01 — Evidence
Evidence Summary
CBD's evidence base looks different depending on where you look: thin and non-supportive in human tumor outcomes, but consistent and mechanistically detailed in animal and cell models. Reading across all three tiers together gives the fairest picture.
Human
Clinical Record
Pharmacokinetics & safety data
The most rigorous human trial to date found no meaningful difference between CBD and placebo on overall symptom burden in advanced cancer patients — and a direct follow-up test of CBD's anti-inflammatory effect in that same patient group also found nothing. No large randomized trial has yet demonstrated tumor regression or survival benefit from CBD as a standalone therapy.
A companion trial combining CBD with THC reported a narrow, pain-specific benefit — not a broader symptom improvement — alongside more frequent sedation. Robust clinical data do exist from prescription cannabidiol programs, defining dose ranges, plasma exposure, hepatic safety patterns, and drug–drug interaction risks.
Animal
Preclinical Signal
Xenograft & tumor models
Animal models provide the most consistent evidence that CBD affects tumor behavior directly, spanning breast, glioblastoma, colorectal, lung, and pancreatic models — but response depends heavily on tumor genetics and context, not a uniform effect.
- Reduced tumor growth in several xenograft and syngeneic models
- In colorectal models, response depended on a single gene (p53 status)
- Fewer metastasis-supporting immune cells reported in breast models
- Increased tumor-cell apoptosis signaling across multiple tumor types
In Vitro
Cell Model Data
Extensive; concentration-dependent
The deepest mechanistic detail of any evidence tier — but the concentrations that produce these effects in a dish often exceed what oral dosing achieves in the bloodstream. In colorectal cells, a single protein (p53 status) was the strongest predictor of whether CBD triggered cell death at all.
- Elevated intracellular ROS and ER stress markers
- Intrinsic (mitochondrial) apoptosis signaling
- Genes for building new fat shut down in ovarian cancer cells
- Autophagy pathway modulation, effect direction varies by tumor type
Human
Clinical Record
The clearest answer human research has produced so far is a negative one. In the most rigorous test of CBD alone in cancer patients to date, a large randomised, placebo-controlled trial found no meaningful difference between CBD and placebo on overall symptom burden (primary endpoint: ESAS total symptom distress score) — patients on both arms improved by roughly the same amount. CBD was reported to be safe and well tolerated, but the trial did not support a symptom benefit from CBD alone.[1] A direct follow-up test in the same patient cohort measured serial inflammatory blood markers (CRP and a cytokine panel) through day 28 and found no difference between arms either — the study authors state directly that they were unable to demonstrate an anti-inflammatory effect of CBD in cancer patients. Two independent negative results from the same trial population, not one, anchor how the rest of the human evidence should be read.[2]
Continue reading — full research detail+
Adding THC to the mix shifted the picture, but only narrowly. A companion trial testing a combined THC and CBD oil reported a small, statistically real improvement specifically on pain (ESAS pain subscore) — not on overall symptom distress, where placebo actually performed better. That narrow gain came with a cost: sedation-type side effects were more frequent in the cannabinoid arm.[3]
In glioblastoma, one small retrospective study reported longer-than-expected survival in patients using CBD alongside standard treatment, compared against historical outcome benchmarks rather than a concurrent control group. That finding cannot be trusted at face value: there was no control group, the sample was small, and patients weren't randomly selected — this is the kind of result that generates a hypothesis, not one that confirms it.[4]
A separate glioblastoma trial (the GEINO-1601 Phase Ib trial) answered a narrower, more modest question — is a combined THC/CBD solution tolerable alongside chemotherapy and radiotherapy — and the answer was yes: no serious side effects were attributed to the cannabinoid component. That trial was never designed to test whether CBD affects survival, and its survival data haven't been reported yet.[5]
Signal maturity: taken together, human research currently answers questions about safety and tolerability — including a specific, direct test of the anti-inflammatory hypothesis, which was also negative — not about whether CBD affects cancer itself. No randomised trial has yet shown that CBD affects tumour growth or survival in humans.
Animal
Preclinical Signal
In breast cancer models (syngeneic 4T1 and MVT-1 mouse models), CBD did more than slow tumour growth — it appeared to change the tumour's surrounding environment. Alongside reduced primary tumour size and fewer lung metastases, researchers reported fewer of the immune cells known to support metastatic spread (M2 tumour-associated macrophages), plus a drop in the enzymes tumours use to break through surrounding tissue (MMP2/MMP9). Because this was tested in mice with intact immune systems rather than immune-deficient models, the finding carries more translational weight than typical xenograft data.[6,7]
Continue reading — full research detail+
In glioblastoma, CBD's clearest win came as a partner to chemotherapy, not a replacement for it — and only in tumours carrying one specific genetic marker (MGMT-promoter methylation status). Tumours with that marker became significantly more responsive to standard chemotherapy (temozolomide) when CBD was added (via RAD51 inhibition, impairing the tumour's DNA-repair capacity), extending survival in treated mice; tumours without the marker saw no such benefit. That split is exactly what makes it useful — it points to who might respond, not just that something happens. A second study reported that CBD given beforehand reduced tumour burden and lowered markers linked to treatment resistance (IDO and PD-L1) in the tumour's stem-like cell population.[8,9]
Colorectal cancer models delivered the most precise finding in the whole animal dataset: a single gene decided whether CBD worked at all (p53 status — functional p53wt xenografts responded; p53-null xenografts did not). Tumours with a functioning copy of that gene shrank under CBD; tumours without it didn't respond. Separately, CBD was reported to reverse a cellular change linked to tumour invasion (epithelial-to-mesenchymal transition) and to reduce tumour size through suppression of a growth-signalling pathway (Wnt/β-catenin), and to work together with an immunotherapy antibody (anti-PD-1 checkpoint blockade) by reshaping the tumour's immune environment (M2-to-M1 macrophage repolarisation).[10,11,12]
In lung cancer, blocking a single protein (ICAM-1, via neutralising antibody) was enough to cancel out CBD's anti-metastatic effect completely — a level of precision that moves this from correlation to a demonstrated mechanism, not just an observed outcome.[13]
Even in pancreatic cancer, one of the hardest cancers to treat, CBD combined with THC was reported to extend survival in a model engineered to closely mirror human disease (the KPC genetically engineered mouse model), including a fully intact immune system. Associated shifts in bile acid metabolism and gut bacteria were also reported, though whether CBD or THC drove these changes isn't yet clear.[14]
Signal maturity: animal models currently provide the most consistent evidence that CBD affects tumour behaviour directly, spanning breast, glioblastoma, colorectal, lung, and pancreatic biology — but translation from mouse to human remains unproven, and no animal finding here has yet been confirmed in a human oncology trial.
In Vitro
Cell Model Data
Colorectal cancer cells reveal CBD's most important on-off switch: a single protein, already known to matter across cancer biology, determines whether CBD triggers cell death at all (p53 status — via induction of the pro-apoptotic protein Noxa). In cells where that protein functions normally, CBD reliably triggered a specific cell-death signal; in cells where it's missing or mutated, that signal didn't appear. This single distinction — not tumour subtype, not driver mutation status — was the strongest predictor of response across the colorectal work, and CBD's activity here was also reported to hold regardless of the cancer's usual driver mutations (KRAS, BRAF).[10,15,22,23,37]
Continue reading — full research detail+
In glioblastoma cells, CBD did something counterintuitive: it flipped a signalling pathway that tumours normally rely on to survive into one that suppresses them instead (NF-κB/RELA), by blocking a specific modification step the tumour needs (Ser-311 phosphorylation). A candidate marker for who might respond emerged from this work — tumours starting with lower internal oxidative stress (baseline ROS) appeared more sensitive — offering a possible way to identify likely responders before treatment. Separately, specific receptors shared with stem-cell markers (GPR55 and TRPV1, co-expressed with SOX2/OCT4) were identified as CBD's dominant target in the tumour's stem-like cell population.[16,24]
In breast cancer cells, CBD's signature effect is switching off a protein tied to invasive, metastasis-prone behaviour (Id-1) — the first time a non-toxic compound had been shown to do this — while simultaneously switching on a related protein linked to cells becoming less aggressive (Id-2). A second, independent effect triggered a cellular stress response (ER stress, via the PI3K/Akt/mTOR axis) that worked alongside programmed cell death rather than replacing it (beclin1-mediated autophagy).[17,6,38]
Ovarian cancer cells provide the cleanest single proof that CBD can starve tumour cells of the fat they depend on: CBD was reported to shut down the genes those cells use to build new fat (FASN, ACACA, SCD1, SREBP-1, via CB1 receptor signalling), confirmed by directly measuring depleted fat levels afterward. Restoring that fat from outside the cell reversed the downstream stress and cell-cycle effects entirely — evidence that the fat disruption, not something else, was driving the result.[18]
In pancreatic cancer cells, CBD was reported to block a growth-signal messenger (the long non-coding RNA MALAT1, upstream of PI3K/Akt/mTOR signalling), suppressing the invasive transition tumour cells rely on to spread.[40]
Signal maturity: cell-model research offers the deepest mechanistic detail of any evidence tier, identifying specific genes, proteins, and pathways CBD interacts with — but distance from the human clinical setting is greatest here, and mechanism alone does not confirm real-world benefit.
Advertisement
02 — Pathways
Pathway Interaction Profile
CBD engages several distinct biological pathways relevant to tumor behavior, grouped below by the functional role each one supports. This includes direct anti-tumour mechanisms and, further down, a separate set of pathways supporting the body's own resilience.
CBD's Contain classification rests on reported suppression of the groundwork tumours lay before they spread — inflammatory signalling that primes surrounding tissue, the vascular growth a tumour needs to feed a new site, the cellular transition that lets cancer cells detach and invade, and the stem-like state that lets dormant cells reactivate later.
Block Seeding & Niche Formation
Research concerning formation of supportive pre-metastatic niches at distant sites.
Angiogenesis / VEGF / HIF-1α
CBD has been reported to impair new blood vessel formation in tumour models, reducing endothelial cell migration and VEGF/HIF-1α signalling — the pathway tumours rely on to grow a blood supply into a new site.[19]
NF-κB / TNF-α / IL-6 inflammatory axis
CBD has been consistently reported to reduce NF-κB-linked inflammatory signalling in tumour and colorectal cancer models — the primary mechanistic basis for CBD's Contain classification, since this axis drives chronic tumour microenvironment priming.[20]
Prevent Tumour Cell Shedding
Research concerning invasion and escape from existing lesions (EMT and ECM breach).
EMT & metastatic invasion
CBD has been reported to reverse the cellular transition tumour cells use to detach and invade (epithelial-to-mesenchymal transition) across colorectal, lung, and breast cancer models, acting through distinct mechanisms in each — Wnt/β-catenin suppression, ICAM-1/TIMP-1 upregulation, and EGFR pathway blockade respectively.[11]
Prevent Dormant Reactivation
Research concerning wake-up signalling and reactivation of dormant disseminated tumour cells.
Cancer stemness (CD44, ALDH, Nanog/Sox2)
CBD has been reported to downregulate a protein linked to stem-like, treatment-resistant tumour cell behaviour (Id-1) in breast cancer and glioma models, while in glioblastoma specifically targeting receptors (GPR55, TRPV1) co-expressed with stem cell markers SOX2 and OCT4.[17]
CBD's Starve classification is anchored in reported disruption of how tumour cells fuel themselves — from the fat-synthesis machinery cells rely on to build new membrane, through direct interference with the mitochondrial machinery that generates energy, to controlled oxidative pressure that taxes a tumour cell's own defences.
Lipid Axis Pressure
Research concerning membrane synthesis and lipid-driven signalling capacity.
Lipogenesis (FASN / ACACA / SCD1 / SREBP-1)
In ovarian cancer cells, CBD was reported to shut down the genes tumour cells use to build new fat, confirmed by directly measured fatty acid depletion — the clearest standalone evidence of CBD acting as a Starve-pathway compound.[18]
Metabolic Flexibility Suppression
Research concerning metabolic adaptation and switching between fuel sources under pressure.
Mitochondrial Electron Transport Chain (ETC I–V)
CBD has been reported to act directly on mitochondrial Complex I and IV in colorectal cancer cells, generating oxidative stress that sits upstream of programmed cell death.[10]
Autophagy & lysosomal system
CBD has been reported to induce autophagy markers in colorectal and breast cancer models, with direction and outcome depending on context — in breast cancer this process works alongside programmed cell death, while in colorectal cancer it can partially protect cells against CBD's effects.[37]
Redox Buffering Taxation (Controlled)
Research concerning tumour-cell redox buffering and vulnerability to oxidative pressure, separate from host redox protection.
NRF2–GSH redox axis
CBD exposure has been reported to raise intracellular oxidative stress and strain tumour cells' own antioxidant defences — the mechanistic bridge between CBD's Starve classification and its downstream Attack-pathway activity.[21]
CBD's Weaken classification reflects reported attrition of the signalling tumour cells depend on to keep growing and tolerate internal stress — cellular stress responses, the central growth pathways tumours co-opt to keep proliferating, and — in one glioblastoma-specific case — a pathway CBD appears to flip from working for the tumour to working against it.
Attrition Pressure
Research concerning cellular stress vulnerability and net tumour-cell attrition under sustained conditions.
ER stress & unfolded protein response (UPR)
CBD has been reported to trigger a cell-stress cascade (PERK/ATF4/CHOP) leading toward programmed cell death; in colorectal cancer this response was reported to depend on KRAS mutation status, suggesting differential sensitivity by tumour genetics.[22]
Expansion Suppression
Research concerning proliferation, cell-cycle progression, and the capacity of lesions to add durable mass.
PI3K–AKT–mTOR
CBD has been reported to suppress this central survival pathway across ovarian, breast, colorectal, glioblastoma, and pancreatic cancer models, reaching it through different upstream routes in each cancer type.[18,40]
RAS–RAF–MEK–ERK (MAPK)
In colorectal cancer, CBD-driven activation of this pathway has been reported to push cells toward cell death rather than the growth signal it normally carries; computational modelling has also predicted direct CBD binding to several proteins in this cascade, though this remains unconfirmed biologically.[23]
Wnt / β-catenin
In colorectal cancer specifically — where this pathway is especially relevant given how often it's disrupted in colorectal tumours — CBD has been reported to suppress Wnt/β-catenin signalling, reducing nuclear β-catenin and reversing invasion-linked gene activity.[11]
Metabolic Weakening — GBM-specific
Research concerning tumour metabolic competence and adaptive capacity over time.
NF-κB / RELA functional conversion
In glioblastoma specifically, CBD has been reported to convert this pathway from a tumour-survival driver into a tumour-suppressing one, by blocking a specific activation step (Ser-311 phosphorylation); a candidate biomarker emerged from this work — tumours with lower baseline oxidative stress appeared more sensitive.[16]
CBD's Attack classification is corroborated by real animal-model evidence, not just cell studies — a p53-status-dependent xenograft response and an MGMT-methylation-dependent survival benefit in mice both directly confirm that these mechanisms operate in a living tumour, not only in a dish. The concentration gap covered under Pharmacokinetics and Administration below still applies, and no human oncology trial has tested any of these mechanisms directly.
Direct Tumour-Directed Killing
Research concerning regulated tumour-cell death (apoptosis, ferroptosis, necroptosis).
Intrinsic apoptosis (mitochondrial / Bcl-2)
This is the most consistently documented cell death mechanism across cancer types; in colorectal cancer specifically, this route was reported to depend on functional p53 — confirmed directly in animal models, where p53-wild-type xenograft tumours shrank under CBD and p53-null xenografts did not.[10]
DNA damage & repair / PARP
In glioblastoma tumours carrying a specific epigenetic marker (MGMT-promoter methylation), CBD has been reported to block a DNA-repair protein (RAD51), converting otherwise-survivable chemotherapy damage into lethal damage — confirmed in orthotopic mouse models, with a measured survival benefit absent in tumours lacking that marker.[8]
Ceramide / S1P axis
In glioma models, cannabinoid receptor engagement has been reported to trigger accumulation of a lipid signalling molecule (ceramide) that inhibits tumour survival signalling — though CBD's specific contribution versus THC requires careful attribution in mixed-formulation studies.[24]
Immune-Mediated Killing (Re-enabled)
Research concerning immune surveillance and cytotoxic execution capacity.
Immune checkpoints & myeloid skewing (M2/MDSC)
CBD has been reported to shift tumour-supportive immune cells (M2 macrophages) toward an anti-tumour state in colorectal and breast cancer models, and to work together with anti-PD-1 checkpoint immunotherapy through this immune reprogramming.[12]
CBD's Protect classification now covers two distinct kinds of evidence. One is clinical-outcome evidence tied specifically to cancer treatment itself, which has no defined mechanism by nature and is detailed below rather than carrying a pathway card — and here the human record is genuinely negative. The other is mechanism-based evidence that CBD supports the body's own tissue resilience independent of any drug interaction, which does carry a real pathway card, set out further down — though the one human test that measured this exact question found nothing, a tension this page states plainly rather than resolving one way or the other.
Oncology Host-Status
Symptom burden — the primary randomised trial (n=144) found no benefit on overall symptom distress versus placebo.[1]
Inflammatory biomarkers — a direct sub-study of that same trial cohort measured serial CRP and a cytokine panel through day 28 and found no difference between arms. The authors' own conclusion: unable to demonstrate an anti-inflammatory effect of CBD in cancer patients.[2] This rules out a specific, plausible alternative explanation for the primary trial's negative result — that CBD works biologically but the symptom scale used wasn't sensitive enough to detect it.
Combination pain benefit — a 1:1 THC:CBD combination showed a statistically significant but clinically modest pain-subscore improvement in a companion trial; no overall symptom-burden benefit, and CBD's specific contribution can't be isolated from THC's.[3]
Inflammatory Regulation
Human and preclinical research on systemic inflammatory regulation, distinct from immune-cell surveillance and organ-specific injury.
Systemic anti-inflammatory mechanism
CBD has been reported to reduce pro-inflammatory cytokine production (TNF-α, IL-6, IL-1β) in LPS-challenged mice, and to reduce cytokine and chemokine output in human microglial and keratinocyte cell models — a real, animal-corroborated mechanism, not purely a cell-dish finding.[27,28] This sits in real tension with the finding directly above: the one human test that measured this exact question — CRP and cytokine levels in cancer patients on CBD — found no effect. Both facts are true at once. Whether the animal mechanism simply doesn't translate to humans, or the cancer-patient trial wasn't the right test of it, isn't something the current evidence resolves.
Block Seeding & Niche Formation
Research concerning formation of supportive pre-metastatic niches at distant sites.
NF-κB / TNF-α / IL-6 inflammatory axis
Preclinical investigations report reduced activity of inflammatory signaling pathways commonly implicated in tumor progression and metastatic priming. Chronic inflammation sustains the tumor microenvironment; modulation of this axis is the primary mechanistic basis for CBD's Contain classification.
Prevent Tumor Cell Shedding
Research concerning invasion and escape from existing lesions (EMT and ECM breach).
CBD has been reported to reverse the cellular changes tumor cells use to detach and invade in colorectal, lung, and breast cancer models — one of the more consistently observed anti-invasion effects across cancer types, reached through distinct mechanisms in each.
Lipid Axis Pressure
Research concerning membrane synthesis and lipid-driven signalling capacity.
Lipogenesis (FASN / ACACA / SCD1 / SREBP-1)
In ovarian cancer cells, CBD has been reported to shut down the genes tumor cells use to build new fat — confirmed by directly measured fatty acid depletion, and the clearest standalone evidence of CBD acting as a Starve-pathway compound.
Metabolic Flexibility Suppression
Research concerning metabolic adaptation and switching between fuel sources under pressure.
Mitochondrial Electron Transport Chain (ETC I–V)
CBD has been reported to act directly on mitochondrial function in colorectal cancer cells, generating oxidative stress that sits upstream of programmed cell death — a consistently directional effect, unlike some of CBD's other metabolic interactions.
Expansion Suppression
Research concerning proliferation, cell-cycle progression, and the capacity of lesions to add durable mass.
CBD has been reported to suppress this central growth-survival pathway across ovarian, breast, colorectal, glioblastoma, and pancreatic cancer models — one of the more consistently observed effects across the preclinical record, reached through different upstream routes depending on tumor type.
Direct Tumor-Directed Killing
Research concerning regulated tumour-cell death (apoptosis, ferroptosis, necroptosis).
Intrinsic apoptosis (mitochondrial / Bcl-2)
Confirmed in animal models, not just cell studies: p53-status determines response directly in xenograft tumours, not only in a dish. The Bcl-2 family checkpoint is a key mediator. No human oncology trial has tested this mechanism.
Inflammatory Regulation
Human and preclinical research on systemic inflammatory regulation, distinct from immune-cell surveillance and organ-specific injury.
Systemic anti-inflammatory mechanism
CBD has been reported to reduce inflammatory cytokine production in animal and cell models — but the one human test that measured this exact question, in cancer patients, found no effect. Both facts are real; this page states them together rather than picking one.
Advertisement
03 — Pharmacokinetics
Pharmacokinetics and Administration
How CBD moves through the body — and what it competes with for clearance — has more bearing on real-world use than the mechanistic evidence alone. This is the practical layer that shapes formulation, timing, and co-medication decisions.
Absorption
CBD is lipophilic with low, variable oral bioavailability (roughly 6–19%) due to first-pass hepatic metabolism. A high-fat meal can raise peak concentration 2–5× and total exposure 2–4× versus fasted state. Consistency of plasma exposure depends heavily on formulation and co-administration with dietary fat.
The Concentration Gap
Most anti-cancer effects reported in cell studies require concentrations several times higher than what oral dosing can achieve in the bloodstream — even at the highest doses tested. A few effects sit close enough to matter; most don't. This gap is essential context for reading the In Vitro evidence above.
Clinical Dose Context
Human studies span a wide range — from roughly 20 mg/day in a self-titrated palliative-care trial up to 1,500 mg twice daily in pharmacokinetic studies. Most commercially available retail products deliver doses at the low end of this range — a relevant gap when interpreting oncology-adjacent research.
Formulation Effects
Plain oil capsules are the least efficient delivery option. Reformulated products — enhanced lipid capsules, self-emulsifying delivery systems, nanoemulsions — have been reported to reach meaningfully higher blood levels, though none has been shown to reach tumor tissue specifically.
Metabolism
Metabolized primarily by CYP3A4, CYP2C19, and CYP2C9. CBD also inhibits these enzymes — meaning it can elevate circulating levels of co-administered drugs that share these clearance pathways. This is clinically significant in polypharmacy settings.
Co-Dosing Considerations
CBD interferes with the clearance of many drugs metabolized via CYP3A4 or CYP2C19 — too many to list here. Pay particular attention to chemotherapy agents like temozolomide and to warfarin, both documented interactions with real clinical consequence. A full oncology drug panel review is recommended before use.
Absorption
The biggest practical constraint on using CBD orally isn't the mechanism — it's getting enough of it into the bloodstream in the first place. Oral bioavailability is low and highly variable, estimated at 6–19%, and gets worse rather than better at higher doses: at 3,000 mg, bioavailability drops to roughly 6.5%, so escalating the dose doesn't produce a proportional rise in blood levels (high first-pass hepatic metabolism; logP ~6.3).[29] Most of what ends up circulating isn't even the active molecule — the majority is 7-carboxy-CBD, an inactive metabolite; parent CBD and the pharmacologically active 7-hydroxy-CBD metabolite make up the smaller remainder.[29]
What a dose is taken with matters more than almost any other variable: a high-fat meal was reported to raise peak plasma concentration (Cmax) 2–5-fold and total exposure (AUC) 2–4-fold compared with a fasted state.[29] For anyone with an unpredictable appetite or dietary restrictions — a common reality during active cancer treatment — that food effect turns dosing into something closer to a moving target than a fixed number.
The Concentration Gap
This is the number that should anchor how everything in Evidence Summary above gets read: most of the anti-cancer effects reported in cell models require concentrations 3–12 times higher than what oral dosing, even at the top of the clinically tested range, can achieve in a patient's bloodstream.[31]
| Benchmark | Concentration | Interpretation |
|---|---|---|
| In vitro IC50 range (anti-cancer) | 2–20 µM | 630–6,289 ng/mL equivalent — the full spread from weakest to strongest reported effect |
| Most reported anti-tumour concentrations | 5–15 µM | 1,572–4,717 ng/mL equivalent — the narrower band where most published effects actually cluster |
| Cmax at 1,500 mg oral CBD (steady state, BID) | ~1.7 µM | 541 ng/mL — the highest concentration oral dosing has been shown to reach in humans[31] |
| GPR55 antagonism (colorectal) | 1–2.5 µM | Sits right at the edge of what maximum oral dosing can reach |
| Id-1 suppression (breast) | 1–5 µM | Lower end overlaps achievable Cmax; upper end does not |
A small number of effects sit close enough to achievable levels to matter: GPR55-mediated reduction in cell adhesion in colorectal cell models (1–2.5 µM) and Id-1 suppression in breast cancer cell models (1–5 µM) both fall within or near the range oral dosing can reach. Everything else in the in vitro record above sits meaningfully out of reach at standard oral doses.[7]
Clinical Dose Context
| Context | Dose | Source |
|---|---|---|
| OTC / consumer retail | 20–300 mg/day | Multiple observational |
| Advanced cancer palliative RCT (Hardy 2023) | ~20 mg/day median (self-titrated) | Phase IIb — negative result[1] |
| GBM adjunct (GEINO-1601) | THC+CBD up to 80 mg/day combined | Phase Ib — tolerability only[5] |
| GBM retrospective (Aviram 2022) | 400–600 mg/day oral | Uncontrolled case series[4] |
| FDA-approved Epidiolex (epilepsy) | Up to 20–25 mg/kg/day | Prescribing information[30] |
| Phase I PK, multiple dose (Taylor 2018) | 750–1,500 mg twice daily | Cmax ~541 ng/mL at 1,500 mg BID[31] |
The most precisely characterized dosing regimen in the human literature is 1,500 mg taken twice daily, studied specifically to define CBD's pharmacokinetics rather than any clinical endpoint (Taylor 2018). At steady state, that regimen produced a peak plasma concentration (Cmax) of approximately 541 ng/mL (~1.7 µM), total exposure (AUCτ) of approximately 3,236 ng·h/mL, a terminal half-life near 60 hours alongside a shorter effective half-life of 10–17 hours, oral clearance of 1,111–1,909 L/h, and an unusually large volume of distribution (20,963–42,849 L) reflecting extensive uptake into fatty tissue rather than staying in the bloodstream. Steady state itself was reached quickly — within about two days — with a modest 1.8–2.6-fold accumulation from twice-daily dosing.[31]
Formulation Effects
Formulation determines whether a clinically meaningful blood concentration is achievable at all — plain oil capsules are the least efficient option available. Conventional sesame or MCT oil delivers the 6–19% bioavailability described above and remains highly dependent on being taken with food. Reformulating the same molecule changes that picture substantially: an enhanced lipid matrix capsule was reported to reach 5.7 times higher Cmax and 3.3 times higher AUC than a standard unformulated isolate capsule under identical fed conditions, self-emulsifying drug delivery systems (SEDDS) reached up to 4.4 times higher Cmax, and nanoemulsion formulations reduced the fed/fasted swing and produced more consistent, person-to-person pharmacokinetics with a faster time to peak.[29,36] Sublingual delivery bypasses part of first-pass metabolism entirely; one piperine-containing pro-nano liposphere spray was reported to reach 4-fold higher Cmax than a standard oromucosal spray.[29,36]
None of these formulation strategies has been validated for reaching tumour tissue concentrations in humans — they change blood levels, not confirmed tumour exposure. This distinction matters for CBD specifically because of its own pharmacokinetics: the large volume of distribution described above reflects uptake into body fat generally, not confirmed concentration at a tumour site. A lipophilic compound distributing widely into adipose tissue is not the same as that compound reaching a tumour in comparable amounts.[29,31,36]
Metabolism and Pharmacogenomics
CBD's own clearance and its effect on other drugs run through the same small set of liver enzymes — the mechanistic root of most of its drug interaction risk. Phase I metabolism runs primarily through CYP3A4 and CYP2C19/CYP2C9 (7-hydroxylation), with smaller contributions from CYP2C8, CYP1A2, and CYP2B6; CBD inhibits these same enzymes to varying degrees (most strongly CYP2E1 and CYP2C19, more weakly CYP3A4 and CYP2C9), meaning it interferes with the very pathways that clear it from the body.[32] Phase II clearance runs through glucuronidation via UGT1A7, UGT1A9, and UGT2B7.[32]
Two factors were reported to meaningfully change how much CBD ends up in the blood for a given dose. Genetic variation in CYP2C9 (the *2 and *3 alleles, associated with reduced enzyme function) was associated with substantially higher plasma CBD levels — up to a 7-fold difference between individuals given an identical dose in the same study.[31,32] Separately, moderate-to-severe hepatic impairment was reported to increase CBD's total exposure (AUC) 2.5–5.2-fold — directly relevant given how often liver function is already compromised by primary liver cancer, hepatic metastases, or prior systemic treatment.[31,32]
Co-Dosing Considerations
CBD's enzyme inhibition creates real interaction risk across several categories of oncology-relevant medication. Each row is flagged by the most cautious guidance its cited evidence supports.
Discuss whether to combine, separate, or avoid CBD and a medication with your treating oncology team or physician.
| Flag | Interaction |
|---|---|
| Avoid | CYP3A4-metabolised chemotherapeutics and targeted therapies — CBD inhibition can unpredictably alter plasma exposure, increasing toxicity risk; taxanes (paclitaxel, docetaxel), EGFR inhibitors (osimertinib, erlotinib), venetoclax, imatinib, and sunitinib are of particular concern.[32] |
| Avoid | Anticoagulants (warfarin) — CYP2C9 inhibition by CBD leads to warfarin potentiation, elevated INR, and haemorrhage risk; a clinically documented interaction. INR monitoring is essential with any concurrent CBD use at meaningful doses.[33] |
| Avoid | Temozolomide — CBD inhibits UGT1A7, UGT1A9, and UGT2B7; TMZ is a UGT substrate, creating potential for increased TMZ plasma levels and myelotoxicity, relevant specifically in GBM treatment.[32] |
| Caution | Antiepileptic drugs (clobazam, lamotrigine, valproate) — well-characterised from epilepsy trials: elevated active N-clobazam metabolite, amplified valproate hepatotoxicity risk; relevant in GBM patients on antiepileptic prophylaxis.[32] |
| Caution | Checkpoint immunotherapy (pembrolizumab, nivolumab) — retrospective oncology data reported cannabis use during immunotherapy correlated with worse clinical outcomes in some cohorts; the mechanism is unclear and not established as causal, but it warrants discussion and monitoring.[35] |
| Caution | Opioid analgesics metabolised via CYP2D6 (codeine, oxycodone) — CBD inhibition can alter opioid metabolism, producing unpredictable analgesia or toxicity; a clinically relevant combination given how commonly opioids are used alongside CBD in palliative and supportive oncology care.[32] |
| Monitor | Statins metabolised via CYP3A4 (atorvastatin, simvastatin, lovastatin) — CBD may elevate statin plasma concentrations; LFT monitoring recommended. Rosuvastatin, not a CYP3A4 substrate, carries lower risk.[32] |
| Monitor | Corticosteroids (dexamethasone) — a bidirectional CYP3A4 interaction: CBD may alter dexamethasone levels, and dexamethasone may induce CYP3A4, reducing CBD exposure; relevant in GBM patients on cerebral oedema prophylaxis.[32] |
| Monitor | All hepatically metabolised drugs at CBD doses approaching clinical ranges — baseline and periodic ALT/AST monitoring recommended; risk is amplified in patients with hepatic impairment or hepatotoxic co-medications.[31,32] |
Advertisement
04 — Onset & Washout
Onset and Washout
CBD produces two distinct kinds of effect on two very different timelines. The distinction matters for interpreting the research, and for discussing timing with a clinician if you are also managing other medications or an upcoming procedure.
Immediate Onset
Direct effects on inflammatory tone and general comfort begin quickly but don't persist — daily consistency is what maintains them, not a single dose.
Steady State
With consistent daily dosing, plasma levels level off in about 2–3 days. Twice-daily dosing produces a modest build-up above single-dose levels — one more reason consistency matters more than any individual dose.
Accumulated Effect
Downstream effects on the tumor microenvironment and inflammatory signaling require sustained exposure to emerge, but appear to persist for some time after stopping.
Dosing Pattern in Studies
In the studies reporting benefit, effects built with steady daily dosing over time rather than occasional or as-needed use. This describes how CBD was studied, not a recommended regimen.
Washout
How long CBD's influence can take to clear before it stops being a relevant factor.
Earliest point to consider introducing a new medication. Interaction risk should be discussed with your care team as soon as possible — not held until this window.
A longer interval to allow before surgery, a medical procedure, or starting a medication with a narrow safety margin.
Two Distinct Clocks
CBD produces two mechanistically distinct kinds of effect, and they run on two very different timelines — collapsing them into a single "does it work" question is the most common way this compound gets misread.
The direct-pharmacology clock (Clock A) is fast and shallow: effects on endocannabinoid tone, CB2 signalling, and general inflammatory and comfort measures begin within hours of a dose. But they fade within a day, because CBD's short effective half-life (10–17 hours, detailed under Pharmacokinetics and Administration above) means a single dose doesn't linger in the bloodstream long enough to sustain a direct pharmacological effect.[31]
| Clock A — Direct Pharmacology | Clock B — Downstream Phenotype | |
|---|---|---|
| Latency | Fast — within hours | Medium — 7 to 14 days |
| Persistence | Short — fades within 24 hours | Days to weeks after stopping; not indefinite |
| What it covers | Endocannabinoid tone, CB2 signalling, inflammatory / comfort effects | Anti-invasion pressure, tumour-microenvironment modulation, pathway-level changes |
The downstream-phenotype clock (Clock B) is slow and deep: its effects — anti-invasion pressure, tumour-microenvironment modulation, the transcriptional and pathway-level changes detailed throughout Evidence Summary and Pathway Interaction Profile above — require sustained, repeated exposure to emerge at all. Once established, these effects appear to persist for some period after stopping, but they are not durable indefinitely; renewed exposure is likely needed to maintain the pressure. A single dose, or an occasional one, realistically only ever touches Clock A.
Steady State and Accumulation
With consistent daily dosing, CBD reaches steady-state plasma levels quickly — within about two to three days — and twice-daily dosing produces a modest 1.8 to 2.6-fold accumulation above single-dose levels, the same figures already detailed under Clinical Dose Context above.[31] Practically, this means the variable that matters is consistency, not any individual dose: a missed day or two doesn't just delay progress, it measurably erodes the plasma exposure the dosing pattern described below depends on.
Dosing Pattern in Studies
The underlying tension here is the same one that runs through this entire profile: the mechanistic case for CBD is well established (see Pathway Interaction Profile above), but the gap between what's mechanistically plausible and what's clinically achievable — detailed in the Concentration Gap section under Pharmacokinetics and Administration above — limits how much real-world confidence that mechanism deserves. That gap, not any single weak result, is what keeps CBD's oncology-relevant case moderate rather than strong.
The evidence points to CBD behaving as a continuity compound rather than a pulse-credible one. In the studies, single-day or occasional dosing realistically only reached Clock A's direct pharmacological effects (tone, inflammation, comfort), while the anti-metastatic and tumour-microenvironment effects that make CBD oncology-relevant required multi-day to multi-week continuity to emerge — and retail products dosed at the lower end of the range are unlikely to reach mechanistically meaningful exposure at all (see Clinical Dose Context above). This describes how CBD was studied, not a recommended dosing pattern.
Washout
Washout takes longer than CBD's plasma half-life alone would suggest, and the reason is the same one raised under Formulation Effects above: CBD doesn't just clear from the bloodstream, it also releases slowly from fat stores it accumulated in during use. In rats, adipose tissue was reported to hold CBD at concentrations roughly 10 to 100 times higher than liver or muscle tissue after sustained oral dosing.[39] No study has measured this directly in human tumour tissue, but the same lipophilic-accumulation behaviour is the working assumption behind why washout windows extend well past what plasma clearance alone implies. Tissue accumulation is also relevant to the central nervous system specifically, which matters directly for GBM given the mechanistic findings already described in Evidence Summary and Pathway Interaction Profile above.
Two thresholds are worth knowing separately. A minimum washout of 3–7 days is the earliest point to consider introducing a new medication that shares CBD's metabolic pathway — but interaction risk should be raised with a care team as soon as CBD use begins, not held until this window closes. A longer washout of 7–14 days is the interval to allow before surgery, a medical procedure, or starting a medication with a narrow safety margin, precisely because it accounts for the slower, variable release from adipose tissue rather than assuming plasma clearance tells the whole story.
Advertisement
05 — Safety
Safety Profile
CBD is generally well tolerated on its own, but its adverse effect profile and drug interaction risk carry more weight in an oncology context than in general use. The considerations below are what typically warrant closer attention.
Somnolence — dose-related sedation reported in clinical trials. Relevant in outpatient and combination therapy contexts.
Gastrointestinal disturbance — nausea, diarrhea, and appetite changes observed particularly at higher doses.
Hepatic enzyme elevations — dose-dependent ALT and AST increases documented, particularly at higher mg/kg ranges and in combination with other hepatically metabolized drugs.
CYP-mediated drug interactions — clinically relevant. Warfarin potentiation is documented. Risk extends to standard oncology agents sharing CYP3A4 / CYP2C19 clearance pathways.
Adverse Effects in Human Trials
The clearest safety signal from human cancer trials is a reassuring one at the individual-patient level: neither the Hardy 2023 advanced-cancer RCT nor the GEINO-1601 glioblastoma tolerability study reported any severe (grade 3–4) adverse event attributable to CBD alone.[1,5] What was reported instead were mild-to-moderate (grade 1–2), dose-dependent effects.
Somnolence and sedation were the most consistently reported, carrying particular relevance in outpatient oncology settings or when CBD is combined with opioids or other CNS-active medications. Diarrhoea, nausea, and appetite changes increased in frequency at higher doses — a non-trivial consideration in patients already managing gastrointestinal toxicity from systemic cancer treatment. Fatigue and asthenia were reported in 37.9% of patients in the GEINO-1601 trial, relevant specifically when CBD is added to a temozolomide-based regimen.[5] Dizziness and headache were more prominent with THC co-administration than with CBD alone.[3,5]
Hepatic Safety
Dose-dependent elevation in liver enzymes (ALT and AST) is the single most clinically significant adverse signal for CBD in an oncology context. The FDA's prescribing information for Epidiolex carries a hepatotoxicity warning based on this pattern, and rare cases of drug-induced liver injury requiring hospitalisation have been reported.[30] The clearest documented example of an additive risk is the valproate–CBD interaction, where both drugs independently carry hepatotoxicity risk and the combination compounds it — the same general principle extends to any other hepatically metabolised drug, not valproate specifically.[34]
This risk is not evenly distributed. As detailed under Pharmacokinetics and Administration above, hepatic impairment increases CBD's own plasma exposure 2.5 to 5.2-fold[31,32] — meaning patients with hepatic metastases, primary liver cancer, or otherwise compromised liver function face substantially higher exposure than the trial populations where these hepatic safety signals were first documented, not simply an equal risk added on top of a shared baseline. Baseline liver function tests and periodic monitoring are a prerequisite for use at doses approaching the clinical range.
06 — Sourcing
Sourcing Guide
Formulation is the biggest factor in whether a CBD product can deliver anything close to what the research above describes. Brand quality, ease of access and compound concentrations matter too. Our Sourcing Guide offers a curated list of products available on the retail market we found to answer all of those concerns.
CBD Sourcing Guide07 — Literature
References
Last reviewed: July 2026