Objective: The aim of this study was to compare the therapeutic effects of immediate-release morphine (IR) and controlled-release oxycodone (CR) tablets on cancer pain in patients with an analgesia history of opioid or non-opioid analgesics by titration. Methods: This study was completed by six medical institutions, together, which adopted perspective and double-blind open-label methods to randomly divide cancer pain patients, from December 2014 to December 2017, into four groups: naive IR, naive CR, tolerant IR, and tolerant CR. Results: Kaplan-Meier analysis showed that remission of pain in the naive CR group (n = 58) was superior to that in the naive IR group (n = 47) (P = 0.031). Pain control in the tolerant CR group (n = 52) was superior to that in the tolerant IR group (n = 49) (P < 0.001). Differences were observed in titration cycles among the four groups (P < 0.001). There were differences in total titration doses among the four groups (P < 0.001). Consumption of oxycodone hydrochloride tablets in the naive CR group was lower than that that in the naive IR group (All P < 0.001) duringW1 & W2, whereas no significant differences were observed between the tolerant IR group and tolerant CR group during W1 (P = 0.061) & W2 (P = 0.060). The incidence rate of additional drug delivery in the naive CR group was less than that in the naive IR group (P = 0.007). There was a lower incidence rate of additional drug delivery in the tolerant CR group, compared with that in the tolerant IR group (P = 0.014). There were no significant differences in adverse effects among the four groups (All P > 0.05). Conclusion: As a therapeutic oral drug, controlled-release oxycodone may obtain more stable analgesic effects than immediate-release morphine, whether the patient has an analgesia history of opioid analgesics or not. Fewer doses are required during treatment, thus it should be used in clinic for cancer pain.