The use of erlotinib in daily practice: a study on adherence and patients' experiences

Adherence to oral pharmacological therapy is a complex and multifactorial issue that can substantially alter the outcome of therapy [6, 7]. Adherence (synonymous with compliance) has recently been defined by the International Society for Pharmacoeconomics and Outcome Research (ISPOR) as the extent to which a patient acts in accordance with the prescribed interval and dose of a dosing regimen [8]. A patient is optimally adherent if no doses are missed, no extra doses are taken, and no doses are taken in the wrong quantity or at the wrong time. Adherence is measured over a period of time and reported as the adherence rate, which is the percentage of dose taken in relation to what was prescribed [9]. There are several methods to measure adherence, including self reports, pill counts, electronic monitoring systems, analyses of pharmacy dispensing databases or assessment of blood or urine samples. There is no golden standard measurement and all methods have limitations [9, 10]. The major limitation of measuring adherence is the so called Hawthorne effect: the measuring of adherence itself influences the adherence because the awareness of patients that adherence is being measured may influence their behaviour. Adherence rates for many chronic drug therapies have been shown to range between 35-70% [11]. The consequences of poor adherence are poor health outcomes and increased health care costs [6].

Cancer patients are generally thought to have higher adherence rates than other patients because they are highly motivated by the gravity of their disease [12, 13]. However, it has been shown that cancer patients have similar adherence rates to those of patients with other diseases [9, 14, 15]. Treatment duration plays a role in the adherence to the regimen. When the medication is continued over a longer period of time, patients become less adherent [16].

In oncology adherence has been studied mainly in two subpopulations, both using long-term medication. In the first population, breast cancer patients on adjuvant hormonal therapy like tamoxifen the adherence rate has been subject of several studies [9]. The reported adherence rates range from 50% to 98% [13, 16, 17]. Several studies concerning adherence to oral medication have been published in the second subpopulation, patients with Chronic Myeloid Leukemia (CML) [15, 18, 19]. Non adherence appear to be associated with poorer response to imatinib [15, 18]. Noens et al have shown that patients with suboptimal response had higher percentages of imatinib not taken (23%) than those with optimal response (7%). Marin et al have demonstrated in CML that there was a strong correlation between adherence rate and response. Adherence was the only critical factor for achieving molecular response.

Another frequently overlooked problem is over-adherence. This may be more an issue in oncology patients than in other patients and may, in the case of oral anticancer agents, lead to substantially increased toxicity [10]. In the study by Nilsson et al. [14], 30% had an oversupply of their medication.

Oncologists may not always consider the issue of adherence. Yet, suboptimal adherence may prove to be the greatest barrier to the effective use of new oral agents [9, 18].

Several factors have shown to be associated with non-adherence. These include patient variables like demographic factors, patients beliefs, disease variables, various aspects of treatment regimes, side- effects and quality of life.

The common sense model of self-regulation (CSM) developed by Leventhal et al [20] is a commonly used theoretical model to understand patients perceptions of illness. According to the CSM peoples' perception of and beliefs about their illness is an important factor in their reactions and behaviour to health threats. Illness perceptions can easily be measured with the Brief Illness Perceptions Questionnaire (Brief IPQ) [21].

Another factor which may influence adherence is patients beliefs about the medication. These representations are usually determined with the Beliefs about Medicine Questionnaire (BMQ) [22]. Associations between adherence and BMQ scores have been shown[23‐27].

Few studies have focused on patients' reasons for not adhering to oral anticancer agents. Eliasson et al [28] have reported that finding ways to deal with the side-effects leads to better adherence in patients with CML. In clinical trials adverse events are generally reported by clinicians, e.g. the National Cancer Institute's Common Terminology Criteria for Adverse Events (CTCAE). Basch et al have shown the value of patients' reported adverse symptom events in cancer patients, which better reflect their daily health status [29, 30].

Erlotinib (Tarceva^®) is an oral anticancer agent, which is registered for the second- and third-line treatment of NSCLC since 2006 [31] Erlotinib is a tyrosine kinase inhibitor (TKI) of epidermal growth factor receptor (EGFR-TKI).

Diarrhea and skin rash are the most common side-effects of erlotinib. The incidence and severity of diarrhea seems to be related to the schedule and dose of erlotinib [32]. Rudin et al [33] have not observed a relationship between the Area Under the Curve (AUC) concentration and diarrhea. However, they did find a relationship between trough plasma concentration (C_trough) of erlotinib and skin rash. Hidalgo et al [32] also detected a relationship between AUC and skin rash. Interestingly, patients who experienced skin rash had significantly longer survival [34, 35].

Hidalgo et al [32] have suggested that an erlotinib plasma concentration of 0.5 μg/ml would be sufficient for relevant antiproliferative activity. Peak concentrations (Cmax) of erlotinib of 1.737 ± 0.777 μg/ml are reached 2.44 ± 1.94 hour (Tmax) after administration of 150 mg a day. Erlotinib has a plasma volume of 233 L and is for 95% bound to plasma proteins like albumine and α1-acid-glycoprotein. The plasma elimination half-life is approximately 36 hours [36]. Erlotinib is mainly metabolized by the liver (90%), only a small part is excreted by the kidneys (9%). It is primarily metabolized by cytochrome P450 3A4 (CYP3A4) and in a lesser extent by CYP1A2, It is a substrate of the multi-drug transporter ATP-binding cassette transporter ABCG2 [37].

Several drug interactions, smoking and food may influence plasma concentrations of erlotinib. Drugs that increase the pH of the upper gastrointestinal tract may alter the solubility of erlotinib, thereby reducing its bioavailability. Inhibitors or inducers of CYP3A4 and CYP1A2 influence the metabolism of erlotinib, which may result in increased or decreased plasma concentrations [31] Smoking results in a significantly lower plasma concentration [38]. When erlotinib is taken together with food the bioavailability of erlotinib increases almost two-fold [39]. It is therefore advised by the manufacturer to take erlotinib at least one hour before or two hours after a meal [31]. Serum total bilirubin, serum α1-acid glycoprotein, serum albumin, creatinine clearance and gender are significant co-variables that explain interindividual variability for clearance [36].

The primary objective is to get insight into the relationship between adherence to erlotinib and the plasma concentration of erlotinib and the relationship between adherence and side-effects in patients with NSCLC. Furthermore, the relationships between patient characteristics, disease characteristics, side-effects, quality of life, patient beliefs and attitude towards disease and medication, adherence, dose adjustments and plasma concentration of erlotinib in patients with NSCLC will be explored.

A multicenter, prospective observational cohort study in which patients starting treatment with erlotinib will be followed for a period up to 16 weeks. At baseline and in weeks 3-4, 8-9, 12 and 15-16, patients will be asked to fill out a questionnaire. Furthermore, in weeks 3-4, 8-9 and 15-16, blood samples are collected, which will be analysed for plasma concentration of erlotinib. Adherence will be measured using a medication event monitoring system (MEMS). Information from the physician will be retrieved from the patient treatment file.

The protocol was approved by the Medical Ethics review board of the VU University Medical Center, Amsterdam, The Netherlands.

Patients will be recruited by the Departments of Pulmonalogy from twelve hospitals in the Netherlands. After the patient and doctor have decided on treatment with erlotinib, patients will be informed about the study and will receive written information to take home. Within two days the researcher will contact the patient by phone to further inform the patient of the study and to ask for his/her consent. If the patient decides to participate in the study, he or she will sign the informed consentform.

Patients with advanced NSCLC under treatment in one of the participating hospitals starting with the use of erlotinib can be included. Patients younger than 18 years and those with insufficient knowledge of the Dutch language will be excluded. According to the sample size calculation (see Statistics) a number of 65 patients will be enrolled.

Several methods are used to determine the variables. Most information is collected directly from patients: they will fill out questionnaires, will use an electronic monitoring pill-container and will donate blood samples. Furthermore, medical information will be retrieved from the patients' file.

The data will be described by means of frequency analysis. Correlation coefficients will be calculated for relationships between adherence, patient characteristics, disease characteristics, side effects, quality of life, patient beliefs and attitude towards disease and medication, dose adjustments and plasma concentration of erlotinib in patients with NSCLC. In particular, the correlation coefficient between plasma concentration and side-effects; the relationship between dose and plasma concentration in daily practice; the influence of different factors on adherence; the influence of different factors on side-effects; the influence of different factors on quality of life; the influence of different factors on dose adjustment and discontinuation. By the use of a multivariate regression analysis the influence of adherence (dependent variable) on plasma concentration and the influence of side-effects (dependent variable) on adherence will be calculated. The dependent variables will be adjusted for potential confounders. For all analyses a two-tailed significance level will be 0.05, all p values below this level will be considered statistically significant. The statistical analyses will be performed with SPSS 15.0 (SPSS Inc. Chicago, Illinois, USA).