CYP450 and CBD: the mechanism behind drug interactions

What CYP450 enzymes are (and why they matter)

The liver as a drug-metabolism factory

Most prescription drugs aren't excreted in their original form. The liver chemically modifies them — that's "metabolism" — before they're cleared through urine or bile. Cytochrome P450 enzymes do the bulk of that work. There are dozens of CYP enzymes; a handful do the heavy lifting on prescription drugs. Perucca 2020 reviews the CBD-specific picture in detail.

The big three for CBD interactions

CYP3A4 — metabolizes roughly 50% of all prescription medications. Examples: statins (atorvastatin, simvastatin), calcium channel blockers (amlodipine, diltiazem), benzodiazepines (alprazolam, diazepam), corticosteroids (prednisone, methylprednisolone, dexamethasone), some opioids (oxycodone, fentanyl), DOACs (apixaban, rivaroxaban), tacrolimus, cyclosporine, many antibiotics.

CYP2C9 — metabolizes warfarin, NSAIDs (celecoxib, meloxicam, ibuprofen, naproxen), phenytoin, glipizide, losartan partially.

CYP2C19 — metabolizes proton pump inhibitors (omeprazole, esomeprazole), some SSRIs (citalopram, escitalopram), clopidogrel (this one is a prodrug — CYP2C19 activation is required, so CBD inhibition is opposite-directional and reduces antiplatelet effect), clobazam.

The secondary players

CYP2D6 — metabolizes some beta-blockers (metoprolol, carvedilol), some SSRIs (fluoxetine, paroxetine), tamoxifen, codeine (prodrug, opposite-directional like clopidogrel), tramadol.

CYP1A2 — caffeine, theophylline, some antipsychotics. Less central to CBD interactions.

The non-CYP pathway: UGT glucuronidation

Some drugs are cleared via UGT (UDP-glucuronosyltransferase) enzymes rather than CYPs. CBD also affects UGT pathways. Drugs cleared via UGT include some opioids (morphine), bilirubin (relevant to liver-function readouts), thyroid hormones (minor), and methotrexate (partially). Most drug-interaction discussion focuses on CYP because CYP covers more drugs, but UGT inhibition matters for specific cases.

How CBD inhibits CYP enzymes

Competitive inhibition

CBD binds to CYP enzymes' active sites and competes with the prescription drugs that should be there. As long as CBD is present at relevant concentrations, the drug gets metabolized more slowly. This is "competitive" inhibition because the more drug present, the more it can outcompete CBD; conversely, the more CBD present, the more inhibition. Yamaori 2011 first documented this in vitro at clinically achievable concentrations; Doohan 2021 extended the finding across 12 cannabinoids, showing most inhibit CYP2C9.

The Bansal 2023 clinical evidence

The most-cited clinical evidence is Bansal 2023 — a randomized, double-blind, placebo-controlled trial in 18 healthy adults. CBD-dominant cannabis at roughly 750mg-equivalent/day raised CYP3A and CYP2C substrate plasma exposure 56–207%. That's not a marginal effect; it's clinically meaningful, and it confirms the in vitro mechanism translates to humans at research doses.

The breadth

Stöllberger 2023 reviewed 403 CYP-substrate drugs and found CBD has pathway overlap with 67–68% of frequently prescribed medications. Nachnani 2024 systematically reviewed 31 case-report studies covering 889 patients on CBD plus prescriptions and found 58% had unexpected serum drug levels after CBD initiation. Most patients on prescriptions are taking at least one CBD-affected drug.

The dramatic case — Geffrey 2015

For magnitude: Geffrey 2015 documented pediatric epilepsy patients on clobazam having clobazam plasma levels rise 60% and active-metabolite levels rise 500% after CBD initiation. The cohort was small and the CBD doses were high (epilepsy-research range), but the magnitude is what shared-pathway accumulation looks like in a vulnerable population.

Time course

CYP inhibition by CBD persists longer than CBD's plasma half-life. Even after CBD has been largely cleared from blood, the enzyme inhibition can persist for hours. Spacing CBD and a prescription doesn't fully eliminate the interaction concern.

Pharmacogenomic variation — why people respond differently

Genetic variants in CYP enzymes mean baseline metabolism varies meaningfully across the population. A "normal metabolizer" has the expected enzyme activity; "poor metabolizers" clear drugs more slowly at baseline; "ultra-rapid metabolizers" clear them faster. The same drug at the same dose has different plasma profiles in different people — and the CBD-induced inhibition layers on top of whatever baseline you started with.

The numbers that matter:

• CYP2D6: roughly 7% of European-descent populations are poor metabolizers (CYP2D6*4 is the most common allele). For CYP2D6-metabolized drugs (metoprolol, fluoxetine, codeine activation), poor metabolizers already have altered baseline drug exposure.
• CYP2C9: variants like CYP2C9*2 and *3 reduce metabolic capacity. Warfarin dosing genotype-guidance accounts for this; INR variability in CBD-warfarin combinations is plausibly larger in *2/*3 carriers.
• CYP2C19: similar variant structure; relevant to PPI and SSRI dosing.

Anderson 2022 documented a CBD-fluoxetine interaction modulated by CYP2D6*4 — the patient's poor-metabolizer status amplified the CBD effect to clinically meaningful magnitude. Practical implication: if you've had pharmacogenomic testing (especially for psychiatric or cardiovascular drugs) and know you're a poor metabolizer for any CYP, the CBD-Rx interaction picture is more cautious in your case.

How dose, format, and timing modulate the picture

Dose matters

Bansal 2023 used research doses (~750mg-equivalent/day). Wellness CBD use is typically 25–75mg/day — a fraction of the trial dose. The CYP inhibition is dose-dependent, so wellness doses produce smaller (not zero) effects. There is no published "no-effect threshold" — we don't know the exact dose below which CYP inhibition becomes clinically negligible — so honest practice is to treat any oral CBD dose as having some CYP impact in patients on CYP-substrate prescriptions.

Format matters more than people realize

Millar 2018's systematic review documents the bioavailability and first-pass profile:

• Oral CBD (swallowed) — highest first-pass effect through the liver. Bioavailability roughly 6%. CBD passes through the CYP system in concentrated form before reaching systemic circulation. Largest CYP impact per mg.
• Sublingual CBD (oil under the tongue) — bypasses some first-pass via dense sublingual capillaries. Bioavailability roughly 13–19%, more than double swallowed-oral. The bolus through the liver is smaller, so CYP inhibition per dose is smaller. This is the route Reclaim's oil is designed for (hold 60–90 seconds before swallowing).
• Topical CBD — minimal systemic absorption. The CBD applied to skin doesn't reach plasma levels that meaningfully inhibit liver CYPs. Lowest-risk format from a CYP-interaction standpoint.
• Transdermal patches — deliver CBD systemically but at lower peak plasma than oral. Sits between sublingual and topical for CYP impact.

Food matters

A high-fat meal increases CBD AUC roughly 5× and Cmax roughly 14× vs fasted (Taylor 2018). Higher peak CBD means more CYP inhibition. The food effect cuts both ways: it's why "take with food for absorption" matters for CBD's intended effects, and why timing matters for the CYP-interaction calculation.

Timing

CYP inhibition outlasts CBD plasma levels. Spacing CBD and a prescription by a few hours helps a little for absorption-window drugs (like levothyroxine) but doesn't fully eliminate CYP-pathway interactions for most drugs.

How to use this when reading per-drug pages

When you read any per-drug interaction page in our hub, ask these five questions:

1. Is the drug CYP-metabolized? Many BP meds, NSAIDs, statins, SSRIs, benzos, corticosteroids — yes. ACE inhibitors, biologics, dabigatran — no.
2. Which CYP enzyme? CYP3A4 is the biggest; CYP2C9 affects warfarin and NSAIDs; CYP2D6 affects some beta-blockers and SSRIs.
3. Are there other CBD-affected drugs in your stack? Stacking compounds the inhibition.
4. What's your route of CBD use? Topical = lowest impact; oral = highest.
5. Do you have known pharmacogenomic variation? If yes, the picture is more cautious.

Per-drug detail at: methotrexate, Celebrex, meloxicam, blood thinners, biologics, prednisone, and others on the hub.