Somewhere between 60 and 70 percent of cancer patients are already using cannabis products during their treatment. They are doing so largely without guidance, because the oncologists treating them — through no fault of their own — often lack the evidence base required to offer meaningful recommendations. Cannabis research has been constrained for decades by regulatory frameworks that classified it alongside hard drugs, and the catch-up has been uneven. The laboratory science is now substantial. The clinical trial data is thinner and more complicated.

A 2024 narrative review published in Cancers by researchers at Soroka Medical Center and Ben Gurion University of the Negev sets out to bridge that gap. The paper is explicitly addressed to both clinicians and patients — a relatively unusual framing in a peer-reviewed oncology journal. It covers cannabis history, pharmacology, methods of consumption, symptom management across five domains, anti-tumour activity, and side effects. It is comprehensive in scope and candid about where the evidence runs out.

This is the framework through which we will examine it — not as a summary, but as an honest accounting of what the research does and does not support.

The Knowledge Problem — Why Oncologists Can't Answer Their Patients' Questions

The review opens with an observation that will resonate with anyone who has navigated cancer care: patients seeking to integrate cannabis into their treatment encounter frustration when their oncologists lack adequate information to provide guidance. This is not a failure of individual physicians. It is a structural consequence of decades of suppressed research.

Cannabis was removed from the US pharmacopeia in 1941, following mounting legal restrictions that classified it alongside other controlled substances. Research into its medicinal applications slowed significantly for more than half a century. By the time the endocannabinoid system was properly characterised — cannabinoid receptor 1 was identified in the early 1990s — the scientific and clinical infrastructure needed to study cannabis properly was still decades behind where it would have been without the interruption.

The result is a knowledge asymmetry that operates in both directions. Patients who have heard promising anecdotal accounts of cannabis and cancer arrive with questions that their physicians cannot confidently answer. Physicians who are aware of the preclinical evidence but lack access to clinical trial data are uncomfortable offering guidance that might be wrong in either direction — either overstating benefit or unnecessarily discouraging something that might help.

The Endocannabinoid System — The Biological Context That Makes This All Possible

To understand why cannabis interacts with cancer in the ways it appears to, you need to understand the endocannabinoid system. This is not optional background — it is the mechanism through which all the therapeutic effects described in this review operate.

Cannabis sativa contains over 500 chemical compounds, of which at least 100 are phytocannabinoids. The most studied are delta-9-tetrahydrocannabinol (THC), which produces psychoactive effects, and cannabidiol (CBD), which does not. The plant also contains terpenes and flavonoids that contribute to its biological activity through what researchers call the entourage effect — the enhanced benefit of compounds working together rather than in isolation.

CB1 and CB2 receptors are expressed not just on neurons and immune cells but on tumour cells themselves. CB1 and CB2 agonists selectively inhibit production of VEGF-A — a potent driver of angiogenesis — in activated immune cells, which has direct relevance to tumour blood vessel formation. This is the biological scaffolding upon which the anti-tumour evidence rests.

How People Actually Use Cannabis — And Why It Matters Clinically

Before turning to therapeutic effects, the review addresses something often skipped in academic treatments of this subject: the practical pharmacology of different consumption methods. This matters for oncology patients because onset time, bioavailability, and dose control vary substantially depending on how cannabis is consumed — and getting the dose wrong has real consequences in patients who are already immunocompromised and managing complex medication regimens.

Appetite and Weight — What the Numbers Actually Show

Appetite loss and cancer-related cachexia — the progressive wasting syndrome seen in advanced cancer — are among the most distressing consequences of both the disease and its treatment. Cannabis has long been understood to stimulate appetite, and the review provides specific data on how this effect compares to standard pharmacological options.

Studies indicate that cannabis can increase caloric intake by approximately 40%, with the effect distributed across the day rather than concentrated at mealtimes. Interestingly, the caloric increase is driven primarily by snacks — particularly sweet solid foods — which has implications for nutritional counselling. The effect on actual weight gain is more variable.

The data here is nuanced. On raw appetite improvement numbers, the established pharmaceutical megestrol outperforms dronabinol. But the trials were not designed to test whole-plant cannabis preparations, which differ from isolated synthetic cannabinoids in ways the research is only beginning to characterise. Higher CBD strains appear to produce less appetite stimulation than high-THC preparations — a ratio consideration with direct relevance for product selection in clinical settings.

Pain Management — The Case That Is Most Developed

Pain is the symptom domain in which the cannabis evidence is deepest, and the Soroka review covers it with appropriate complexity. Cancer pain is not a single entity. It arises from bone metastasis, spinal cord compression, chemotherapy-induced peripheral neuropathy, pathological fractures, and nerve compression — each with somewhat different pharmacological requirements.

The current standard involves opioid analgesics, which carry risks of dependence and dose-escalation that are particularly problematic in patients already managing complex treatment regimens. The review describes cannabinoids as a potential alternative or adjunctive therapy — one that engages different pain mechanisms entirely, through CB1 and CB2 receptors rather than opioid receptors, meaning their analgesic effects are not blocked by opioid antagonists.

The chemotherapy-induced peripheral neuropathy data is particularly interesting. A retrospective analysis of 513 patients treated with oxaliplatin found that cannabis significantly reduced the rate of neuropathy — 15.3% in cannabis users versus 27.9% in controls. The protective effect was more pronounced in patients who began cannabis before starting oxaliplatin treatment (75% protection) versus those who started cannabis afterward (46%). This temporal finding — that early introduction matters — has direct implications for when cannabis should be discussed with patients, not just whether.

The only published controlled trial on cannabis for chemotherapy-induced peripheral neuropathy — involving 16 patients randomised to nabiximols or placebo — found no statistically significant difference between groups on average pain scores. However, responder analysis revealed clinically significant pain reduction in a subset of patients, with a mean reduction of 2.6 points on a 0–10 scale and a number needed to treat of five. The trial was small. The question it raises is not closed.

Nausea and Vomiting — The Established Indication

If there is one domain in which cannabis has the clearest established evidence for cancer patients, it is chemotherapy-induced nausea and vomiting. Multiple national academies of science, systematic reviews, and a Cochrane analysis have concluded that oral cannabinoids are effective antiemetics in adults undergoing chemotherapy. The biological mechanism runs through CB1 receptors on dopaminergic and noradrenergic neurons in brain regions governing the emetic response.

The review highlights a phase II crossover trial that is worth examining in detail. Eighty-one cancer patients receiving emetogenic intravenous chemotherapy, with persistent nausea and vomiting despite standard antiemetics, were randomised to THC:CBD capsules (2.5 mg each, three times daily) or placebo across two chemotherapy cycles, with patients choosing their preferred treatment for a third cycle.

The 83% patient preference figure deserves emphasis. In a population already dealing with significant adverse effects of cancer treatment, 83% of patients preferred a therapy that produced more side effects than placebo — because those side effects were less burdensome than uncontrolled nausea. This is a patient-centred outcome measure that purely statistical analyses can obscure.

The review notes that the American Society of Clinical Oncology's expert panel remains cautious, citing insufficient data to formally recommend medical cannabis for nausea prevention. This is a legitimate scientific conservatism — but it sits in some tension with the reality that most cancer patients are already making their own decisions without formal guidance.

Sleep — An Underexplored but Clinically Significant Domain

Sleep disturbance affects up to 19% of the general population and is substantially more prevalent among cancer patients. It is also among the least well-studied applications of cannabis in oncology. The review's treatment of this domain is appropriately cautious about what the evidence can and cannot support.

Short-term, high-dose CBD may assist in reducing sleep onset latency and prolonging sleep duration — possibly through CBD's anxiolytic properties rather than through direct sedation. Nabiximols studies involving cancer patients in pain have reported subjective improvements in sleep quality, though the review notes these may reflect reduced pain rather than changes to sleep biology itself. This distinction matters for product selection.

Anti-Tumour Effects — The Evidence Hierarchy

This is the domain that attracts the most attention and generates the most confusion — both in the popular press and in clinical conversations. The review addresses it systematically, distinguishing between preclinical findings, early clinical results, and the significant gap between them.

The preclinical case is substantial. Cannabinoids interfere with cancer cell biology through multiple mechanisms: they induce apoptosis (programmed cell death) directly, block tumour angiogenesis by inhibiting VEGF-A production, suppress metastasis, trigger autophagy, and inhibit cell proliferation. These effects have been demonstrated across lung, breast, prostate, glioblastoma, and ovarian cancer models, among others. The endocannabinoid anandamide inhibits proliferation in prostate cancer cell lines by downregulating epidermal growth factor receptor expression. THC and the synthetic cannabinoid JWH-133 reduce tumour growth, metastases, and angiogenesis in breast cancer mouse models through Akt pathway inhibition.

The most clinically advanced anti-tumour evidence involves glioblastoma multiforme — the most aggressive form of brain cancer. A pilot trial involving intracranial THC administration in recurrent glioblastoma patients found tumour proliferation reduction in two of nine patients. The subsequent nabiximols plus temozolomide trial produced the result that now anchors the entire clinical anti-tumour discussion.

A finding with very different implications also appears in the review — one that deserves equal attention. A study of 68 metastatic cancer patients beginning immunotherapy found that cannabis users demonstrated a median time to tumour progression of only 3.4 months, compared to 13.1 months in non-users. Median survival was 6.4 months in cannabis users versus 28.5 months in non-users. The anti-inflammatory properties of cannabis may interfere with the mechanism by which immunotherapy activates the immune system against tumours — a critical interaction that has not yet been adequately characterised in randomised trials.

Side Effects — An Honest Accounting

The review does not advocate uncritically. It presents the adverse effect profile of cannabis with the same rigour applied to the therapeutic evidence. This balance is one of the paper's genuine strengths.

Acute psychoactive effects of THC include euphoria, anxiety, sensory distortions, altered time perception, and paranoia. At higher doses — particularly from concentrated vaporised products — arrhythmia and myocardial infarction risk increase in susceptible individuals. The most common adverse effects of synthetic cannabinoids in a review of over 3,600 toxicity reports were tachycardia (30%), agitation (13.5%), drowsiness (12.3%), nausea and vomiting (8.2%), and hallucinations (7.6%). Deaths and severe outcomes were rare (0.2% and 0.1–0.09% respectively).

For chronic use, the concerns are different: tolerance development, withdrawal effects including severe depressive episodes, increased systolic hypertension risk, ischaemic stroke risk, and ventricular arrhythmia risk. Cannabis also reduces immune response to some infections — a consideration that cannot be ignored in immunocompromised cancer patients.

Non-psychoactive CBD presents a substantially better risk-benefit profile. Its absence of psychoactivity eliminates several of the acute concerns, and it constitutes up to 40% of whole-plant cannabis extracts. The challenge is that the anti-tumour and symptom management literature does not always clearly distinguish between CBD-dominant, THC-dominant, and balanced preparations — making clinical translation of specific findings more complicated than headlines suggest.

What the Review Asks of Researchers, Clinicians, and Regulators

The Soroka review is ultimately a call to action addressed to three audiences simultaneously. For researchers, it identifies the specific gaps: adequately powered randomised controlled trials with standardised preparations, clear patient stratification, and outcomes that capture quality of life alongside disease progression. For clinicians, it offers a framework for evidence-based conversations with patients who are already using cannabis and need guidance rather than dismissal. For regulators, it documents that the knowledge gap is not a scientific problem but a structural one — created by decades of regulatory restriction and maintainable only by continued restriction.

The paper's most useful contribution may be its honest acknowledgement of what the evidence does and does not support. Cannabis is not a cancer cure. It is a complex plant producing biologically active compounds that interact with fundamental aspects of cancer biology in ways that are scientifically credible, reproducible across multiple research groups, and — in the glioblastoma case — clinically promising. It is also a compound with genuine risks, genuine drug interactions, and at least one documented context — immunotherapy — where its use may be harmful.

That complexity is precisely what patients deserve to hear, and exactly what this review tries to provide.