Objectives Explain the potential benefits of a caregiver-assisted pain coping skills training intervention. Describe the relative benefits of the caregiver-assisted pain coping skills intervention compared to enhanced treatment-as-usual for patients with advanced cancer and their family caregivers. Describe challenges of delivering a behavioral intervention to patients with serious illness. Importance. Pain is common among patients with advanced cancer and causes distress for both patients and their caregivers. Cognitive-behavioral pain coping skills interventions can improve pain and pain-related outcomes but have rarely been tested in advanced cancer. Objective(s). To conduct a multi-site RCT testing the efficacy of a caregiver-assisted pain coping skills training (CG-CST) intervention for advanced cancer. Method(s). Patients with stage III-IV cancer and moderate-severe pain and their family caregivers were recruited from four academic medical centers and one hospice/palliative care organization. They were randomized to CG-CST or enhanced treatmentas-usual (E-TAU). Dyads in both conditions received educational resources on pain management; those in CG-CST received three 60-minute sessions via videoconference. Caregiver outcomes (self-efficacy for helping the patient manage pain, caregiver strain, caregiving satisfaction, psychological distress) and patient outcomes (self-efficacy for pain management, pain intensity and interference, psychological distress) were collected at baseline and post-intervention. Results. 202 dyads enrolled and were randomized to E-TAU (N¼101) or CG-CST (N¼101). 171 dyads (92 E-TAU, 79 CG-CST) completed post-intervention assessments. Mixed models for repeated measures were used to estimate pre-post changes in outcomes; effect sizes (ES) for within and between group changes were calculated. Both conditions led to comparable improvements in patient self-efficacy (E-TAU/CG-CST ES¼0.42/0.43), caregiver self-efficacy (E-TAU/CGCST ES¼0.43/0.42), patient pain severity (E-TAU/ CG-CST ES¼-0.27/-0.32), pain interference (E-TAU/ CG-CST ES¼-0.44/0.42), and patient psychological distress (E-TAU/CG-CST ES¼-0.31/-0.27) (all p's<.02). Compared to E-TAU, CG-CST improved caregiving satisfaction (differential ES¼0.41, p<.01) and tended to improve caregiver anxiety (differential ES¼-0.30, p¼.06). Conclusion(s). Contrary to expectations, the CGCST intervention did not improve pain outcomes relative to enhanced TAU, although it did lead to improved caregiving satisfaction and decreased caregiver anxiety. Differential retention in the study arms and temporal improvement trends may have impacted findings. Impact. Despite the promise of behavioral pain coping interventions, challenges in retaining seriously ill patients in such interventions may dampen their effects.