In this study we evaluated the efficacy and tolerability of sublingual fentanyl (SLF) for breakthrough pain (BTP) in adult opioid-tolerant cancer patients. Patients received one dose of placebo, SLF 100, 200 and 400 microg in random order at four pain episodes. The primary efficacy endpoint was pain intensity difference (PID) from baseline. Twenty-seven patients received study medication. Overall PID increased significantly with SLF 400 microg versus placebo (8.57 mm, p <0.0001). Improvements were statistically different from placebo at 15 min (p = 0.005). SLF 100 and 200 microg showed a numerical trend towards improved pain relief. A dose that gave a clinically important reduction in pain (PID > 20 mm) was identified by 95% of patients. Reduced use of rescue medication (p < 0.001, SLF 400 microg) and improved global assessment of treatment (p = 0.0146, SLF 400 microg) confirmed these differences as clinically important. Nausea and dizziness were the most common treatment-related adverse effects. SLF appears to be a fast, effective and well-tolerated treatment for BTP.