This pilot trial sought to investigate whether any of three doses of American ginseng (Panax quinquefolius) might help cancer-related fatigue. A secondary aim was to evaluate toxicity. Eligible adults with cancer were randomized in a double-blind manner, to receive American ginseng in doses of 750, 1,000, or 2,000 mg/day or placebo given in twice daily dosing over 8 weeks. Outcome measures included the Brief Fatigue Inventory, vitality subscale of the Medical Outcome Scale Short Form-36 (SF-36), and the Global Impression of Benefit Scale at 4 and 8 weeks. Two hundred ninety patients were accrued to this trial. Nonsignificant trends for all outcomes were seen in favor of the 1,000- and 2,000-mg/day doses of American ginseng. Area under the curve analysis of activity interference from the Brief Fatigue Inventory was 460-467 in the placebo group and 750 mg/day group versus 480-551 in the 1,000- and 2,000-mg/day arms, respectively. Change from baseline in the vitality subscale of the SF-36 was 7.3-7.8 in the placebo and the 750-mg/day arm, versus 10.5-14.6 in the 1,000- and 2,000-mg/day arms. Over twice as many patients on ginseng perceived a benefit and were satisfied with treatment over those on placebo. There were no significant differences in any measured toxicities between any of the arms. There appears to be some activity and tolerable toxicity at 1,000-2,000 mg/day doses of American ginseng with regard to cancer-related fatigue. Thus, further study of American ginseng is warranted.