Chemotherapy-induced nausea is the condition cannabis medicine was built on. Before the sleep blends and the wellness branding, there was a 1985 FDA approval for synthetic THC, handed to chemo patients who kept vomiting through every antiemetic their oncologists had. Forty years later, this is still the strongest evidence base in the field. A Cochrane review. Two FDA-approved cannabinoid drugs. A line in the NCCN oncology guidelines. And since 2024, a placebo-controlled phase III trial showing that adding oral THC and CBD tripled the complete response rate in patients who were failing modern antiemetics.
The dose that did it was 2.5mg of THC, three times a day. Half a standard dispensary gummy. The industry that prints "micro" on anything under 5mg has spent a decade implying that real effects start at 10, and the best-run cannabis trial in oncology quietly demonstrated the opposite.
So this page is the closest thing this library has to firm clinical ground. It also carries the strictest conditions of any condition page we publish, because the people it applies to are on drug regimens where interactions have real consequences. Read the interaction section before anything else if you take protease inhibitors.
What the evidence actually says
The 2017 National Academies report, still the most rigorous evidence review in the field, put chemotherapy-induced nausea and vomiting in its top category: conclusive or substantial evidence that oral cannabinoids are effective antiemetics. Only chronic pain and MS spasticity shared that tier. Everything else on this site sits below it.
The foundation is old, and it's worth being honest about how old. The 2015 Cochrane review (Smith et al.) pooled 23 randomized trials of dronabinol and nabilone. Patients on cannabinoids were nearly six times more likely to report complete absence of vomiting than patients on placebo (relative risk 5.7). Strong signal. The catch: those trials ran between 1975 and 1991, and not one of them compared cannabinoids against ondansetron or anything else an oncologist would reach for today. Beating the antiemetics of the Carter administration is a low bar.
Which is why the Grimison trial matters so much. Published in the Journal of Clinical Oncology in 2024, it enrolled 147 patients on moderately or highly emetogenic chemotherapy who still had refractory nausea or vomiting despite full modern prophylaxis. Dexamethasone plus a 5-HT3 antagonist in 97% of them, an NK-1 antagonist in 80%, olanzapine in 10%. These were the patients the good drugs had already failed. On top of all that, they took capsules of 2.5mg THC and 2.5mg CBD (or placebo) three times daily, from the day before chemo through day five.
Complete response, meaning no vomiting, no retching, no rescue medication for five days, went from 8% on placebo to 24% on THC:CBD. An absolute gain of 16 percentage points in the hardest population you could pick. In the earlier crossover phase, when patients were asked blind which cycle they preferred, most picked the cannabis extract, side effects and all.
That preference detail says more than the p-value. These are people who know exactly what a bad week feels like, comparing two of them.
The 2.5mg lesson
The trial dose deserves its own section because it contradicts almost everything dispensary shelf design tells you. Participants took 2.5mg THC per dose, balanced 1:1 with CBD, up to three times daily. Daily THC total: 7.5mg. A single Wana gummy contains more than that.
The FDA-labeled dronabinol regimen runs higher (5 mg/m² before chemo, repeated up to four to six times daily, which lands most adults at 8 to 10mg per dose), but dronabinol is pure THC with no CBD to blunt the edge, and the label exists precisely because that regimen makes a meaningful share of patients feel high, dizzy, or sedated. NCCN's breakthrough recommendation sits at 5 to 10mg every four to six hours. The Grimison protocol got its result at a fraction of that, with CBD along for the ride, in patients who mostly weren't cannabis users.
Even at 2.5mg, this wasn't free. Sedation hit 18% of the cannabis group versus 7% on placebo. Dizziness, 10% versus zero. Transient anxiety, 4% versus 1%. For a cancer patient already flattened by treatment, added sedation is a genuine cost, not a footnote. Anyone telling you low-dose THC has no side effects in this population hasn't read the safety table.
Timing is most of the game
Here is the part that trips people up in practice. Edibles are the right format for this condition and the wrong format for the moment most people reach for them.
An oral cannabinoid taken while you're actively vomiting has one likely destination, and it isn't your bloodstream. Every protocol that works treats cannabinoids as prevention taken on a schedule around infusion days. The dronabinol label: first dose 1 to 3 hours before chemotherapy. The Grimison protocol: start the day before, continue through day five. Nobody in any successful trial was unwrapping a gummy mid-crisis.
Onset supports the same logic. A traditional edible needs 60 to 90 minutes to work, sometimes longer during chemo when gastric emptying slows. Build levels ahead of the infusion and maintain them on a clock, whether or not you feel sick in the moment. If you need rescue during acute vomiting, that's what your prescribed rescue medications are for.
Drug interactions: the section that isn't optional
Disclosure matters more here than for any other condition we cover. Surveys keep finding that a large share of cancer patients using cannabis never tell their care team, usually out of vague embarrassment. Your oncologist has heard it before, prescribes a THC analog when needed, and would rather adjust your antiemetic plan than discover the interaction in your bloodwork.
Where the evidence thins out
Three honest gaps. First, anticipatory nausea, the conditioned kind that starts in the parking garage before the needle appears. The animal data (Parker's group has spent two decades on this) points at CBD and even non-psychoactive cannabinoid acids, but controlled human trials barely exist. If your nausea is anticipatory, you're extrapolating.
Second, dispensary edibles are not trial capsules. The evidence base runs on pharmaceutical dronabinol, nabilone, and a standardized 1:1 extract with batch-controlled content. A state-licensed gummy with a clean COA is a reasonable approximation of the Grimison capsules. A hemp-shop product of uncertain provenance is not, and after the federal hemp ban lands on November 12, the distinction stops being academic anyway.
Third, cannabinoid hyperemesis syndrome. Long-term heavy THC use can itself cause cycles of intractable vomiting, classically relieved by hot showers. If you were a daily consumer before diagnosis and your nausea reads as strange, cyclical, or resistant to everything, cannabis belongs on the suspect list rather than the treatment list. Tell your team about your use history either way.
What to do with all this
The pattern that mirrors the evidence: a low-dose edible at 2.5mg THC with matched CBD, started the day before infusion, taken three times daily through roughly day five, layered on top of your full prescribed antiemetic regimen, with your oncology team informed of everything. Products sold as 1:1 ratio gummies at 2.5mg or 5mg (cut in half) get you closest to the trial protocol. Skip anything marketed on potency.
This is the one condition where "the science supports it" is a plain statement instead of a stretch. It earned that status through four decades of trials at doses the recreational market would consider a rounding error, taken on a schedule, next to conventional medicine instead of in place of it. Use it the way it was studied and it's the most defensible edible use case there is.