This study was performed to evaluate the activity of single-agent temozolomide in two groups of chemotherapy-naïve non-small cell lung cancer (NSCLC) patients, with (12 patients) and without (13 patients) brain metastases (BM). Patients in both groups were treated with temozolomide 200 mg/m(2)/day, administered orally for 5 consecutive days of a 28-day cycle. Treatment was continued for up to six cycles, disease progression or unacceptable toxicity. The median number of received cycles was only one in the group with and two in the group without BM, and early disease progression was the main reason for treatment discontinuation. Toxicity was moderate-in the group of patients with BM, the most frequently observed grade 3 or 4 side-effects included thrombocytopenia (17%), granulocytopenia (17%), lethargy (17%); other neurological (17%) and other genitourinary toxicity (17%). Patients without BM experienced anaemia (15%), thrombocytopenia (23%), nausea (15%) and lethargy (15%). This trial was designed according to Simon one-sample two-stage testing procedure and both groups of patients were assessed separately. No objective response was observed in either group and the study was closed after the first step of accrual with the conclusion of a lack of therapeutic activity of single-agent temozolomide in patients with stage IV NSCLC.