FDA Drug Approval Summary: Gefitinib (ZD1839) (Iressa®) Tablets

On May 5, 2003, gefitinib (Iressa®, ZD1839) 250-mg tablets received accelerated approval by the U.S. Food and Drug Administration as monotherapy treatment for patients with locally advanced or metastatic non-small cell lung cancer (NSCLC) after failure of both platinum-based and docetaxel chemotherapies. Information provided in this summary includes efficacy and safety results of relevant clinical trials. Effectiveness was demonstrated in a randomized, double-blind, phase II, multicenter trial comparing two oral doses of gefitinib (250 mg/day versus 500 mg/day). Two hundred sixteen patients were enrolled. The 142 patients who were refractory to or intolerant of a platinum and docetaxel comprised the evaluable population for the efficacy analysis. A partial tumor response occurred in 14% (9 of 66) of patients receiving gefitinib 250 mg/day and in 8% (6 of 76) of patients receiving gefitinib 500 mg/day. The overall objective response rate for both doses combined was 10.6% (15 of 142 patients) (95% confidence interval 6.0%-16.8%). Responses were more frequent in females and in nonsmokers. The median duration of response was 7.0 months (range 4.6-18.6+ months). Other submitted data included the results of two large trials conducted in chemotherapy-naive, stage III and IV NSCLC patients. Patients were randomized to receive gefitinib (250 mg or 500 mg daily) or placebo, in combination with either gemcitabine plus cisplatin (n = 1,093) or carboplatin plus paclitaxel (n = 1,037). Results from those studies showed no benefit (response rate, time to progression, or survival) from adding gefitinib to chemotherapy. Consequently, gefinitib is only recommended for use as monotherapy. Common adverse events associated with gefitinib treatment included diarrhea, rash, acne, dry skin, nausea, and vomiting. Most toxicities were Common Toxicity Criteria grade 1 or 2. Interstitial lung disease (ILD) has been observed in patients receiving gefitinib. Worldwide, the incidence of ILD is about 1% (2% in the Japanese postmarketing experience and about 0.3% in a U.S. expanded access program). Approximately one-third of the cases were fatal. Physicians should promptly evaluate new or worsening pulmonary symptoms. If ILD is confirmed, appropriate management includes discontinuation of gefitinib. Gefitinib was approved under accelerated approval regulations on the basis of a surrogate end point response rate. No controlled gefitinib trials, to date, demonstrate a clinical benefit, such as improvement in disease-related symptoms or greater survival. Accelerated approval regulations require the sponsor to conduct further studies to verify that gefitinib therapy produces such a benefit. The Oncologist 2003;8:303-306 The Oncologist 2003;8:303-306 www.TheOncologist.com Correspondence: Martin H. Cohen, M.D., U.S. Food and Drug Administration, HFD-150, 5600 Fishers Lane, Rockville, Maryland 20857, USA. Fax: 301-594-0499; e-mail: cohenma@cder.fda.gov Received May 19, 2003; accepted for publication June 20, 2003. ©AlphaMed Press 1083-7159/2003/$12.00/0 INTRODUCTION Gefitinib (Iressa®, ZD1839; Astra Zeneca, Inc.; London, UK; http://www.astrazeneca.com) (Fig. 1) is a member of a new class of oral drugs that inhibits intracellular tyrosine kinase (TK) activity including that of the epidermal growth factor receptor (EGFR)-TK. Gefitinib binds at the The Oncologist FDA Commentary Cohen, Williams, Sridhara et al. 304 ATP site of the tyrosine kinase region, a region that is highly conserved across the various transmembrane tyrosine kinases. The maximum plasma concentrations resulting from clinically relevant gefitinib doses are 0.5-1 μM or more, similar to or greater than the 50% inhibitory concentration values of other intracellular transmembrane tyrosinespecific protein kinases. Therefore, gefitinib cytotoxicity could be the result of inhibition of downstream signal proteins or ATP-dependent kinases other than EGFR-TK. Gefitinib is extensively metabolized in the liver by the cytochrome P450 3A4 enzyme. Over a 10-day period, approximately 86% of an orally administered radioactive dose is recovered in the feces with less than 4% of the dose in the urine. Following daily oral administration, steadystate plasma levels are reached in 10 days and are twofold higher than those achieved following single doses. In preclinical studies, the antiproliferative activity of gefitinib, alone or in combination with cytotoxic drugs, was investigated in human ovarian (OVCAR-3), breast (MCF10A ras; ZR-75-1), and colon (GEO) cancer cell lines, which express EGFR and transforming growth factor alpha. Gefitinib inhibited colony-forming ability in a concentration-dependent manner. Combining gefitinib with platins (cisplatin, oxaliplatin, carboplatin), taxanes (paclitaxel, docetaxel), topoisomerase inhibitors (doxorubicin, etoposide, topotecan), or the antimetabolite raltitrexed resulted in a markedly greater apoptotic cell death than that induced by single-agent treatment. In studies with colon cancer (GEO) xenografts, combined treatment with gefitinib and cytotoxic agents produced tumor growth arrest and extended the survival of tumor-bearing animals. The submitted new drug application (NDA) sought accelerated approval for gefitinib as monotherapy for patients receiving third-line treatment for locally advanced or metastatic non-small cell lung cancer (NSCLC). At present, there are four cisplatin-containing doublets (docetaxel, gemcitabine, paclitaxel, vinorelbine) and single-agent vinorelbine approved for the first-line treatment of this patient population. Docetaxel is approved for second-line therapy. Third-line treatment is an unmet need. STUDY DESIGN The sponsor submitted two trials in previously treated NSCLC patients. The first trial was in the third-line treatment of NSCLC and provided the primary evidence supporting approval. Failure of prior platinum and docetaxel had to have been the result of treatment intolerance or disease progression within 90 days of the last chemotherapy. The trial was a randomized, double-blind, phase II, multicenter comparison of two doses of gefitinib tablets (250 mg/day versus 500 mg/day). The second trial enrolled patients who had failed one or two previous chemotherapy regimens, at least one having contained a platinum, and provides supportive evidence of gefitinib safety. Two hundred sixteen patients at 30 U.S. medical centers were entered in the third-line treatment trial. One hundred forty-two patients with documented failure of platinum and docetaxel were evaluable for the primary efficacy analysis. Patient demographics and disease characteristics of the eligible treatment population are summarized in Table 1. Approximately 75% of the eligible study patients had adenocarcinoma histology (either alone or mixed with squamous cell histology). Thirty-two percent of patients receiving gefitinib 250 mg/day had never smoked. The median time from lung cancer diagnosis to study randomization was 19.6 months. Among the 142 evaluable patients, there were 15 partial responses. Response rates and durations are summarized in Table 2. Partial responses occurred in 9 of 66 patients receiving gefitinib 250 mg/day and in 6 of 76 patients receiving gefitinib 500 mg/day. The response rate was similar in the 74 study patients not evaluable for the primary efficacy analysis (9.5%). Exploratory analyses of response rates in different subgroups of patients were performed (Table 3). Due to small sample sizes, the results lack precision, as evidenced by the wide confidence limits of the estimates. Responding patients were predominantly female (11/15) and had adenocarcinoma (12/15). Response rates did not vary with World Health Organization (WHO) performance status (0-1 versus 2) or number of prior therapies (2 versus 3 versus 4). While disease-related symptom improvement and quality of life (QOL) were evaluated in both trials, the meaningfulness of that evaluation is questionable. Because Iressa® 250 mg/day and 500 mg/day had comparable efficacy results, there was no comparator regimen for the QOL or symptomrelief analysis. Methodologic issues exist, including the absence of blinding, the censoring of early progressors, and the use of concomitant medications that might have contributed to symptom relief. Therefore, QOL and symptom results are not presented here. Two large trials were conducted in chemotherapynaive, stage III and IV NSCLC patients. Two thousand one O