Edibles and Immunosuppressants: What CBD Did to Tacrolimus and Everolimus, What Nobody Has Measured at a Gummy Dose, and Why the Transplant Clinic Owns the Test

On most of our interaction pages the dose gap is the calming part: the study used thirty gummies' worth of CBD, you eat one, and the fair advice is to tell the prescriber and stop worrying. Transplant drugs are where we stop offering that comfort. Tacrolimus, everolimus and sirolimus are dosed to a blood level inside a window a few nanograms wide, and the exits from that window are kidney damage on one side and rejection on the other.

A phase 1 trial at Indiana University measured tacrolimus exposure roughly tripling on CBD, the Epidiolex label prints 2.5-fold for everolimus, and since 2023 the Prograf label has carried its own section titled Cannabidiol Drug Interactions (Zortress and generic sirolimus have one too). What stays thin is everything below the trial dose. No controlled study has tested a nightly gummy against a transplant drug, and none has tested THC with one, by PubMed searches on November 7, 2026.

So our position is blunt. The stakes are high, the numbers at retail doses are thin, and one test settles the level question for one patient: a trough, which the transplant clinic already draws and can read against a range it already uses. The gummy is the part of the regimen nobody in that clinic can see unless someone says so. (Symptom evidence lives on our Crohn's, ulcerative colitis and rheumatoid arthritis pages, the wider enzyme map on our CBD interactions hub.)

Why does a small blood-level change matter with these drugs?

Because the dose is set by a blood level and the window is narrow. The Prograf label gives kidney recipients on mycophenolate a tacrolimus trough range of 4 to 11 ng/mL, and Zortress gives everolimus 3 to 8. The top of each range is under three times the bottom, so a 2.5-fold rise crosses nearly all of it.

Section 2.6 of the Prograf label ties both toxicity and efficacy failure to the trough, and names adding or stopping an interacting drug as a reason to test more often. Absorption is the weak point: only 17 to 23 percent of a swallowed dose reaches the blood in transplant patients, by section 12.3, so anything interfering in the gut has a lot of drug to let through. If a 3.1-fold rise carried straight across (a single-dose study in healthy people cannot promise that), a kidney recipient at 5 ng/mL lands at 15.5 against a ceiling of 11. At 2.5-fold, an everolimus patient at 5 lands at 12.5 against a ceiling of 8.

It also runs the other way. A clinic that cuts the dose around a CBD habit has built that habit into the regimen, so quitting or switching brands can drop the level, and low is the rejection side. Rein's 2020 nephrology review flags this for intermittent use and inaccurate labels. In one German report a child's everolimus level swung between 1.7 and 12.3 µg/L on a fixed dose as the CBD changed.

What does CBD do to tacrolimus, everolimus and the other transplant drugs?

It raises most of them. A phase 1 trial tripled tacrolimus exposure, the Epidiolex label reports everolimus up 2.5-fold and mycophenolic acid up 35 percent, and troughs rose in 76 percent of 25 patients on everolimus or sirolimus. Cyclosporine has a stable small series, one case report and little else.

The table lists each drug, what happened to its level, the dose measured, the evidence and the source.

The completed trial posted results on ClinicalTrials.gov on September 30, 2026: 2.9-fold in 19 people who do not express CYP3A5, 3.8-fold in 9 who do, and about 2.5-fold in the 10 with chronic kidney disease, the group nearest a transplant clinic. A PubMed search on November 7 found no peer-reviewed paper on those final numbers yet.

Elsewhere on this site we use the label's midazolam result to wave away CYP3A4 scare lists, and it stands: the probe did not move at 750mg of CBD twice daily. It does not rescue tacrolimus, which tripled at a lower dose. In So's trial the peak jumped and the half-life held, which points to more drug getting in through the gut, and the authors reach for grapefruit as the comparison. They cannot yet say whether gut CYP3A4, CYP3A5 or the P-gp pump is responsible, though their test-tube work found CBD hit CYP3A5 harder. Prograf's label tells patients to avoid grapefruit outright in section 2.1. Cannabidiol gets close monitoring and a possible dose cut in 5.17 and 7.3.

Mycophenolate is the odd one out. CellCept's label reports the same 35 percent rise for isavuconazole and asks only for monitoring, never mentions cannabidiol, and sets no target blood level. It asks for blood counts, so the trough argument does not carry over.

Does a CBD gummy or a THC edible change transplant drug levels?

Nobody has measured it. No controlled study has tested a gummy-sized dose against any transplant drug, by a PubMed search on November 7, 2026. The tacrolimus trial averaged 826mg of CBD a day, or 33 gummies at 25mg. Below that sit a small kidney series, two edible case reports and a handful of Reddit posts.

Green Roads' Relax Berries, 25mg apiece on our CBD gummy ranking, are the yardstick. The everolimus dose is 70 of them a day for a 70kg adult, and the mycophenolate dose is 42. That gap looks comforting until you read So's framing: the authors put the median over-the-counter dose at 0.6 mg/kg a day, 42mg at 70kg, with a range reaching 5.6 mg/kg, or 392mg, and they leave open whether the interaction survives there. Cuñetti's Uruguayan kidney recipients took 4 to 12 gummies' worth for three weeks, and the authors called tacrolimus levels variable and asked for longer follow-up.

The two edible case reports involve THC products and both are messy. Hauser's patient, a man recovering from a stem cell transplant, took gummies a family member brought in. After a switch to oral tacrolimus his level hit 43.8 and then 45.8 ng/mL, and he spent three days in intensive care. He was also on posaconazole, an antifungal the Prograf label answers with a one-third tacrolimus dose. Moadel's patient, a liver recipient who had taken lozenges up to her transplant, developed encephalopathy, and the team had no explanation until she produced the bottle. She denied using them during the admission. Neither case proves the cannabis did it. Bansal's 20mg THC brownie inhibited none of five probe enzymes, and P-gp was not among them.

The closest thing to a gummy-range CBD reading comes from strangers online. Allen and colleagues searched 63,561,233 Reddit posts and found 190 mentioning both an immunosuppressant and a cannabis product. Sixty-six voiced worry, including someone whose partner on Prograf wanted to try edibles and hoped to talk first to an experienced user with a heart transplant, which is a narrower specialty than most clinics list. Four reported a problem, among them a Prograf level of 30 on about 4mg blamed on CBD, and sirolimus climbing on 15mg of CBD and 12mg of cannabis a day while tacrolimus held. None of it can be verified, and no trial has gone there.

Does cannabis raise infection risk for people on immunosuppressants?

Smoked cannabis carries the case reports, and edibles barely appear. Invasive mold infections have been tied to smoking in transplant recipients, and in 2025 genome sequencing matched one patient's fungal infection to that patient's own medical cannabis. A 2026 review in Clinical Infectious Diseases calls the size of the risk unknown.

The transplant reports are few: fatal aspergillosis in a marrow recipient (Hamadeh 1988), treated aspergillosis in a kidney recipient (Marks 1996), and a kidney recipient with a marijuana history who lost the graft to mucormycosis (Hoerschgen 2023). The 2025 match was Cryptococcus in a myeloma patient, and the Pittsburgh group's review says it remains the only conclusive link and that US states have no uniform standard for microbial testing. In 2016 insurance claims, people coded for cannabis use had fungal infections 3.5 times as often after adjusting for age and immune status: 40 of 53,217 against 6,294 of 21,559,558. Absolute numbers are small, 40 percent of the infected cannabis users also had tobacco codes, and claims cannot say where a fungus came from.

Route is the difference. Gwinn's 2023 review, citing the American Society of Transplantation's paper, puts 89 percent of reported fungal infections after smoking and 4 percent after edibles. A PubMed search on November 7, 2026 pairing invasive fungal disease with edible or oral cannabis returned four records, and the only patient among them had smoked. That does not make a gummy sterile (the same review says few reports have checked how common mycotoxins are). The logic reaches autoimmune patients too, since the Humira and Remicade boxed warnings both name aspergillosis.

CBD is also an immune drug in its own right, which a trough cannot see. Blood cells from 23 volunteers on CBD up to 5 mg/kg twice daily proliferated less when stimulated in the lab, T cells by 37 percent, adding to tacrolimus's effect, with IL-2 and TNF-alpha down and IL-6 up (Gisch 2026). An Israeli phase 2 gave 300mg a day to 48 stem cell recipients on cyclosporine and methotrexate and reported a hazard ratio of 0.3 for grade II to IV acute graft-versus-host disease against 101 historical controls, with no randomized follow-up in the same day's search. The Epidiolex label lists infections in 40 to 41 percent against 31 on placebo, in severe epilepsy at 10 to 20 mg/kg a day. Whether that stacking helps a transplant patient or hurts one has no answer in these papers.

What do transplant programs and outcome studies say about cannabis?

They disagree, which matters to anyone waiting for an organ. In a 2018 survey of 49 US liver programs, 14 percent transplanted active marijuana users and another 28 percent listed users who stopped by surgery. Cohorts split as well: no harm in 1,225 kidney recipients in Maryland, and more biopsy-proven rejection in 2,091 in Toronto.

The Toronto hazard ratio was 1.55. A 2026 matched study of heart recipients found rejection at 1.50 with mortality and infection rates similar, and in Medicare data on 52,689 kidney recipients a cannabis dependence or abuse diagnosis in the first year carried about double the graft loss, which those authors say partly reflects other behaviors. None of these abstracts breaks out edibles or separates a drug effect from missed doses. Heart transplant clinicians surveyed in 2024 had no consensus, and the American Society of Transplantation found approval rules differing between organs at the same institution. In one Colorado survey of 538 liver recipients, 23.8 percent used marijuana, 55.3 percent of them by eating it. The Maryland cohort counted users partly by urine screen, and our drug-testing page owns how long an edible stays detectable.

Do edibles interact with methotrexate or biologics?

No human study has measured it. PubMed searches on November 7, 2026 found no pharmacokinetic study pairing a cannabinoid with methotrexate, azathioprine, a TNF blocker or a JAK inhibitor. Antibodies are proteins rather than pills, methotrexate's overlap with CBD is the liver, and for these patients the live infection question is smoke.

The nearest methotrexate papers are a mouse study (Xu 2026) and a rat study (Ilhan 2025), both friendly to CBD and a long way from a rheumatology clinic. None of the methotrexate, Humira, Remicade or Xeljanz labels mentions a cannabinoid. Our rheumatoid arthritis page holds the liver-test detail and our Crohn's page the infliximab and adalimumab labels. Two wrinkles: tofacitinib is a pill whose label runs metabolism mainly through CYP3A4 with a minor CYP2C19 share, and Neoral carries a rheumatoid arthritis indication, which puts those patients back in the first half of this page.

What the transplant team can use

Whether to order an extra trough is the clinic's call, and the Prograf, Zortress and sirolimus labels all give it grounds to do so. Four details from the patient give it something to read against the range it already uses.

1. Which immunosuppressant and the last trough, flagged if it is tacrolimus, sirolimus or everolimus.
2. CBD and THC in milligrams per day, by product, and whether it is a gummy, an oil or a cream.
3. The start date, any stop or brand switch, and whether it goes down with food.
4. Any azole antifungal, plus any new diarrhea, which So lists among the things that move tacrolimus and the Epidiolex label reports in up to 20 percent of patients.

How to open that conversation is covered in our guide to using edibles for medical purposes.