BACKGROUND Epidural analgesia is a common technique for managing perioperative and obstetric pain. Patients with cancer who cannot tolerate opioids or not responding to conventional treatment may benefit from epidural analgesia. Therefore, this systematic review aimed to analyze the efficacy and safety of epidural analgesia in patients with intractable cancer pain. METHODS We searched PubMed, Embase, and the Cochrane Central Register of Controlled Trials to identify studies on patients with cancer who received epidural analgesia. We assessed the quality of all included studies using the risk-of-bias tool or Newcastle-Ottawa scale. The primary outcome was pain relief after epidural analgesia, and the secondary outcome was quality of life, analgesic consumption, and adverse events. The studies were grouped based on the medications used for epidural analgesia. A descriptive synthesis was performed following the Synthesis Without Meta-analysis reporting guideline. RESULTS Our systematic review included nine randomized controlled trials (n = 340) and 15 observational studies (n = 926). Two randomized controlled trials suggested that epidural opioids were not superior to systemic opioids in relieving pain. Epidural opioids combined with local anesthetics or adjuvants, including calcitonin, clonidine, ketamine, neostigmine, methadone, and dexamethasone, offered better analgesic effects. No significant difference in pain relief between an intermittent bolus and a continuous infusion of epidural morphine was observed. Epidural opioids had more analgesic effects on nociceptive pain than neuropathic pain. The methods used to evaluate the quality of life and the corresponding results were heterogeneous among studies. Six observational studies demonstrated that some patients could have decreased opioid consumption after epidural analgesia. Adverse events, including complications and drug-related side effects, were reported in 23 studies. Five serious complications, such as epidural abscess and hematoma, required surgical management. The heterogeneity and methodological limitations of the studies hindered meta-analysis and evidence-level determination. CONCLUSIONS Coadministration of epidural opioids, local anesthetics, and adjuvants may provide better pain relief for intractable cancer pain. However, we must assess the patients to ensure that the benefits outweigh the risks before epidural analgesia. Therefore, further high-quality studies are required.