Gefitinib is a selective inhibitor of the epidermal growth factor receptor (EGFR) tyrosine kinase, commonly expressed in solid human tumours of epithelial origin. Inhibition of EGFR tyrosine kinase activity inhibits tumour growth, metastasis and angiogenesis and increases tumour cell apoptosis.
Clinical characteristics of never smoker, adenocarcinoma histology and female gender have been shown to be independent predictors of positive EGFR mutation status. Asian patients also have a higher incidence of EGFR mutation-positive tumours.
Most NSCLC tumours with sensitizing EGFR kinase mutations eventually develop resistance to IRESSA treatment with a median time to disease progression of 1 year. In about 60% of cases, resistance is associated with a secondary T790M mutation for which T790M targeted EGFR TKIs may be considered as a next line treatment option. Other potential mechanisms of resistance have been reported following treatment with EGFR signal blocking agents including bypass signalling such as HER2 and MET gene amplification and PIK3CA mutations. Phenotypic switch to small cell lung cancer has also been reported in 5-10% of cases.
The randomised phase III first line IPASS study was conducted in patients in Asia1
with advanced (stage IIIB or IV) NSCLC of adenocarcinoma histology who were ex-light smokers (ceased smoking ≥15 years ago and smoked ≤10 pack years) or never smokers (see Table 1).
China, Hong Kong, Indonesia, Japan, Malaysia, Philippines, Singapore, Taiwan and Thailand.
Click on icon to see table/diagram/image
Pre-planned exploratory analyses were conducted on the biomarker data at the time of the primary analysis. A total of 437 patients had evaluable data for EGFR mutation analysis. PFS was significantly longer for IRESSA than carboplatin/paclitaxel in EGFR mutation positive patients (n=261, HR 0.48, 95% CI 0.36 to 0.64, p<0.0001), and significantly longer for carboplatin/paclitaxel than IRESSA in EGFR mutation negative patients (n=176, HR 2.85, 95% CI 2.05 to 3.98, p<0.0001). Patients were considered EGFR mutation positive if one of 29 EGFR mutations was detected by Amplification Refractory Mutation System (ARMS) using DxS EGFR 29 mutation detection kit. Patients were deemed EGFR mutation negative if samples were successfully analysed and none of the 29 EGFR mutations was detected. PFS results in the subgroup with unknown EGFR-mutation status (hazard ratio with gefitinib, 0.68; 95% CI, 0.58 to 0.81; P<0.0001) were similar to those for the overall population.
In EGFR mutation positive patients, ORR was superior for IRESSA (71.2%) vs carboplatin/paclitaxel (47.3%) (OR 2.751, 95% CI 1.646 to 4.596, p=0.0001). In EGFR-mutation negative patients, ORR was superior for carboplatin/paclitaxel (23.5%) vs IRESSA (1.1%) (OR 0.036, 95% CI 0.005 to 0.273, p=0.0013).
Quality of life outcomes differed according to EGFR mutation status. In EGFR mutation-positive patients, significantly more IRESSA-treated patients experienced an improvement in quality of life and lung cancer symptoms vs carboplatin/paclitaxel (see Table 2).
Click on icon to see table/diagram/image
An analysis of overall survival (OS) was performed after 954 deaths (78% maturity), which demonstrated no statistically significant difference in OS for IRESSA versus carboplatin/paclitaxel in the overall study population (HR 0.901, 95% CI 0.793 to 1.023; p=0.1087). Median OS: IRESSA, 18.8 months; carboplatin/paclitaxel, 17.4 months.
Subgroup analyses of OS by EGFR mutation status showed no significant difference in OS for IRESSA versus carboplatin/paclitaxel in the subgroup of patients with known mutation positive (HR 1.002, 95% CI 0.756 to 1.328; median OS 21.6 months vs.21.9 months) or negative (HR 1.181, 95% CI 0.857 to 1.628; median OS 11.2 months vs. 12.7 months) tumours. The OS outcome in the subgroup of patients with unknown mutation status (HR 0.818, 95% CI 0.696 to 0.962; median OS 18.9 months vs. 17.2 months) was consistent with the overall population.
In the IPASS trial, IRESSA demonstrated superior PFS, ORR, QOL and symptom relief with no significant difference in overall survival compared to carboplatin/paclitaxel in previously untreated patients, with locally advanced or metastatic NSCLC, whose tumours harboured activating mutations of the EGFR tyrosine kinase.
In a phase III double blind clinical trial comparing IRESSA to placebo in patients with advanced NSCLC who had received 1 or 2 chemotherapy regimens, IRESSA did not significantly prolong survival in the overall population (HR 0.89, CI 0.77 to 1.02, p=0.09, Median 5.6 vs 5.1 months for IRESSA and placebo respectively), or in patients with adenocarcinoma (HR 0.84, CI 0.68 to 1.03, p=0.09, Median 6.3 vs 5.4 months for IRESSA and placebo respectively). IRESSA also did not produce any significant benefits in terms of quality of life or symptom control. Pre-planned subgroup analyses showed a statistically significant increase in survival for patients of Oriental ethnicity treated with IRESSA compared to placebo (HR=0.66, CI 0.48 to 0.91, p=0.01, Median 9.5 vs 5.5 months), and for patients that had never smoked treated with IRESSA compared to placebo (HR=0.67, CI 0.49 to 0.92, p=0.01, Median 8.9 vs 6.1 months).
In patients with high EGFR copy number (n=114) there was a non-significant trend towards improved survival in patients receiving IRESSA compared to those receiving placebo (HR 0.61, 95% CI 0.36 to 1.04, p=0.067). Analysis of EGFR gene copy number was measured by fluorescence in situ hybridisation (FISH) using the LSI EGFR SpectrumOrange/CEP 7 SpectrumGreen probe. Patients were considered to have high EGFR gene copy if their tumour had high polysomy (≥4 copies in ≥ 40% of cells) or gene amplification (presence of tight EGFR gene clusters and a ratio of gene/chromosome per cell ≥2, or ≥15 copies of EGFR per cell in ≥10% of analysed cells).
Analysis of EGFR protein expression data showed that EGFR positive patients had no significant survival benefits when treated with IRESSA compared to placebo (N=264, HR=0.77; CI 0.56 to 1.08, p=0.13) than EGFR negative patients (N=115, HR=1.57; CI 0.86 to 2.87, p=0.14), though neither of the subset analyses were statistically significant. EGFR status was not tested for the majority of patients. For patients in whom EGFR status was not tested (N=1313, HR=0.84; CI 0.73 to 0.98, p=0.03), the HR was similar to that seen in the overall study population as would be expected. A positive EGFR expression status was defined as having at least 10% of cells staining for EGFR in contrast to the 1% cut-off specified in the DAKO EGFR pharmDx kit instructions.
For patients with an EGFR gene mutation, there was insufficient data to allow a meaningful evaluation of survival.
The overall trial data suggest that the patients most likely to benefit from treatment with IRESSA are patients of Oriental ethnicity or patients who have never smoked.
The IFUM study was a single arm, multicentre study conducted in Caucasian patients (n=106) with activating, sensitizing EGFR mutation positive NSCLC to confirm that the activity of gefitinib is similar in Caucasian and Asian populations. The ORR accordingto investigator review was 70% and the median PFS was 9.7 months. These data are similar to those reported in the IPASS study.
Circulating Tumour DNA (ctDNA): In the IFUM trial, mutation status was assessed in tumour and ctDNA samples derived from plasma, using the Therascreen EGFR RGQ PCR kit (Qiagen). Both ctDNA and tumour samples were evaluable for 652 patients out of 1060 screened. The sensitivity of EGFR mutation testing in ctDNA using the Qiagen Therascreen EGFR RGQ PCR kit was 65.7%, with a specificity of 99.8%. The positive and negative predictive values of ctDNA were 98.6% and 93.8%, respectively (Table 3). Objective response rate in the IFUM full analysis set of patients who were tDNA positive was 69.8% (95% CI: 60.5% to 77.7%). The ORR in those patients who were tumour and ctDNA mutation positive was 76.9% (95% CI: 65.4% to 85.5%) and in those who were tumour only mutation positive but ctDNA negative 59.5% (95% CI: 43.5% to 73.3%). (See Table 3.)
Click on icon to see table/diagram/image
These data are consistent with the pre-planned exploratory Japanese subgroup analysis in IPASS. In that study ctDNA derived from serum, not plasma was used for EGFR mutation analysis using the EGFR Mutation Test Kit (DxS) (N= 86). In that study, concordance was 66%, sensitivity was 43.1 %, specificity was 100%. The positive and negative predictive values were 100% and 54.7%, respectively.
Following intravenous administration, gefitinib is rapidly cleared, extensively distributed and has a mean elimination half-life of 48 hours. Following oral dosing in cancer patients, absorption is moderately slow and the mean terminal half-life is 41 hours. Administration of gefitinib once daily results in 2 to 8-fold accumulation with steady state exposures achieved after 7 to 10 doses. At steady state, circulating plasma concentrations are typically maintained within a 2 to 3-fold range over the 24-hour dosing interval.
Following oral administration of IRESSA, peak plasma concentrations of gefitinib typically occur at 3 to 7 hours after dosing. Mean absolute bioavailability is 59% in cancer patients. Exposure to gefitinib is not significantly altered by food. In a trial in healthy volunteers where gastric pH was maintained above pH 5, gefitinib exposure was reduced by 47% (see 'PRECAUTIONS' and 'INTERACTIONS').
Mean volume of distribution at steady state of gefitinib is 1400L indicating extensive distribution into tissue. Plasma protein binding is approximately 90%. Gefitinib binds to serum albumin and α1-acid glycoprotein.
Metabolism: In vitro
data indicate that CYP3A4 is the major P450 isozyme involved in the oxidative metabolism of gefitinib.
studies have shown that gefitinib has limited potential to inhibit CYP2D6. In a clinical trial in patients, gefitinib was co-administered with metoprolol (a CYP2D6 substrate). This resulted in a small (35%) increase in exposure to metoprolol, which is not considered to be clinically relevant.
Gefitinib shows no enzyme induction effects in animal studies and no significant inhibition (in vitro
) of any other cytochrome P450 enzyme.
Three sites of biotransformation have been identified in the metabolism of gefitinib: metabolism of the N-propylmorpholino-group, demethylation of the methoxysubstituent on the quinazoline and oxidative defluorination of the halogenated phenyl group. Five metabolites have been fully identified in faecal extracts and the major component was O-desmethyl gefitinib, although this only accounted for 14% of the dose.
In human plasma 8 metabolites were fully identified. The major metabolite identified was O-desmethyl gefitinib, which was 14-fold less potent than gefitinib at inhibiting EGFR stimulated cell growth and had no inhibitory effect on tumour cell growth in mice. It is therefore considered unlikely that it contributes to the clinical activity of gefitinib.
The production of O-desmethyl gefitinib has been shown, in vitro
, to be via CYP2D6. The role of CYP2D6 in the metabolic clearance of gefitinib has been evaluated in a clinical trial in healthy volunteers genotyped for CYP2D6 status. In poor metabolisers, no measurable levels of O-desmethyl gefitinib were produced. The range of gefitinib exposures achieved in both the extensive and the poor metaboliser groups were wide and overlapping but the mean exposure to gefitinib was 2-fold higher in the poor metaboliser group. The higher average exposures that could be achieved by individuals with no active CYP2D6 may be clinically relevant since adverse experiences are related to dose and exposure.
Gefitinib total plasma clearance is approximately 500 mL/min. Excretion is predominantly via the faeces with renal elimination of drug and metabolites accounting for less than 4% of the administered dose.
In population based data analyses in cancer patients, no relationships were identified between predicted steady state trough concentration and patient age, body weight, gender, ethnicity or creatinine clearance.
In a phase I open-label study of single dose gefitinib 250 mg in patients with mild, moderate or severe hepatic impairment due to cirrhosis (according to Child-Pugh classification), there was an increase in exposure in all groups compared with healthy controls. An average 3.1-fold increase in exposure to gefitinib in patients with moderate and severe hepatic impairment was observed. None of the patients had cancer, all had cirrhosis and some had hepatitis. This increase in exposure may be of clinical relevance since adverse experiences are related to dose and exposure to gefitinib.
Gefitinib has been evaluated in a clinical trial conducted in 41 patients with solid tumours and normal hepatic function, or moderate or severe hepatic dysfunction due to liver metastases. It was shown that following daily dosing of 250 mg IRESSA, time to steady state, total plasma clearance and steady state exposure (Cmax,ss
) were similar for the groups with normal and moderately impaired hepatic function. Data from 4 patients with severe hepatic dysfunction due to liver metastases suggested that steady state exposures in these patients are also similar to those in patients with normal hepatic function.
Toxicology: Preclinical Safety Data Relevant to the Prescriber:
Gefitinib showed no genotoxic potential.
There was, as expected from the pharmacological activity of gefitinib, a reduction in female fertility in the rat at a dose of 20 mg/kg/day. When administered during organogenesis, there were no effects on rat embryofetal development at the highest dose (30 mg/kg/day), however in the rabbit, there were reduced foetal weights at 20 mg/kg/day and above. There were no compound induced malformations in either species. When dosed to the rat throughout gestation and parturition, there was a reduction in pup survival at a dose of 20 mg/kg/day (see 'USE IN PREGNANCY & LACTATION').
Following oral administration of carbon-14 labelled gefitinib to rats 14 days post partum, concentrations of radioactivity in milk were higher than in blood (see 'USE IN PREGNANCY & LACTATION').
Data from non-clinical (in vitro
) studies indicate that gefitinib has the potential to inhibit the cardiac action potential repolarization process (e.g. QT interval). Clinical experience has not shown a causal association between QT prolongation and gefitinib.
A 2 year carcinogenicity study in rats resulted in a small but statistically significant increased incidence of hepatocellular adenomas in both male and female rats and mesenteric lymph node haemangiosarcomas in female rats at the highest dose (10 mg/kg/day) only. The hepatocellular adenomas were also seen in a 2 year carcinogenicity study in mice, which demonstrated a small increased incidence of this finding in male mice dosed at 50 mg/kg/day, and in both male and female mice at the highest dose of 90 mg/kg/day (reduced from 125 mg/kg/day from week 22). The effects reached statistical significance for the female mice, but not for the males. The clinical relevance of these findings is unknown.