BackgroundAdvance (anticipatory) care planning (ACP) requires discussions between patients and healthcare professionals about planning for future deterioration in health. ACP improves care coordination but uptake is limited and often deferred.AimTo assess the feasibility and acceptability to patients, carers, and GPs of a primary care ACP intervention for people with incurable oesophageal, gastric, or pancreatic cancer.Design and settingA 12-month feasibility randomised controlled trial (RCT) in a Scottish Cancer Network.MethodPatients aged ≥18 years starting palliative oncology treatment were randomised 1:1 to an ACP intervention or standard care. Patients in the intervention group received an oncologist letter supporting them to request a GP review along with a patient information leaflet about ACP. Pre-specified analyses with masking included trial recruitment and retention, ACP completion, and quality-of-life questionnaires (EuroQol EQ-5D-5L and ICECAP Supportive Care Measure) at baseline, 6, 12, 24, and 48 weeks. Qualitative interviews with purposive sampling explored patient, carer, and GP experiences.ResultsOf 99 eligible participants (269 screened), 46% were recruited (n = 46) and randomised; 25 to intervention and 21 to control. By 12 weeks, 45% (n = 9/20) of the individuals in the intervention and 59% (n = 10/17) in the control group had a documented ACP plan. By 24 weeks, 30% (n = 14) had died; in the remaining participants quality of life was maintained at 24 weeks except for physical symptoms. Social norms associating ACP with dying were prevalent among 23 participants interviewed. No psychological or clinical harms were identified.ConclusionAn RCT of ACP for people with incurable cancer in primary care is feasible. Patient, carer, and GP attitudes and behaviours determined acceptability and timing of care planning.