OBJECTIVES: The objective of this study was to determine the effects of physical therapy, including massage and exercise, on pain and mood in patients with advanced terminal cancer. DESIGN: The design was a randomized controlled pilot study. SUBJECTS: Twenty-four (24) patients with terminal cancer were randomly assigned to one of two treatment groups. INTERVENTIONS: Group A received a physiotherapy intervention consisting of several massage techniques, mobilizations, and local and global exercises. Group B received a simple hand contact/touch to areas of pain (cervical area, shoulder, interscapular area, heels, and gastrocnemius), which was maintained for the same period of time as the intervention group. All patients received six sessions of 30-35 minutes in duration over a 2-week period. OUTCOMES: Outcomes were collected at baseline, at 1 week, and at a 2-week follow-up (after treatment completion) by an assessor blinded to the treatment allocation of the participants. Outcomes included the Brief Pain Inventory (BPI, 0-10 scale), Memorial Pain Assessment Card (0-10 scale), and Memorial Symptom Assessment Scale (MSAS Physical, Psychological, 0-4 scale). Baseline between-group differences were assessed with an independent t-test. A two-way repeated-measures analysis of variance was used to examine the effects of the intervention. RESULTS: There were no significant between-group baseline differences (p>0.2). A significant group × time interaction with greater improvements in group A was found for BPI worst pain (F=3.5, p=0.036), BPI pain right now (F=3.94, p=0.027), and BPI index (F=13.2, p<0.001), for MSAS Psychological (F=8.480, p=0.001). CONCLUSIONS: The combination of massage and exercises can reduce pain and improve mood in patients with terminal cancer. A sustained effect on pain and psychologic distress existed; however, parameters such as physical distress and the least pain were no greater in the intervention group as compared to the sham.