Predictors of Response to Gefitinib (Iressa®)

Abstract & Commentary

Synopsis: Gefitinib, an inhibitor of EGFR tyrosine kinase, has been demonstrated to produce clinical responses in a small number of patients with advanced non-small-cell lung cancer. Prior reviews had indicated that women and patients with adenocarcinoma histology had a somewhat higher response rate. In this review of 3 consecutive phase II trials, other clinical features were sought that might predict treatment response. Multivariable analysis revealed the presence of bronchioalveolar features and being a never smoker were additional, independent predictors of response.

Source: Miller VA, et al. J Clin Oncol. 2004;22:1103-1109.

Activation of the epidermal growth factor receptor (EGFR) by ligand binding or mutation stimulates cell growth, proliferation, invasion, and metastasis and prevents apoptosis.1 Inasmuch as EGFR has been shown to be present in all non-small-cell lung cancer (NSCLC) tumors, the receptor itself, or its ligands or downstream effectors have become principal targets in the treatment of these tumors. Theoretically, the inhibition of EGFR and its cellular-mediated effects should retard the growth of NSCLCs driven by EGFR signaling. Gefitinib (Iressa®), a synthetic anilinoquinazoline, inhibits EGFR tyrosine kinase and has been shown to produce radiographic regressions with symptomatic benefits in several phase II trials.2,3 Clinical and pathologic features including female sex, and adenocarcinoma have been demonstrated to influence response to gefitinib. It has been postulated that additional pretreatment variables may also predict sensitivity to this agent.

Miller and colleagues from the Memorial Sloan-Kettering Cancer Center examined 139 patients with NSCLC participating in 1 of 3 consecutive studies of gefitinib monotherapy. These patients were analyzed for the prognostic significance of stage, sex, age, performance status, bone metastasis, number of prior chemotherapy regimens, use of prior cisplatin, carboplatin or docetaxel, smoking history, and tumor cell type. The latter variables (tumor cell type and smoking history) had not been evaluated in this regard previously.

Overall, 90% of the 139 patients received 250 mg daily and 2 patients received less than 225 mg daily. The observed response rate to gefitinib in the clinical trials was 15% (95% CI, 9-21%). None of the 38 patients with a NSCLC other than an adenocarcinoma experienced a radiographic regression (0% observed rate; 95% CI, 0-9%; P < .001). The adenocarcinoma group was further separated into 2 subgroups—tumors with or without elements of bronchioloalveolar cell carcinoma. The presence of bronchioloalveolar features was found to be significant with regard to treatment (38% vs 14% response rate; P < .001). Additionally, 13 (36%) of the 36 never smokers experienced regressions as compared with 8 (8%) of the 104 current or former smokers (P < .001). Furthermore, patients with Karnofsky performance status = 80% were more sensitive to gefitinib than those with Karnofsky performance status = 70% (22% vs 8%; P = .03). Multivariable analysis demonstrated that the presence of bronchioloalveolar features in an adenocarcinoma (or pure bronchioloalveolar cell carcinoma) and a history of never smoking are independent predictors of gefitinib sensitivity.

Comment by William B. Ershler, MD

In previous NSCLC treatment analyses,2,3 it was apparent that women—and patients with adenocarcinoma—were somewhat more likely to respond to Gefitinib treatment. Now we can add bronchioalveolar features and non-smoking history to those clinical features that would offer predictive value for treatment response. Where we go with these observations remains a question. Should male patients with other histological types of NSCLC be treated with other drugs? Even with lower response rates, prescribing an oral agent that has associated low levels of toxicity certainly deserves some consideration at a time when alternatives are only marginally more successful and have a greater profile of adverse consequences.

Yet, the real hope is that laboratory-oriented investigators will take these clinical observations back to the bench. As indicated by Miller et al, tumor samples from Gefitinib-responding patients will be compared with samples taken from resistant patients by microarray to study the EGFR pathway in an effort to identify molecular markers that might be useful in identifying individual patients likely to respond to gefitinib.


1. Mendelsohn J. Clin Cancer Res. 2000;6:747-753.
2. Fukuoka M, et al. J Clin Oncol. 2003;21:2237-2246.
3. Kris MG, et al. JAMA. 2003;290:2149-2158.

William B. Ershler, MD INOVA Fairfax Hospital Cancer Center, Fairfax, VA; Director, Institute for Advanced Studies in Aging, Washington, DC, is Editor of Clinical Oncology Alert.