Purpose: The anesthetic ketamine is widely used for pain related to cancer, but the evidence to support its use in this setting is weak. This study aimed to determine whether ketamine is more effective than placebo when used in conjunction with opioids and standard adjuvant therapy in the management of chronic uncontrolled cancer pain. Ketamine would be considered of net benefit if it provided clinically relevant improvement in pain with limited breakthrough analgesia and acceptable toxicity. Patients and Methods: In this multisite, dose-escalation, double-blind, randomized, placebo-controlled phase III trial, ketamine or placebo was delivered subcutaneously over 3 to 5 days. Results: In all, 185 participants were included in the primary analysis. There was no significant difference between the proportion of positive outcomes (0.04; 95% CI, -0.10 to 0.18; P = .55) in the placebo and intervention arms (response rates, 27% [25 of 92] and 31% [29 of 93]). Pain type (nociceptive v neuropathic) was not a predictor of response. There was almost twice the incidence of adverse events worse than baseline in the ketamine group after day 1 (incidence rate ratio, 1.95; 95% CI, 1.46 to 2.61; P < .001) and throughout the study. Those receiving ketamine were more likely to experience a more severe grade of adverse event per day (odds ratio, 1.09; 95% CI, 1.00 to 1.18; P = .039). The number of patients needed to treat for one additional patient to have a positive outcome from ketamine was 25 (95% CI, six to ∞). The number needed to harm, because of toxicity-related withdrawal, was six (95% CI, four to 13). Conclusion: Ketamine does not have net clinical benefit when used as an adjunct to opioids and standard coanalgesics in cancer pain.