Cannabis and Chemotherapy: The Drug Interactions Oncologists Watch For

About a third of cancer patients use cannabis. About one in five tells their care team. That gap is where the actual danger in this subject lives, because the real risks are not the ones patients imagine, and every one of them is easier to manage when the oncologist knows.

This guide covers how common co-use is, the enzyme problem underneath the drug interactions, the immunotherapy controversy told honestly from both sides, and what the oncology profession itself now recommends.

The short answer

Cannabis is neither the cancer treatment some sellers claim nor the harmless sideline many patients assume. It shares liver enzymes with a long list of drugs, it carries a real infection risk when smoked during immunosuppression, and there is an unresolved question about immunotherapy that deserves more respect than either side usually gives it.

How common this actually is

The best count comes from a survey run across twelve National Cancer Institute designated cancer centers, and it found that 32.9 percent of patients had used cannabis since diagnosis, while only 21.5 percent reported having spoken to their health care providers about it.

An earlier single-center survey found 24 percent active users, three quarters of whom wanted information from their cancer team and mostly were not getting it there. The professional body's own framing puts use at 20 to 40 percent of adults with cancer.

The silence runs both directions. A national survey of oncologists found 80 percent discussing cannabis with patients and nearly half recommending it, while less than 30 percent considered themselves knowledgeable enough to make such recommendations.

Patients are not telling doctors, and the doctors being told do not feel equipped. That is the system this article is trying to patch.

The enzyme problem

Most drug interactions with cannabis run through one mechanism, and it is worth understanding once.

The liver clears a huge share of modern medicine through cytochrome P450 enzymes, especially one called CYP3A4. THC is metabolized by CYP3A4 and CYP2C9. CBD is metabolized by CYP3A4 and inhibits CYP2C19 and CYP3A4. When two drugs share an enzyme, each can slow the clearance of the other, and levels drift upward.

This is documented, not theoretical. The Canadian Medical Association Journal's review records that very high INR levels and bleeding have been reported with combined use of warfarin and marijuana, and that the transplant drug tacrolimus can rise roughly threefold alongside cannabinoids. The FDA label for prescription CBD warns that it produces a 3-fold increase in plasma concentrations of N-desmethylclobazam and tells clinicians to consider dose reduction for drugs cleared by a string of named enzymes.

Now note which drugs travel those same pathways: many oral chemotherapy agents, targeted therapies, anti-nausea drugs, steroids and blood thinners. A patient adding high-dose CBD oil to that mix, silently, is adjusting doses nobody chose.

The practical point is not that interactions make cannabis unusable. It is that they make disclosure non-optional.

The immunotherapy question, both sides

This is the sharpest controversy in the field, and most articles tell you half of it. The alarm came from Israel. A retrospective study of 140 patients on nivolumab, a checkpoint inhibitor, found the response rate fell from 37.5 percent on nivolumab alone to 15.9 percent in the nivolumab plus cannabis group. A follow-up prospective study found cannabis users progressing at a median of 3.4 months against 13.1 for non-users, with worse survival.

Those numbers are alarming, and they went around the world. Here is the other half.

Both studies were small and observational, and the cannabis users were sicker at baseline. A reanalysis reported that over a quarter of the published p-values could not be replicated and described gross inaccuracies. Then came bigger, better data: a pooled analysis of individual patient records from four Canadian immunotherapy trials, 684 patients, found no statistically significant difference in overall survival for patients taking or not taking cannabinoids at baseline. A 2026 commentary by two oncologists adds that subsequent studies have not demonstrated worse outcomes in patients on immunotherapy using cannabis. 

The honest summary: an early warning signal that larger and cleaner data has mostly failed to confirm, and a question still open enough that anyone starting a checkpoint inhibitor should have this exact conversation with their oncologist rather than resolving it from a blog, including this one.

The unexamined day is a wasted opportunity. Reflect on what you did, what you learned, and how you can improve.

John Dewey

What is well supported

Cannabis in oncology has one legitimately strong use case, and it is older than most patients realize.

Dronabinol, pharmaceutical THC, has been FDA-approved since 1986 for nausea and vomiting associated with cancer chemotherapy in patients who have failed to respond adequately to conventional antiemetic treatments. Nabilone holds the same indication. The National Cancer Institute's clinician reference states plainly that the FDA has not approved cannabis as a treatment for cancer or any other medical condition, while listing those approved cannabinoids for side effects.

The 2024 professional guideline draws the same two lines. It recommends against cannabis or cannabinoids as a cancer-directed treatment outside a clinical trial, and it finds the strongest supportive evidence for refractory chemotherapy-induced nausea and vomiting when added to standard antiemetics, delivered with a call for open, nonjudgmental communication.

Relief, plausibly yes. Cure, no. Anyone selling the second word is selling.

The smoking-specific risk

One risk in this subject has nothing to do with chemistry and everything to do with biology.

Chemotherapy suppresses white blood cells. Smoked plant material carries fungal spores: a classic study found 73 percent of cannabis cigarettes from chronic users grew Aspergillus, with spores passing easily into air samplers during smoking. A CDC claims analysis found cannabis users 3.5 times more likely to have a fungal infection.

For a healthy adult this is nearly irrelevant. For a neutropenic patient mid-cycle it is the difference between a habit and a hazard. During active treatment, anything inhaled and combusted is the wrong format, whatever your view of everything else in this article.

The conversation script

Since only a fifth of patients are having it, here is what the conversation needs to contain, and it takes two minutes.

What you take, including CBD oil from the supermarket, because CBD is the stronger enzyme inhibitor. How you take it, because smoked, vaped and oral carry different risks. How much and how often, because interaction risk scales with dose. And the ask: whether anything in the treatment plan, especially blood thinners, oral chemotherapy or a checkpoint inhibitor, shares a clearance pathway.

Oncologists cannot report you, and the guideline explicitly instructs them toward nonjudgmental communication. The worst realistic outcome of disclosure is a dose adjustment. The worst realistic outcome of silence is in the warfarin case reports.

Where a seed company honestly fits

Nowhere near treatment, and possibly somewhere near the year after it.

People rebuilding an appetite and a sleep schedule after treatment ends are a real part of our audience, and what they consistently want is gentleness and predictability, not power. That is a breeding direction, and it is the opposite of the potency race.

GG4 Auto is the heavier of our two picks here, GG4 crossed into our Super Auto line at 26% THC, 70 to 75 days from seed at 80 to 100 cm, creamy coffee and spice with a deep physical calm, and honestly labeled: this one is not for beginners. RS11 x Banana OG is the brighter option, a 50/50 at 32%, 65 to 70 days, grapefruit, melon and cherry over a relaxing, focused effect.

Neither is medicine, and nothing in this article is medical advice. Four decades of breeding has taught us who actually buys seeds in the months after a hard year, and what they thank us for later is never the THC number.

The short version

Roughly a third of cancer patients use cannabis and only about a fifth tell their care team, while most oncologists discussing it do not feel expert in it.

The interaction risk runs through shared liver enzymes, with documented cases of warfarin bleeding and tripled tacrolimus levels, and the FDA's own CBD label warning of threefold rises in co-administered drugs. The immunotherapy alarm from two small Israeli studies has not been confirmed by larger pooled trial data, and remains a live question for anyone on a checkpoint inhibitor.

Pharmaceutical THC has been approved for chemotherapy nausea since 1986, the 2024 guideline supports cannabinoids only for refractory nausea and recommends against any cancer-directed use, and smoking during immunosuppression carries a real fungal risk. Tell your oncologist. It is a two-minute conversation.

Barney's Farm has been developing premium cannabis genetics since the 1980s, with over 40 Cannabis Cup wins. Explore our full seed catalog and find strains bred for every climate and skill level.