CBD and Liver Enzymes: What the New JAMA Study Actually Shows

A clear-eyed look at the data, the media spin to come, and what patients and clinicians actually need to know

What You’ll Learn in This Post:

❇️ What the July 7, 2025 JAMA Internal Medicine study tested, and why its methodology matters

❇️ What the findings actually say—and don’t say—about CBD and liver enzyme elevations

❇️ Where this study fits among the broader CBD safety literature

❇️ How to interpret lab changes like ALT/AST elevations in clinical context

❇️ What this means for patients, clinicians, and regulators alike

A Study Worth Attention

On July 7, 2025, JAMA Internal Medicine published a rigorously designed clinical trial titled: “Cannabidiol and Liver Enzyme Level Elevations in Healthy Adults: A Randomized Clinical Trial” by Jeffry Florian, PharmD; Pablo Salcedo, PhD; Keith Burkhart, MD; Aanchal Shah, PhD; Lakshmi Manasa S. Chekka, PhD; Dro Keshishi, PhD; Vikram Patel, MD; ShanChao Yang, MD; Melanie Fein, MD; Ryan DePalma, MD; Murali Matta, PhD; David G. Strauss, MD; and Rodney Rouse, MD. (Read PDF here)

This randomized, double-blind, placebo-controlled trial studied 201 healthy adults who were given 5 mg/kg/day of purified CBD isolate (Epidiolex-grade) for four weeks. Weekly liver panels tracked changes in key enzymes—ALT, AST, ALP, and bilirubin.

The topline finding? 8 participants in the CBD group (5.6%) had ALT or AST elevations >3× the upper limit of normal. Zero placebo participants did. Yet no one developed symptoms, bilirubin abnormalities, or liver injury. All enzyme elevations reversed after CBD cessation.

Now let’s dissect what this study really tells us—and how it fits into the broader landscape of cannabinoid safety.

What This Study Got Right

In a field often flooded with retrospective surveys, unblinded cohort data, and inconsistent cannabis formulations, this trial stands out for its methodological discipline. The authors of “Cannabidiol and Liver Enzyme Level Elevations in Healthy Adults,” published July 7, 2025 in JAMA Internal Medicine, got several things exactly right.

1. It Was a True Randomized Controlled Trial

This was no observational study or secondary analysis. It was a prospective, randomized, double-blind, placebo-controlled trial—the gold standard of clinical research. The investigators pre-specified endpoints, maintained dosing consistency, and ensured blinding across all arms.

2. Dosing and Product Consistency

Rather than aggregating data from various unstandardized products, the study used Epidiolex-grade purified CBD isolate at a tightly controlled dose of 5 mg/kg/day, given as two divided doses. That translates to roughly 350 mg/day for a 70 kg adult. It was pharmaceutical-grade, devoid of confounders like THC, terpenes, or other cannabinoids.

3. High-Frequency, High-Fidelity Monitoring

Participants received weekly liver enzyme testing (ALT, AST, ALP, bilirubin) and daily electronic diaries to report symptoms and potential adverse events. This level of oversight is rare in cannabinoid research, where follow-ups are often months apart or missing altogether.

4. Transparent Safety Outcomes

Rather than simply noting enzyme changes, the researchers tracked whether these changes met recognized safety thresholds (e.g., >3× ULN), whether they were accompanied by other signs of liver stress (e.g., bilirubin elevation), and whether they resolved. The data were not sensationalized—just clearly reported.

5. No Evidence of Clinical Injury

Despite the enzyme elevations, there were:

• No Hy’s Law cases (ALT/AST elevation + bilirubin elevation = high risk of DILI)

• No symptoms or adverse clinical outcomes

• No persistent lab abnormalities

The enzyme elevations were asymptomatic, transient, and fully reversible.

What This Study Missed

For all its rigor, the study still leaves important gaps. That’s not a criticism—it’s the nature of early-phase research. Still, to responsibly interpret the findings, it’s worth understanding what the study didn’t explore, couldn’t answer, or chose not to measure.

1. It Didn’t Study Real Patients

The trial enrolled only healthy adults—no chronic illness, no prescription medications, and no pre-existing liver conditions. That makes sense for safety profiling, but it limits generalizability. The people actually using CBD for pain, anxiety, inflammation, insomnia, or epilepsy are often older, on polypharmacy, or managing chronic disease. This study tells us how CBD affects pristine systems, not real-world ones.

2. The Dose Was High—Uncommonly So

The participants received 5 mg/kg/day of CBD isolate, split into two doses—around 350 to 500 mg/day for the average adult. That’s far beyond what most consumers use, even at CED Clinic where we are keeping up with latest in patient care and following individuals carefully and over years. This is beyond what most patients receive, especially outside of epilepsy treatment. In typical practice, CBD doses range from 10 to 100 mg/day, and often include additional cannabinoids like THC or terpenes that may modulate metabolism and liver load. The findings here can’t be applied to those more common, lower-dose regimens.

3. It Didn’t Explore Functional or Mechanistic Outcomes

The study showed that ALT and AST increased in a small subset of participants. But why those elevations occurred remains unexamined. Were these participants slow metabolizers? Was there any associated oxidative stress? Inflammation? Mitochondrial changes? No metabolomics, cytokine profiles, pharmacogenomics, or functional testing was done to contextualize the enzymatic response.

4. It Used CBD Isolate Only

While isolate allows clean pharmacology, it doesn’t reflect what’s actually on shelves. Most over-the-counter products are full-spectrum or broad-spectrum, containing other cannabinoids (CBG, CBN, trace THC), terpenes, and botanical compounds. These may affect liver metabolism through entourage effects or CYP450 modulation. The study’s findings can’t be extrapolated to non-isolate products.

5. It Captured Only Short-Term Use

The trial lasted just four weeks. That limits what we know about longer-term liver effects, adaptation, or resolution over time. Many patients use CBD chronically—sometimes for years. Do transient enzyme elevations resolve and stay normal? Do they recur? Do they plateau? This study can’t say.

6. No Symptom Correlation

While the enzyme elevations were asymptomatic, the study didn’t assess whether people felt differently—better or worse—on CBD. Did sleep improve? Did anxiety worsen? Were there subclinical effects on cognition, energy, or digestion? None of that was reported, even though participants kept daily symptom diaries.

7. No Exploration of Inter-Individual Variability

Roughly 1 in 18 participants had significant enzyme elevations. Who were they? Did they share metabolic features? Genetic polymorphisms? Gut microbiome profiles? Alcohol use? Without that information, it’s impossible to assess which users may be at greater risk.

The Big Misunderstanding: Lab Changes ≠ Liver Damage

This is where most conversations around this study tend to unravel—confusing enzyme elevations with evidence of harm.

Let’s clarify:
Just because a liver enzyme rises doesn’t mean liver damage has occurred. In fact, mild ALT or AST elevations are incredibly common, transient, and often meaningless outside a clinical context. They can be caused by:

• A few days of ibuprofen or acetaminophen

• Drinking alcohol

• A new antibiotic

• Strenuous exercise

• Even fasting

Yet when the compound is CBD, those same lab shifts are interpreted with dramatic alarm. That’s not scientific. That’s selective skepticism.

This trial reported that 5.6% of healthy adults given high-dose CBD isolate developed ALT or AST elevations >3× the upper limit of normal. That’s the typical threshold for flagging possible drug-induced liver injury (DILI) in early-phase trials. But let’s look closer:

• Zero participants developed symptoms

• No bilirubin elevations occurred (which would suggest impaired clearance)

• No one met Hy’s Law criteria, the FDA’s red flag for liver toxicity

• All enzyme elevations resolved spontaneously, within 2–3 weeks after stopping CBD

If this were any other compound—especially one with FDA approval—these findings would be reported as reversible, asymptomatic lab variations. But because it’s CBD, the narrative often becomes: “CBD causes liver damage.”

That’s inaccurate, and irresponsible.

What This Study Does—and Doesn’t—Prove

✅ It shows that in healthy adults, high-dose purified CBD can cause temporary elevations in liver enzymes in a small subset of users.
❌ It doesn’t show that CBD causes permanent damage, clinical liver injury, or risk at typical doses.

The lack of clinical symptoms, bilirubin elevation, or persistently abnormal labs matters enormously.

Because liver enzymes are indicators, not verdicts. They fluctuate in response to a wide range of metabolic activity—not just damage. What matters most is:

• Magnitude of change

• Duration of elevation

• Presence of symptoms or concurrent lab changes

• Reversibility

In this case, the signal was mild, transient, asymptomatic, and reversed after stopping. That’s a very different story than “CBD is hepatotoxic.” (= toxic to the liver)

How This Study Compares to the Broader Body of CBD and Liver Safety Research

This trial—rigorous, well-controlled, and pharmacologically specific—is a welcome contribution to the field. But it’s not the first to explore this question, and its findings neither confirm a long-standing danger nor contradict the larger evidence base. To contextualize it properly, we need to compare it to what’s already out there.

What’s Consistent Across Studies

1. Liver enzyme elevations have been observed before
   The FDA’s original approval of Epidiolex (a CBD isolate) in 2018 was based on data from clinical trials in patients with epilepsy, including Dravet and Lennox-Gastaut syndromes. In those trials, liver enzyme elevations—particularly ALT and AST—were dose-dependent and more common in patients taking concurrent valproate, a known hepatotoxic antiepileptic.

   • FDA data showed up to 13% of patients had ALT/AST elevations >3x ULN when taking 20 mg/kg/day of CBD with valproate.

   • In patients not on valproate, the rate was 3%.

   This JAMA study’s rate—5.6% at 10 mg/kg/day—is in line with that existing data.

2. Liver abnormalities are typically asymptomatic and reversible
   Across multiple studies, the pattern is the same: ALT and AST may rise, but true liver injury is rare, and labs normalize upon discontinuation. Even in children and those with neurologic conditions, CBD-induced liver failure is not reported in the peer-reviewed literature.

3. Dose matters
   Higher doses—especially over 10–15 mg/kg/day—are more likely to trigger lab changes. In contrast, most commercial CBD products deliver doses in the range of 10–50 mg total per day, not per kg. That’s a 10– to 50-fold difference compared to the regimens used in this study and others like it.

What’s Different About This Study

1. It studied healthy adults—not patients
   Previous data came almost exclusively from pediatric epilepsy populations or people with serious underlying conditions. This study, by contrast, enrolled 201 healthy adults with no comorbidities or concurrent medications. That gives it cleaner internal validity, but limits external relevance to the people actually using CBD for therapeutic reasons.

2. It excluded real-world formulations
   While earlier studies focused on Epidiolex or even commercial oils, this study used pure CBD isolate in gelatin capsules. That rules out variability, but also misses the entourage effect—where other cannabinoids, flavonoids, or terpenes might influence absorption, metabolism, or risk.

3. Its short duration narrows interpretation
   Four weeks of use tells us something—but not much about long-term CBD safety. Most real-world users take CBD for months or years. We still lack data on chronic exposure at therapeutic doses, especially across diverse demographics and use cases.

4. It removed known confounders, which are part of real life
   This trial excluded people on medications, with liver disease, or with significant alcohol use. That allows cleaner attribution of effects to CBD—but real-world users often do have those risks. We still don’t know how CBD interacts with those variables in typical use scenarios.

📌 Bottom line:
This trial fits comfortably within the safety signal observed in earlier CBD studies—but it also narrows the question so much that its broader relevance becomes limited. It’s valuable for regulators and researchers, but only a partial answer to real-world questions about CBD safety.

What This Study Means for Clinical Practice and Consumer Guidance

This trial should be seen less as a warning sign and more as a benchmark: a tightly controlled study that tells us what can happen in a subset of people, under specific circumstances—not what does happen to most people using CBD therapeutically or casually.

Here’s how to think about the practical takeaways:

🧠 For Clinicians: 

1. Monitor when high doses are involved
   If a patient is using more than 200–300 mg/day of CBD isolate—or titrating up for seizure, severe anxiety, or chronic pain—consider checking baseline liver function tests (LFTs) and repeating them after a few weeks. Especially if the patient is taking other hepatically metabolized drugs (e.g., valproate, statins, azoles, SSRIs).

2. Contextualize enzyme elevations
   A modest bump in ALT or AST doesn’t equal toxicity. If bilirubin, ALP, and synthetic function (INR, albumin) are normal—and the patient is asymptomatic—it’s reasonable to monitor rather than stop CBD prematurely.

3. Personalize dosing discussions
   Emphasize that liver effects, when they occur, are dose-dependent and transient in the vast majority of users. Encourage patients to stick to lower effective doses and to increase only when necessary, ideally with support from a knowledgeable provider.

🌿 For Consumers:

1. Don’t panic about liver enzymes
   A small proportion of users at very high doses might experience elevated liver enzymes—but this doesn’t mean the liver is being harmed. In this study, no one developed liver disease, and all abnormalities reversed after stopping CBD.

2. Stick to appropriate doses
   Most over-the-counter products contain 10–50 mg per serving. That’s far below the doses studied here (300–500 mg/day). If you’re staying within that lower range and not taking other liver-sensitive medications, your risk is likely very low.

3. Read labels—and verify them
   Not all CBD products are created equal. Seek out brands that publish third-party Certificates of Analysis (COAs) with batch-specific dosing and contaminant data. Products with accurate labeling, GMP compliance, and low variability are safer and more predictable.

4. If in doubt, check your labs
   If you’re using high-dose CBD, or combining it with other medications, it’s reasonable to ask your doctor to run a liver panel. Most changes, if they occur, are minor and reversible—but it’s always better to be informed.

🧭 Overall Guidance

This study doesn’t tell patients to stop using CBD. It tells us that, like any compound with physiological activity, CBD can produce measurable effects in the body—some of which merit monitoring when used at pharmacologic doses.

The best response isn’t alarmism. It’s thoughtful, informed use—paired with transparency from manufacturers, education for prescribers, and regulatory clarity for all.

Final Thoughts: Science, Sanity, and CBD Safety

“Cannabidiol and Liver Enzyme Level Elevations in Healthy Adults: A Randomized Clinical Trial,” published July 7, 2025, in JAMA Internal Medicine, doesn’t offer a verdict on CBD. It offers a data point—a rigorously collected one—about what happens to liver enzymes in a small group of healthy people taking high doses of pharmaceutical-grade CBD isolate.

Here’s the grounded truth:

♦︎ Yes, a small percentage of users had elevated ALT or AST levels.

♦︎ No, there was no liver injury, no bilirubin rise, no clinical symptoms, and no lasting harm.

♦︎ Yes, the enzyme changes reversed after CBD was stopped.

♦︎ No, this does not apply to typical users taking far lower doses of full-spectrum products.

♦︎ And yes—we should pay attention. Not because CBD is dangerous, but because real-world cannabis care deserves the same scrutiny, structure, and respect as any other medical intervention.

This study is not a scandal. It’s a call for nuance:

✔️ A call for regulators to distinguish between low-dose, over-the-counter CBD and clinical-grade, high-dose therapies.

✔️ A call for physicians to move beyond reflexive fear and learn how to interpret cannabinoid pharmacology.

✔️ A call for patients to understand that plant-based care isn’t exempt from needing guidance, precision, or evidence.

✔️ And most of all, a call for a public health culture that doesn’t confuse lab noise with clinical signal.

⚡ In short: this study doesn’t tell us to stop using CBD. It tells us to start doing it smarter.

( Dr C: Here’s a new section I’m trying out…. )

🚨 Alarm Bells vs. Actual Evidence:

What Critics Will Say—And What the Study Really Shows

As media headlines start to circle, expect strong opinions. But not all takes are created equal. Here’s a breakdown of the most common alarmist claims we’re likely to hear—and the facts that quietly undercut them.

💣 Claim #1: “CBD Causes Liver Damage”

Counterpoint:
No participants in the trial experienced liver damage.

• No bilirubin elevation (a key marker of liver dysfunction)

• No symptoms like jaundice or abdominal pain

• No Hy’s Law cases (a red-flag combo of elevated ALT + bilirubin)

• No long-term harm—all enzyme elevations resolved after stopping CBD
  🔍 Lab changes ≠ liver injury. Context matters.

📈 Claim #2: “5.6% Had Dangerous ALT Elevations!”

Counterpoint:
True—8 out of 133 in the CBD group saw ALT or AST >3× ULN.
But that definition threshold is a screening tool, not a diagnosis.

• Tylenol, statins, alcohol, and vigorous exercise can also cause transient enzyme bumps.

• No participant got sick.

• The elevation was temporary, asymptomatic, and reversible.

🧪 Drug development guidelines flag this for monitoring—not panic.

🔬 Claim #3: “This Proves CBD Isn’t Safe”

Counterpoint:
This study shows that high-dose, pharmaceutical-grade CBD isolate can transiently affect liver enzymes in healthy people. That’s not the same as declaring CBD broadly unsafe.

• Typical CBD doses are 20–100 mg/day—not the 300–500+ mg/day used here

• Most real-world products are full-spectrum, not pure isolate

• Most users aren’t healthy volunteers—they’re patients using titrated doses for clinical needs

⚖️ Safety isn’t binary. It’s about dose, duration, product, and person.

🥼 Claim #4: “If 5.6% Had Reactions, CBD Shouldn’t Be OTC”

Counterpoint:
Regulatory thresholds for OTC products already tolerate low risks—when properly labeled, dosed, and monitored.

• Even widely used OTC meds (like NSAIDs or acetaminophen) carry known liver risks

• This study supports dose clarity and GMP manufacturing, not prohibition

• Proper oversight can manage these mild risks far better than prohibition ever will

📋 This argues for smarter policy—not stricter prohibition.

🧠 Claim #5: “Enzyme Bumps Mean Cumulative Damage”

Counterpoint:
The study lasted 4 weeks. What it showed was acute, self-resolving enzyme elevation with high-dose CBD—not cumulative toxicity.

• No worsening over time

• No trend toward organ dysfunction

• No signs of progression to liver disease

⏳ We need long-term studies—but there’s no evidence here of building harm.

🎯 Bonus: What the Study Didn’t Show

✔️ Nothing about full-spectrum or broad-spectrum CBD
✔️ Nothing about low-dose daily use
✔️ Nothing about patient populations with chronic illness
✔️ Nothing about cannabinoid synergy (e.g., CBD + THC)
✔️ Nothing about long-term liver outcomes

🧭 This trial answered a narrow question—let’s not let it be misused as a broad indictment.

🔍 Perspective Check: CBD vs. Common Liver Enzyme Offenders

If the idea of liver enzyme elevations sounds alarming, it’s worth asking: compared to what?

Elevations in ALT and AST—like the ones seen in this CBD trial—aren’t rare in medicine. In fact, several widely used and FDA-approved substances are known to trigger transient liver enzyme increases, often at similar or higher rates.

Here’s how CBD stacks up:

⚠️ In this context, the findings of the CBD trial are unremarkable. A 5.6% elevation rate with no associated symptoms or liver dysfunction places CBD in the company of many everyday substances—some of which are available over-the-counter or taken chronically without lab monitoring.

🎯 Takeaway: Mild enzyme bumps aren’t unique to CBD. What matters most is whether those changes are dose-dependent, symptomatic, reversible, and clinically significant. In this trial, the answer is reassuring across the board.

🔎 FAQ: CBD and Liver Enzymes