A randomized phase II trial of medical cannabis to reduce symptom burden in patients with newly diagnosed advanced pancreatic cancer (CanPan).

Abstract

697 Background: Patients with advanced pancreatic adenocarcinoma experience severe, persistent symptoms. Medical cannabis shows promise for symptom management, yet high-quality randomized data are lacking due to regulatory barriers in cannabis research. Partnering with state cannabis programs may represent a novel pathway to conduct cannabis trials. Methods: We conducted a randomized phase II trial of early versus delayed (waitlist control) cannabis for 32 patients with newly diagnosed locally advanced or metastatic pancreatic adenocarcinoma planning to initiate cytotoxic chemotherapy, ≥1 qualifying symptom (pain, nausea, and/or anorexia) and without regular cannabis use, from one community and one academic site. Patients were randomized 1:1 to early (0-8 weeks) or delayed (9-16 weeks) cannabis intervention (certification, education, personalized provision of cannabis products at no cost) through the Minnesota Medical Cannabis Program. The primary study period was 0-8 weeks when only the early arm received the intervention. Patients completed symptom (PRO-CTCAE; ranges 0-4, higher scores=more symptom severity, assessing efficacy and potential harms; weekly) and quality-of-life (PROMIS Global; every 4 weeks) surveys. The primary outcome was feasibility. Secondary outcomes included acceptability, changes in symptom burden, quality-of-life, and opioid use. Results: From October 2024 to May 2025, we enrolled 34 patients, 32 of whom began the study (median age 71 years, 53% women, 56% metastatic disease). Patients reported substantial moderate-to-severe baseline symptom burden: insomnia (85%), pain (77%), and appetite loss (69%). Ten (31%) required opioids with mean (SD) baseline Oral Morphine Equivalents (OME) use of 7 (16) mg. The study met pre-specified feasibility benchmarks (74% enrollment (goal ≥20%), 78% compliance with arm allocation (goal ≥60%), and 75% PRO completion rate (goal ≥50%)). Patients reported high satisfaction and acceptability of cannabis intervention with 84% recommending the intervention to others. At 8 weeks, early arm patients experienced numerically but not statistically significantly higher rates of improvement in pain (44% vs 20%, p=0.35), appetite (56% vs 30%, p=0.37), and insomnia (67% vs 30%, p=0.18). They also experienced a numerically greater decrease in OME (mean (SD); -3 (8) vs +7 (30), p=0.22), and lower rates of new symptoms related to potential cannabis-related harms (dry mouth (22% vs 40%, p=0.22) and dizziness (11% vs 30%, p=0.78)). Conclusions: We demonstrate the feasibility and acceptability of conducting interventional cannabis trials through a model collaboration between investigators and a state cannabis program that can overcome regulatory barriers. The encouraging preliminary efficacy and safety of early cannabis access in managing symptoms supports further exploration. Clinical trial information: NCT06605430 .