Journal of Advances in Medicine and Medical Research

Cannabis and cannabinoid-based products are increasingly used by people with cancer, yet the term “cancer treatment” encompasses two fundamentally different questions: whether cannabinoids can relieve cancer- or treatment-related symptoms, and whether they can modify tumour biology sufficiently to function as anticancer therapy. This critical narrative review evaluates these questions separately and then integrates them through an evidence hierarchy spanning pharmacology, preclinical oncology, randomised supportive-care trials, early cancer-directed studies, observational cohorts, drug-interaction research and contemporary clinical guidance. Literature published from 1 January 1990 to 19 July 2026 was identified through multidisciplinary scholarly searches, supplemented by citation tracking, trial-registry checks and DOI verification. The most clinically credible role is supportive rather than tumour-directed. Cannabinoid medicines can provide benefit for refractory chemotherapy-induced nausea and vomiting when added to contemporary antiemetic care, although much historical evidence predates current prophylaxis. By contrast, randomised evidence does not establish a clinically important benefit for opioid-refractory cancer pain, a dependable opioid-sparing effect, or meaningful reversal of cancer cachexia. Recent placebo-controlled data also fail to show improvement in overall symptom burden with balanced tetrahydrocannabinol–cannabidiol oil, despite limited secondary pain signals accompanied by greater psychomimetic toxicity. Preclinical studies demonstrate context-dependent effects on apoptosis, autophagy, proliferation, angiogenesis, invasion and immune signalling, but effective experimental concentrations, heterogeneous formulations, model limitations and predominantly high risk of bias constrain translation. Human anticancer evidence remains restricted to small glioblastoma studies and non-randomised observations, insufficient to establish efficacy. Immune-checkpoint inhibitor interactions remain unresolved, with conflicting observational signals. The evidence therefore supports a sharply differentiated interpretation: selected cannabinoid medicines may be useful as adjuncts for particular symptoms, whereas cannabis or cannabinoids should not replace established anticancer therapy outside clinical trials. Progress depends on pharmaceutical-grade formulations, exposure-confirmed dosing, mechanism-informed tumour selection, modern supportive-care comparators and adequately powered trials with clinically meaningful endpoints.