Anorexia (loss of appetite) is a prevalent and distressing symptom in people with cancer, with limited effective interventions. Medicinal cannabis has shown promise in improving appetite-related symptoms in people with cancer.

To assess the efficacy of medicinal cannabis for improving appetite-related symptoms in people with cancer, considering measures and outcomes, interventions and toxicity.

Systematic review with narrative approach to synthesis and meta-analysis.

Databases (MEDLINE, CINAHL, CENTRAL), websites and trials registries were searched from inception to February 2021. Included studies were randomised controlled trials (RCT) in English peer-reviewed journals comparing medicinal cannabis with placebo and/or another intervention. Study quality was assessed using the Cochrane risk of bias tool.

Five studies were included that compared medicinal cannabis interventions (dronabinol, nabilone and cannabis extract) either with placebo ( = 4) or megestrol acetate ( = 1). Measures and trial endpoints varied, but efficacy was demonstrated in one trial only, in which dronabinol significantly improved chemosensory perception and other secondary outcomes (taste of food, premeal appetite, proportion of calories consumed as protein) compared with placebo. Cannabis interventions were generally well tolerated across studies, regardless of the product or dose, although the comprehensive measurement of toxicities was limited.

Evidence from RCTs that medicinal cannabis increases appetite in people with cancer is limited. Measures, outcomes and interventions were variable, and toxicities have not been comprehensively evaluated. Future research should carefully consider biological mechanisms to guide more nuanced selection of endpoints and interventions, including product, dose and administration.