Personalized medicine with tyrosine kinase inhibitors: the erlotinib case

Published in: Volume 7 / Year 2013 / Issue 2
Category: Feature – Oncology Drug Treatment
Page: 32-33
Visits: 2034 total, 1 today

Abstract:
Large variability in the pharmacokinetics and pharmacodynamics of tyrosine kinase inhibitors warrants dose stratification rather than the use of fixed doses in all patients. Large population-based studies are required to identify clinically-relevant co-factors that may be used to facilitate dose stratification.

Molecular-targeted therapy with tyrosine kinase inhibitors (TKIs) represents a recent breakthrough in cancer treatment. However, TKIs are mainly administered at a fixed dose, ignoring the potential need for dose individualization. Considering the high inter-individual variability in the pharmacokinetics (PK) of TKIs, a fixed dose may result in suboptimal treatment or excessive toxicity. The PK, toxicity and efficacy of molecular targeted anticancer therapies must be established in order to optimize their prescribed doses. Additionally, wide interpatient variability in the PK of TKIs has been observed [1].

The relationship between certain toxicities and treatment efficacy has been demonstrated. For example, erlotinib may produce a skin rash in some patients which may serve as a surrogate marker of treatment response. However, in clinical practice, the predictive value of skin rash is unclear and is, therefore, of limited use.

Biomarkers have been identified for many drugs and might be useful for determining the optimum dose. Promising alternatives to fixed dosing have been explored. These include therapeutic drug monitoring, genotype- or phenotype-adjusted dosing, and dose modification according to toxicity. However, prospective studies are needed to validate alternate dosing methods in order to facilitate the development of dosing algorithms that may be used to individualize treatment according to patient need. Genotyping and phenotyping are promising strategies for dose optimization although this approach requires validation in large prospective studies. Therapeutic drug monitoring may also have its place in dose optimization, but it requires instrumentation and validation of analytical methods; it is also time consuming and potentially costly.

The tyrosine kinase inhibitor erlotinib requires individualized dosing

Mechanism of action
Erlotinib is a TKI that specifically targets the epidermal growth factor receptor (EGFR) tyrosine kinase, which is highly expressed and occasionally mutated in various forms of cancer. Erlotinib binds in a reversible fashion to the adenosine triphosphate (ATP) binding site of EGFR. For the signal to be transmitted, two membrane receptors of the EGFR family need to combine to form a homodimer. The homodimers then use ATP to autophosphorylate each other whch leads to a conformational change in their intracellular structure. This results in the exposure of a further binding site for proteins, causing a signal cascade to the nucleus. By inhibiting the ATP binding site, autophosphorylation is not possible and the signal is stopped [2].

Indication
Erlotinib is approved for the treatment of locally advanced or metastatic non-small cell lung cancer (NSCLC) after failure of at least one prior therapy. The dose is 150 mg daily two hours before or after a meal.

Pharmacokinetics
Erlotinib exhibits highly variable PK rendering wide variations in plasma concentrations of the same dose in a given population. The oral bioavailability of erlotinib is 80%, but increases with simultaneous food intake. In contrast smoking decreases erlotinib plasma concentrations. Erlotinib is metabolized in the liver by the cytochrome P450 enzyme isoforms, CYP3A4, CYP1A2 and CYP1A1. The main metabolites are biologicallyactive and are excreted in the faeces. Interaction with other drug metabolized by the same cytochrome P450 enzyme subtypes may interact with erlotinib elimination. The elimination half-life is 36 hours meaning that steady state is reached within seven to eight days. These factors point to several reasons for a large inter-individual variation in the PK of erlotinib:

  • Since the absorption and bioavailability of erlotinib is increased by food, it is advised that patients take the drug without food. However, this may be a problem in patients with poor nutritional status (frequently the case) in lung cancer patients.
  • Metabolism by the cytochrome P450 enzyme subtypes, CYP3A4, CYP1A2 and CYP1A1, the activity of which may be variable and determined to a degree by genetic heritage.
  • Interaction with other drugs metabolized by CYP3A4, CYP1A2 and CYP1A1.
  • Presence of biologically-active metabolites, the quantity of which also varies among patients with respect to kinetics and potency.
  • Long elimination half-life and low clearance which is a particular problem in patients with impaired liver function and advanced age.
  • There appears to be a large variation in the volume of distribution of erlotinib.

A high inter-individual variation in PK parameters of erlotinib and other TKIs has been reported previously [1]. Although this would suggest that the efficacy and safety of erlotinib would likely benefit from individualized dosing, the recommended dose is fixed at 150 mg daily.

Population-based PK/PD models
The assessment of population-based PK and pharmacodynamics (PD) is a useful strategy for dose optimization, particularly for cancer drugs which usually have a large inter-individual variability in terms of treatment response and outcome. A population-based PK study allows the identification of characteristics that significantly influence PK parameters, but also takes variance in PD into account. In general, population-based kinetics treat the whole population rather than the individual patient. By doing so, data from many individuals can be analyzed, and a more representative sample of the target population obtained. It is possible not only to describe the mean tendencies in the population, i.e. the typical values, but also to describe the random effects including variability between subjects, between dose occasions and within a subject (residual variability). In this way, it is possible to generate models in which the correlation between relevant population patient characteristics and PK parameters are described for an entire group of patients. Such population-based models may then be used to predict the optimal dosage for individual patients based on the values of given characteristics. This is known as Bayesian forecasting.

To gain further knowledge, plasma concentrations of erlotinib and its main active metabolite, OSI-420, must be assessed in many patients with NSCLC and analyzed by a population-based PK method. The effect of several covariates and single nucleotide polymorphisms in ABCB1, ABCG2, and CYP3A5, as well as the effects of other registered clinical parameters on PK parameters must be evaluated. PK/PD relationships between plasma drug exposure, area under a curve, and skin toxicity should be carefully assessed in a larger population. The non-linear population approach applied to PK data combined with a PK/PD analysis, will reveal cofactors that determine variations in PK related to polymorphisms, as well as outcome and side effects as a basis for more individualized dosing. It will also allow the use of more than one dosage of erlotinib during long-term treatment for NSCLC.

Limitations of sophisticated dosing approaches
Unfortunately, the population-based PK/PD approach does not permit personalized dosing. Rather, it only allows dose stratification. Some patient characteristics can, in some cases, be identified as cofactors of significance that will influence dosing. Examples include age and organ function which can be measured quantitatively, and may provide additional information on optimal dosing. Such methods, however, may include random errors inherent from analysis precision, non-linearity and minor patient-related cofactors not reaching statistical significance with respect to drug turnover and targeting. The dose may, therefore, be stratified with respect to some variables, but far from all. These sophisticated methods of dose stratification are still in their infancy, and are mainly used as research tools. Few have been able to isolate more than a handful of factors that should be considered when modifying the dose.

There are further limitations in the dose optimization of erlotinib. The majority of population-based PK/PD studies have evaluated cytotoxic drugs in order to predict and avoid side effects rather to foresee treatment outcome. The reason is that the PK of cytotoxic drugs can be adequately described; their dependence on statistically-significant patient variables is known which allows the dosage to be tailored. Although the relationship between the PK and PD might be elucidated with respect to side effects occurring in the different tissues of the patient, the most important factor affecting dose optimization in patients with cancer is the tumour itself. The PK of most tumours is different from similar tumours in other patients and is dependent on tumour blood flow and perfusion, transporters for influx or efflux of the drug, internal metabolism and drug capturing, the mechanisms of which are unclear. Further, the quantification of tumour variance is not possible. The same is true for the effect of the drug on cancer cells, which might be of different type, housing different targets and have different binding properties. Knowledge of tumour biology and effects is mostly lacking. Indeed, cancer drugs might be administered according to weight, surface area, by dose banding or at a fixed dose, with similar effects on the tumour.

The large variability in the PK and PD of erlotinib indicates the need for individualized dosing. Thus, the use of a fixed dose of erlotinib based on a small study with a limited number of patients should be reconsidered.

Authors

Professor Per Hartvig-Honoré, PharmD, PhD
Professor in Pharmacokinetics

Khanda Amin, BPharm

Department Pharmacology and Pharmacotherapy
Farma, University of Copenhagen
2 Universitetsparken
DK-2100 Copenhagen, Denmark

Anders Mellemgaard, MD, PhD
Department of Oncology
Herlev Hospital
DK-2100 Copenhagen, Denmark

References

1. KlümpenHJ,SamerCF,MathijssenRH,SchellensJH,GurneyH.Moving towards dose individualization of tyrosine kinase inhibitors. Cancer Treat Rev. 2011 Jun;37(4):251-60. doi:10.1016/j.ctrv.2010.08.006
2. Erlotinib [homepage on the Internet]. c2011 [updated 2012 Nov 27; cited 2012 Dec 12]. Available from: http://en.wikipedia.org/wiki/Erlotinib

Go Back Print

Comments are closed.

Webdesign by I2CT
css.php