonetheless, both approaches have been used to assess the change in the drug concentration as a function of the decrease in its clearance produced by the inhibition of the principal CYP enzyme responsible for the biotransformation necessary for the drug’s elimination. The increase in the AUC with the inhibitor present (AUCi) relative to the AUC without the inhibitor present is related to the fractional inhibition of the enzyme (i) as follows:

Equation 5:

AUCi/AUCo = I/ ( 1-I )

As can be readily appreciated, equation 5 describes a complex hyperbolic relationship (Figure 10.1) between the fractional inhibition of the enzyme and the change in the plasma concentration of the drug whose metabolism has been inhibited. This curve approximates linearity over narrow portions of the curve. As the inhibitions increase, the increase in the plasma concentration of the affected drug increases disproportionately. This fact further explains the differences in clearance of drugs such as desipramine which are observed when fluoxetine and paroxetine are coprescribed at their usually effective minimum antidepressant doses versus citalopram, fluvoxamine, and sertraline (Table 10.1). Fluoxetine and paroxetine produce substantially more than 50% inhibition of the enzyme in contrast to the other 3 SSRIs and thus are on the rapidly ascending portion of this hyperbolic curve.

As stated at the beginning, this discussion has used the effect of the SSRIs on CYP 2D6 to illustrate the enzymological and pharmacokinetic principles relevant to understanding in greater depth the differential effects of drug-induced inhibition of CYP enzyme-mediated clearance of concomitantly prescribed drugs. These principles are relevant to the effects of the SSRIs on other CYP enzymes and to the effects of other drugs on the various CYP enzymes.

FIGURE 10.1

Using the approach explained above, in vitro studies can be used to screen already marketed drugs or drugs in the development for their effects on CYP enzymes and to predict using a knowledge of their K_i, the plasma drug concentration expected under clinically relevant dosing conditions and the plasma:liver partition coefficient whether a clinically meaningful change in the clearance of specific drugs can be expected if they were concomitantly administered with the potential inhibitor. Based on such in vitro modeling, appropriate in vivo studies can be done on a selective basis to confirm such predictions when such an interaction would be predicted to occur in a significant number of patients (ie, the drugs are likely to be frequently coprescribed together) and the consequences are predicted to be clinically important. The advantage of this sequential in vitro and then in vivo approach is the cost and time efficiency of doing in vitro rather than in vivo studies to screen the drugs against a full battery of CYP enzymes.

Given this discussion, readers can anticipate that they will be hearing more and more about the effects of a wide variety of drugs on CYP enzymes. This information will aid the physicians in anticipating pharmacokinetic interactions and will allow them to make appropriate treatment decisions (eg, drug selection, dose adjustments) to avoid adverse consequences of such interactions. The SSRIs and their differential effects on CYP enzymes have been a significant impetus and vehicle for educating physicians about this important advance in our knowledge.

TABLES

Table 2.1 Major Classes of Antidepressants Defined by Principal Mechanisms of Action 16

Table 2.2 Criteria for New Drug Development 19

Table 2.3 The Evolution of Psychopharmacology 20

Table 2.4 TCA (Amitriptyline) Polypharmacy in a Single Pill 25

Table 3.1 Effect of Antidepressants on Serotonin Uptake In Vitro 37

Table 3.2 Effect of Antidepressants on Norepinephrine Uptake In Vitro 38

Table 3.3 Effect of Antidepressants on Dopamine Uptake In Vitro 39

Table 3.4 Relative Potency of the Enantiomers of Citalopram, Fluoxetine and Their Metabolites for Inhibiting the Uptake Pumps for Different Biogenic Amine Neurotransmitters 44

Table 3.5 In Vitro Selectivity Ratio for Different SSRIs and Selected TCAs 46

Table 3.6 Pharmacologic Properties of Antidepressants and Possible Clinical Consequences 48

Table 3.7 Relationship Between Dose, Plasma Level, Potency and Serotonin (5-HT) Uptake 50

Table 3.8 Effect of Uptake Inhibitors and Their Metabolites In Vitro 58

Table 3.9 Effect of Metabolism on the Central MOA and Half-lives of Some SSRIs 60

Table 4.1 STEPS: Factors to Be Considered When Selecting a Medication for a Patient 64

Table 4.2 Safety and Tolerability of TCAs Versus SSRIs 65

Table 4.3 Placebo-adjusted Incidence Rate (%) of Frequent Adverse Effects on Imipramine 68

Table 4.4 Response Rates in Patients With Major Depressive Disorder by Meta-analysis 71

Table 5.1 Common Features of SSRIs With Regard to the Treatment of Major Depression 75

Table 5.2 Comparison of the Placebo-adjusted Incidence Rate (%) of Frequent Adverse Effects for SSRIs 82

Table 5.3 Adverse Events for Each SSRI That Occur ³ 1% More Often Than With Other SSRIs 86

Table 5.4 Placebo-adjusted Incidence (%) of Various Forms of Sexual Dysfunction on Four SSRIs 88

Table 5.5 SSRI Versus Placebo: Response Rate and Relapse Rate 92

Table 6.1 Concentration-dependent Effects of SSRIs 108

Table 6.2 Pharmacokinetic Parameters Relevant to the Use of SSRIs 109

Table 6.3 CYP Enzyme Responsible for Biotransformation of SSRIs 111

Table 6.4 Change in Half-life (t1_2) as a Function of Multiple Dose Administration 118

Table 6.5 Effects of Dose and Age on the Plasma Levels of SSRIs 120

Table 6.6 Effect of Liver Disease on SSRI Metabolism and Pharmacokinetics 127

Table 6.7 Effect of Renal Impairment on Pharmacokinetics of SSRIs (Single Dose) 127

Table 7.1 Unintended Targets of Some SSRIs: CYP Enzymes 131

Table 7.2 History of Our Knowledge of Biotransforming Enzymes 133

Table 7.3 Human CYP Enzymes as Classified by Family, Subfamily and Gene 134

Table 7.4 Two General Classes of CYP Enzymes 134

Table 7.5 What Are the Functions of CYP Enzymes? 136

Table 7.6 Types of Drug Interactions 145

Table 7.7 Pharmacokinetic Interactions: How Do They Present Clinically? 146

Table 7.8 Examples of Metabolically Mediated Pharmacokinetic Drug-drug Interactions and Their Clinical Presentations 152

Table 7.9 Drugs Metabolized by CYP Enzymes 158

Table 7.10 Genetically Determined CYP Enzyme Deficiency 160

Table 8.1 Prevalence of Major Depression in Specific Medically Ill Populations 162

Table 8.2 Concomitant Use of Antidepressant With Other Medications in Different Patient Populations 162

Table 8.3 Could Inhibition of CYP Enzymes Have Other Consequences? 164

Table 8.4 Some SSRIs Lack Selectivity With Regard to Effects on Serotonin Uptake Versus CYP 2D6 165

Table 8.5 Common Myths About CYP Enzymes and SSRIs 166

Table 8.6 Drug-induced Inhibition of CYP Enzymes 167

Table 8.7 The Relative Potency of Five Different SSRIs and Their Metabolites for Inhibiting the Functional Integrity of Three CYP Enzymes 1A2, 2D6 and 3A3/4 Based on In Vitro Studies Using Human Hepatic Microsomes 170

Table 8.8 Effects of Specific SSRIs on Specific CYP Enzymes at Their Usually Effective Antidepressant Dose 176

Table 8.9 In Vivo Studies of Effects of Different SSRIs on CYP 2D6 Function 179

Table 8.10 Relative Potency of the Enantiomers of Fluoxetine and Norfluoxetine for Inhibiting the CYP Enzyme 2D6 180

Table 8.11 Comparison of the In Vivo Effects of Different SSRIs on Specific CYP Enzyme Substrates 193

Table 10.1 Variables Which Determine the Magnitude of the Effect In Vivo: CYP 2D6 as an Example 230

Table 10.2 SSRIs: Brand Names by Country 234

FIGURES

Figure 2.1 Structural Formulas of Several SSRIs 17

Figure 2.2 Schematic Illustration of Relationship Between Drug Site of Action and Effect 19

Figure 2.3 Standard and New Generation Antidepressants Mechanisms of Action 22

Figure 2.4 In Vitro Potency of Amitriptyline as a Representative Tricyclic Antidepressant for Different Sites of Action and Related Mechanisms of Action 27

Figure 3.1 Generic Curve of a Drug's Concentration-dependent Effect on Specific SOA 35

Figure 3.2 Selectivity Ratios for a Series of Uptake Inhibitors Measured In Vitro 41

Figure 3.3 In Vitro Profile of Antidepressants 53

Figure 4.1 Comparative Incidence of Side Effects Between Amitriptyline and Sertraline 70

Figure 5.1 Discontinuation Rate Due to Adverse Events as a Function of Dose for Three SSRIs 84

Figure 5.2 Antidepressant Efficacy as a Function of Dose for Three SSRIs 85

Figure 5.3 Relationship Between Daily Dose of Sertraline, Mean Plasma Levels of Sertraline, and Mean Reduction in Serotonin Uptake by Platelets After 14 Days of Drug Administration at One of Four Fixed Doses 97

Figure 5.4 Estimated Minimum Effective Drug Concentration: Average Plasma Drug Concentration Achieved in the Group Treated With the Minimum, Effective Dose 103

Figure 6.1 Time to Steady-state and Time to 95% Washout 123

Figure 7.1 Drug Metabolism by CYP Enzymes 136

Figure 7.2 CYP Enzyme Reaction Cycle 138

Figure 7.3 Diverse Monooxygenase Activities of CYP Enzymes 139

Figure 7.4 Phases of Xenobiotic Metabolism 140

Figure 7.5 Relationship Between Dosing Rate, Clearance, Steady-state Drug Concentration, and Clinical Response 141

Figure 7.6 Relationship of Pharmacodynamics, Pharmacokinetics and Biological Variance in Determining Overall Result of Drug Treatment 142

Figure 7.7 Dose-response Curves With Risperidone 147

Figure 7.8 Dose-response Curves for Seizure Risk With Clozapine and Bupropion 149

Figure 7.9 Multiple Concentration: Response Curves of Tertiary TCAs 154

Figure 7.10 How Knowledge of Drug-metabolizing Enzymes Will Simplify Understanding of Pharmacokinetic Interactions 156

Figure 8.1 Design of Pharmacokinetic Interaction Studies Using SSRIs as an Example 173

Figure 8.2 Differential In Vivo Effects of Five Different SSRIs on CYP 2D6 Function 185

Figure 8.3 Trough Concentrations of Desipramine in Plasma Correlated With Concentrations of Sertraline Plus Desmethylsertraline in the Sertraline Treatment Group and Fluoxetine Plus Norfluoxetine in the Fluoxetine Treatment Group 188

Figure 8.4 Trough Concentrations of Desipramine in Plasma Correlated With Concentrations of Sertraline Plus Desmethylsertraline in the Sertraline Treatment Group and Paroxetine Plus Norfluoxetine in the Paroxetine Treatment Group 189

Figure 8.5 Relative In Vivo Effects of CYP 3A3/4 Inhibitors on Triazolobenzodiazepines 195

Figure 10.1 Fractional Increase in AUC/ Percent Decrement in Clearance 232