Predictive factors for interstitial lung disease, antitumor response, and survival in non-small-cell lung cancer patients treated with gefitinib.

PURPOSE: Interstitial lung disease (ILD) is a serious adverse effect of gefitinib, but its prevalence and risk factors remain largely unknown. We examined the prevalence of and risk factors for gefitinib-induced ILD associated with practical use of the drug in Japanese with non-small-cell lung cancer (NSCLC).

PATIENTS AND METHODS: Clinical information was retrospectively assembled for NSCLC patients who started gefitinib treatment at affiliated institutions of the West Japan Thoracic Oncology Group between August 31 and December 31, 2002. Medical records of patients who developed pulmonary infiltrates were reviewed by a central committee of extramural experts for identification of patients with gefitinib-induced ILD. Multivariate logistic or Cox regression analysis was performed to identify independent predictive factors for ILD, antitumor response, and survival.

RESULTS: Seventy cases of and 31 deaths from gefitinib-induced ILD were identified among 1,976 consecutively treated patients at 84 institutions, corresponding to a prevalence of 3.5% and mortality of 1.6%. Gefitinib-induced ILD was significantly associated with male sex, a history of smoking, and coincidence of interstitial pneumonia (odds ratios = 3.10, 4.79, and 2.89, respectively). Predictive factors for response were female sex, no history of smoking, adenocarcinoma histology, metastatic disease, and good performance status (PS), whereas predictive factors for survival were female sex, no history of smoking, adenocarcinoma histology, nonmetastatic disease, good PS, and previous chest surgery.

CONCLUSION: ILD is a serious adverse effect of gefitinib in the clinical setting that cannot be ignored. However, patient selection based on sex and smoking history can minimize ILD risk and maximize the clinical benefit of gefitinib.