Seventeen elderly patients with advanced progressive non small cell lung cancer (NSCLC) were treated with oral etoposide at the daily dose of 100 mg for 14 days every 3-4 weeks with pharmacokinetic monitoring. One partial response and 6 stabilizations were documented with a median overall duration of 13 weeks (range 8-32). The median survival was 24 weeks with an apparent advantage for non-progressive patients (40 weeks vs. 18 weeks). The treatment was well tolerated especially by those patients without concomitant illness, suggesting the crucial role of a careful selection of the geriatric population. Toxicity was not related to the etoposide plasma level, but was clearly dependent on comorbidity. A geriatric assessment rather than chronological age therefore appears to be more reliable in the selection of elderly patients for clinical trials. The easy self-management, favorable toxicity profile and synergy with other compounds makes oral etoposide suitable for further clinical-pharmacological studies in elderly patients.