Multi-Institutional Randomized Phase II Trial of Gefitinib for Previously Treated Patients with Advanced Non-Small Cell Lung Cancer

Reviewer: Keith Cengel, MD, PhD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: February 15, 2004

Authors: Masahiro Fukuoka, Seiji Yano, Giuseppe Giaccone, Tomohide Tamura, Kazuhiko Nakagawa, Jean-Yves Douillard, Yutaka Nishiwaki, Johan Vansteenkiste, Shinzoh Kudoh, Danny Rischin, Richard Eek, Takeshi Horai, Kazumasa Noda, Ichiro Takata, Egbert Smit, Steven Averbuch, Angela Macleod, Andrea Feyereislova, Rui-Ping Dong, and José Baselga
Source: Journal of Clinical Oncology Vol 21, No 12, 2003: 2237-2246


In the past two decades, advances in conventional chemotherapy and radiation therapy have yielded only modest increase in median survival of non-small cell lung cancer (NSCLC) patients. For patients with progressive disease after chemotherapy, the prognosis is especially poor with median survival times around 16 weeks. However, recent advances in molecular based therapeutics have generated considerable hope and enthusiasm.

The epidermal growth factor receptor (EGFR) is expressed in the majority of NSCLC tumors and EGFR over-expression is associated with poor prognosis in NSCLC patients. In phase I studies, the EGFR small molecule tyrosine kinase inhibitor gefitinib was well tolerated with characteristic skin rash and GI distress being the most commonly observed toxicities. In addition, patients with NSCLC in these trials showed about a 10% clinical response rate. Therefore, the current multicenter, randomized phase II trial was initiated to evaluate two dose levels of gefitinib in NSCLC patients with disease progression after one or two chemotherapy regimens.


210 patients with refractory NSCLC after 1-2 chemotherapy regimens and a performance status of 0-2 were randomly assigned to either 250 mg/day or 500 mg/day oral gefitinib as single agent therapy. Objective tumor response as assessed by CT, progression free survival (PFS), overall survival (OS) and quality of life as measured by the lung cancer subscale of the FACT-L questionnaire were the primary endpoints of this study.


The response rate (CR + PR) was18% in 250 mg/d vs 19% in 500mg/d (p=ns). The disease control rate (RR + SD) was 54% in 250 mg/d vs 51% in 500 mg/d (p=ns). 40% and 37% of the patient's evaluable for symptom improvement showed increased QoL scores in the 250mg/d and 500mg/d group, respectively. The median time to symptom improvement was 8 days. In patients with stable disease, the symptom improvement rate was 70% in 250mg/d vs 40% in 500mg/d. Median PFS and OS were 2.7 and 7.6 months in 250mg/d vs 2.8 and 8.0 months in 500mg/d, respectively (p=ns). Importantly, gefitinib was quite well-tolerated and the most common toxicities observed were skin and GI, as previously reported.


In this trial, gefitinib produced a clinical response as second line NSCLC therapy that is similar in magnitude to what has been observed in trails of docetaxel chemotherapy. Importantly, these data also show that gefitinib therapy is extremely well tolerated. However, there remain multiple, unresolved issues regarding the wider application of this novel therapy in NSCLC. While gefitinib is a specific EGFR tyrosine kinase inhibitor, there as of yet is no clinically validated marker to help pre-select a responsive patient population. One reason for this is that EGFR is expressed at some level on the majority of NSCLC cells and unlike trastuzumab, which appears to act only on cells that overexpress HER2, gefitinib can produce responses in cells without EGFR overexpression. Moreover, there are multiple other molecular lesions, such as overexpression of the insulin-like growth factor I receptor and expression of oncogenic K-Ras, that can lead to getfitinib resistance even in the presence of EGFR overexpression. This molecular heterogeneity may help explain why gefitinib response rates appeared to be higher in patients with adeno- than squamous cell carcinoma, even though squamous cell carcinomas are thought to much more frequently overexpress EGFR.

Unfortunately, two phase III trials of gefitinib + cytotoxic chemotherapy in locally advanced, previously untreated NSCLC showed no survival benefit for the combination compared to chemotherapy alone. One potential interpretation of these results is that cytotoxic chemotherapy and EGFR tyrosine kinase inhibitors may be mutually antagonistic. However, it is also possible that the cisplatin based chemotherapy given to the patents prior to enrollment in the IDEAL trails may have altered the degree of EGFR-dependence in these tumors.

In this context, it is interesting to note that cisplatin based chemotherapy was found in one study to reduce the percentage of tumors expressing oncogenic K-Ras from 42% to 15%. Therefore, to maximize the clinical application of gefitinib, further studies are needed to clarify the interrelationships between EGFR signaling and other altered signaling pathways in NSCLC. Nevertheless, these results are exciting and suggest that there is great potential the clinical application of small molecule EGFR tyrosine kinase inhibitors in patients with this devastating disease.


Oral gefitinib at either the 250mg/d or 500mg/d dosage provided clinical disease response in about 1/5 and disease control in about 1/2 of patients with chemotherapy refractory NSCLC. These response rates compare favorably to those observed with cytotoxic chemotherapy such as docetaxel in similar patient populations. Moreover, gefitinib was well-tolerated and symptom improvement was observed in about 40% of patients.