Week 11

• Drug concentration and effect are often linked
• Variability in PK


  • There may be variability in pharmacokinetic parameters between subjects, this will result in differences in the concentration-time profile of the drug between subjects
  • If concentration-time profile is highly variable this will mean that the drug effect is also likely to be highly variable
  • For example, a patient has a faster clearance, drug effect at the end of the dosing interval is near zero, while a patient with lower clearance drug effect will still be roughly 60% of the maximum possible effect
• Variability in PD response


  • The pharmacodynamic response, the effect of drugs, may also vary considerably between people
  • Different patients may have different EC50 and Emax values
  • When pharmacokinetic and pharmacodynamic variability are combined it becomes even more difficult to predict the likely response to a given drug dose
  • When there is significant variability in the pharmacokinetic or pharmacodynamics, or both, a on-dose-fits-all approach may not work. Giving everyone the same dose may mean it will be effective to some but toxic to others
• Dealing with PK variability


  1. Pharmacokinetic variability can be minimised by measuring drug concentrations and adjusting the drug dose to ensure that everybody has similar drug concentrations

  2. This accounts for dose-concentration variability's

  3. This displays a concentration-response relationships

  4. If drug concentration can be controlled if can be seen how individuals respond to same drug concentration. What remains is variability in the pharmacodynamic response, that is the drug effect may still change between individuals when the same drug concentration is achieved
• Approaches to dose individualization


  • Therapeutic drug monitoring (TDM)
  • Target concentration intervention (TCI)
• Therapeutic drug monitoring (TDM)


  • Relies on an established therapeutic range
  • The therapeutic range is the range in which it is desirable to achieve the drug concentrations
  • Drug inefficacy and toxicity can still occur within the therapeutic range
  • This range provides the highest probability of benefits with the lowest probability of side effects
  • It does not mean the drug will be effective in 100% of patients or that 100% of patients will experience side effects
• Target concentrations strategy: "target concentration intervention"


  1. Select target concentration

  2. Predict clearance and volume of distribution based on population PK parameters

  3. Calculate a loading dose and maintenance dose rate to achieve the target concentration

  4. Administer the doses and measure drug concentrations

  5. Use the measured concentrations to predict further individualised values of CL and V

  6. Revise the target concentration based on patients response

  7. Repeat
• How is target concentration identified?


  • A positive number on the y-axis indicates a beneficial effect and a negative number indicates an undesired effect
  • A value of 1 on the y-axis represents 100% chance of efficacy and a value of -1 represents 100% chance of undesired effect
• No, everyone has their own optimal target concentration
• Circumstances where TDM and TCI is appropriate


  • Drug has large between-subject pharmacokinetic variability
  • Therapeutic and adverse effects related to drug concentration
  • Narrow therapeutic index
  • Defined therapeutic (target) concentration range
  • Desired therapeutic effect difficult to monitor
• Circumstances where TDM or TCI is appropriate


  • Suspected drug interactions
  • Suspected drug side effects/toxicity
  • Suspected drug abuse
  • Unexplained failure of therapy
  • Suspected non-compliance
• TDM and TCI should only be done if there is a good reason to, such as lack of effect or suspected toxicity
• To do pharmacokinetic-based monitoring a complete and accurate drug history is needed, including the dosing times, exact times when the drug concentrations were taken, knowledge of the patients current clinical picture, the goal of treatment, and the perceived need for the drug
• If the patient is doing well you may not need to adjust their drug dose
• Drugs pharmacokinetically dose individualised


  • Drugs used prophylactically to maintain the absence of a condition such as seizures, cardiac arrhythmias, depressive or manic episodes, asthma relapses or organ rejection
  • Drugs to avoid toxicity as with the aminoglycoside antibiotics which have a narrow therapeutic window
• Therapeutic target ranges for most drugs


  Derived from observation of therapeutic and adverse effects in small groups of patients, there will be some individuals who achieve adequate effects at lower concentrations or experience adverse effects within the therapeutic range
• Most drug responses are graded responses and are continuous throughout the concentration range. That is the therapeutic response does not magically 'turn on' at the lower limit of the range, nor do toxic responses abruptly materialise at the upper limit
• Individual patients may have a different therapeutic range than the typical or recommended range due to pharmacodynamic variability
• If the clinical effect of a drug can be readily measured

  It is usually better to adjust the drug dose based on response
• Where the drug does not have a measurable effect

  Pharmacokinetic-based dosage individualisation is used
• When to take blood samples

  1. Taking blood samples for concentration monitoring depends on the drugs pharmacokinetics and pharmacodynamics

  2. For most drugs that dosed to steady-state a trough (pre-dose) concentration is often preferred

  3. Taking blood samples too close to the dose may yield a blood concentration that reflects the distribution phase of the drug
• If you sample prior to full distribution of the drug the drug concentration will be unusually high and you may recommend an unnecessary dosage reduction
• Trough concentration


  • For most drugs that are dose individualised, the trough concentration is all that is required
  • The trough has practical advantages in that it is the least variable point in the dosing interval and the next dose can be withheld for a short while, until the blood sample has been taken
  • For drugs that have short half-lives, in relation to the dosing interval, samples should be collected close to the time the next dose is planned, meaning sample is collected pre-dose
  • For drugs that have long half-lives, in relation to the dosing interval, a sample collected in the elimination phase, but not necessarily just before the next dose is planned, may be sufficient
• Non-trough concentration measurement


  Useful where the drug effect is correlated with drug concentration taken at a specific time after the dose
• Methods of PK dose adjustment


  • First principles
  • Computerised methods such as log-linear regression and Bayesian forecasting
• First principles dosage adjustment


  1. Doubling the dose doubles the concentration

  2. New dose = current dose x Target Css / Measured Css

  3. New interval = Current interval x measured Css / target Css
• Drugs appropriate for first principle dosing


  • Have well defined target through concentrations
  • Can only be used in a patient at steady-state with their therapy
• Drugs appropriate for first-principle dosage adjustment


  • Theophylline
  • Lithium
  • Some immunosuppressants
  • Carbamazepine
  • Valproate
• Log-linear regression (LLR)


  • Based on the assumption of a one-compartment distribution model
  • Involves taking two blood samples
  • Linear regression is applied to the natural log values of the concentrations to predict the concentration-time course and from that Cmax and Cmin values are extrapolated
• Log-linear regression


  • Requires that the drug concentration assay is exact
  • Can be used to estimates the AUC after once-daily dosing of aminoglycosides
  • Only appropriate for drugs that display one compartment pharmacokinetics
• Bayesian Forecasting


  • A sophisticated, non-linear regression, method that can predict the entire concentration-time course of a drug from one or two concentrations
  • Preferred method for pharmacokinetic-based dose individualisation when 'first principles' methods are not sufficiently accurate
  • Can be used for nearly any drug as long as a pharmacokinetic model has been previously developed to describe the pharmacokinetic behaviour of that drug across a population of people
• Bayesian Forecasting in clinical practise


  • Requires a population pharmacokinetic model that describes the pharmacokinetics of a drug based in compartmental analysis
  • Models for various drugs are currently available in open source and commercial software programs
  • TDMx is a free open-source program that has population pharmacokinetic models for several antibiotics
• Drugs with a narrow therapeutic index


  • Can be more difficult to dose
  • Patients on these drugs often need to be closely monitored
• Therapeutic drug monitoring (TDM)


  Measurement of drug concentrations in biological fluids and adjusting the drug dose to achieve drug concentrations within a desired target range
• Target concentration intervention


  1. Selecting a specific target concentration for a given individual

  2. Using subsequent pharmacokinetic and pharmacodynamic data on that individual to revise dosage and target concentration predictions
• Adjusting the drug dose to achieve a target concentration range

  Reduces variability in the pharmacokinetic phase of drug action
• Considerations during pharmacokinetic-based dosage adjustment


  • Sample timing and handling
  • The drug assay to use for drug measurement
  • The dose adjustment method to apply
• Drug concentrations need to be interpreted in the context of the clinical features of the patient
• Adequate information about the sample and the patient are required to interpret drug concentrations
• Changes in protein binding


  Can alter the interpretation of total (bound and unbound) drug concentrations