To assess the weaning outcome of lung cancer patients with acute respiratory failure (ARF) requiring mechanical ventilation, we retrospectively analyzed the database of the respiratory intensive care unit at a university-affiliated tertiary care hospital. Charts were reviewed for cancer status, biochemistries before respiratory failure, causes of respiratory failure, acute physiology and chronic health evaluation (APACHE) III score, ventilatory settings, data recorded during spontaneous breathing, duration of ventilator days, and weaning outcome. Ninety-five consecutive respiratory failure events in 81 patients were recorded from January 1, 1995 through June 30, 1999. Twenty-six episodes ended with successful weaning (27.4%). Age, gender, and cancer status did not affect the weaning outcome. Serum albumin level, APACHE III score, highest fractional inspired O2 (FiO2) and highest positive end-expiratory pressure, organ failure, ability to shift to partial ventilatory support, and duration of mechanical ventilation could significantly influence the weaning outcome statistically. The overall hospital mortality rate was 85.2%. Our results suggested that lung cancer patients with ARF will have a better chance to wean if the initial APACHE III score was less than 70, use of FiO2 never exceeded 0.6, or less than 2 additional organ systems failed during the treatment course.