Compounds › Verified against primary sources

Dose adjustment for hepatic function

Here it is not just a gap: there is documented harm, with numbers and outcomes — and it hits exactly the people this site attracts

Partially approvedVerified against primary sources

Experimental material. Read before using anything from here.

Nothing on this page is a medical recommendation, prescription or treatment plan. It is an organized translation of protocols circulating in research communities and, where it exists, of what published trials tested. The two are marked differently — and they are not equivalent.

Most of the compounds here have no FDA approval for human use. Several are sold labeled "research use only", which means they have not gone through purity, sterility or dosage controls for human consumption. A community-reported dose is not a validated dose: it is what someone reported having done.

Talk to a licensed health professional before considering any of these compounds. If you already use one and feel anything unexpected, seek care — do not wait for the next routine test.

Specific attention for this compound: Two things change management and are almost never said. First: in cholestatic injury — the kind from SARMs and anabolic steroids — ALT can return to normal while bilirubin worsens; monitoring only AST and ALT gives false reassurance. Second: the typical latency is 2 to 3 months, so finishing a short cycle without symptoms is not proof of safety, and jaundice can appear after you have already stopped. Dark urine, pale stools, itching or yellowed eyes: stop and seek care the same day.

Summary

The renal page's counterpart, by the same method: section 8.7 extracted from the openFDA label endpoint and transcribed. Semaglutide and tirzepatide require no adjustment; bremelanotide was not evaluated in Child-Pugh C; tesamorelin declares the gap. The inversion is sparsentan — a kidney drug, with a black box warning for hepatotoxicity and the instruction to avoid it in any degree of hepatic impairment. The page shows why nobody classifies themselves in Child-Pugh alone (two of the five components are physical examination, not laboratory), brings the trap of one hour of weight training leaving AST and ALT abnormal for seven days in healthy men — with normal bilirubin, GGT and alkaline phosphatase, which is the criterion that separates muscle from liver — and gathers the real harm: 130 of the 839 DILIN cases came from supplements, 45 of them from bodybuilding, with a median of 91 days of jaundice. And the SARMs, with cholestasis, latency of 2 to 3 months and a reported bilirubin peak of 41.5 mg/dL.

The renal page's counterpart — and why this side is worse

The dose adjustment by renal function page ended by saying it did not cover the liver. This one does, by the same method: section 8.7 of each package insert, extracted from the FDA label endpoint and transcribed without cuts.

The difference between the two is large and uncomfortable. On the renal side, the survey mostly returned gaps: where there was data, it said to adjust nothing. Here there are gaps too — but beyond them there is documented harm, with name, number and outcome. And it hits this site's audience squarely: young men, taking things bought online to gain muscle.

There is also an inversion worth recording before anything else: the only compound cited on this site that the package insert says to avoid in any degree of hepatic impairment is a kidney drug.

What the package insert says, literally

Extracted from the same endpoint, on September 4, 2026, section 8.7 of each label.

Compound	Product	What the package insert determines	Detail
Semaglutide	Ozempic / Rybelsus	The recommended dose in patients with hepatic impairment is the same as for those with normal hepatic function.	In a study with different degrees of hepatic impairment, no clinically relevant change in pharmacokinetics was observed
Tirzepatide	Mounjaro	No dose adjustment is recommended.	In a clinical pharmacology study with varying degrees of hepatic impairment, no change in pharmacokinetics was observed
Bremelanotide	Vyleesi	No adjustment in mild to moderate hepatic impairment (Child-Pugh A and B, score 5–9).	Not evaluated in severe hepatic impairment. Use with caution in Child-Pugh C (score 10–15), because these patients may have an increased incidence and severity of adverse reactions
Tesamorelin	Egrifta SV	Not established.	The same passage that declares the renal gap declares the hepatic one: the pharmacokinetics in patients with renal or hepatic impairment have not been established
Sparsentan	Filspari	Avoid use in patients with any degree of hepatic impairment (Child-Pugh A to C), due to the potential risk of severe liver injury.	Has a black box warning for hepatotoxicity and is only distributed under a restricted program, with mandatory enrollment of the prescriber, the dispenser and the patient

The black box warning on the kidney drug

Sparsentan is one of the six drugs on the kidney disease page — approved for IgA nephropathy and, since 2026, for focal segmental glomerulosclerosis. Its black box warning does not talk about the kidney. It talks about the liver, and the sentence from the label deserves to be read in full:

"Some endothelin receptor antagonists have caused elevations of aminotransferases, hepatotoxicity, and liver failure."

Note the subject of the sentence: it is not this drug, it is the class. The risk was inherited from chemical relatives, and the label assumes it as a precaution before the product has accumulated cases of its own. It is the exact opposite of the reasoning that circulates about research peptides, where the absence of reported cases is presented as proof of safety — when it is almost always just absence of surveillance.

And there is the lesson of scale: a drug with a phase 3 trial, an agency, a package insert and a mandatory follow-up program still carries liver risk to the point of requiring patient enrollment. Approval is not absence of risk; it is risk measured, named and watched.

Why you cannot classify yourself

Every package insert above speaks of Child-Pugh A, B or C. This is where the hepatic side separates from the renal one in a way almost nobody notices.

For the kidney, the classification comes out of a calculation: creatinine, age and sex go into the CKD-EPI equation and a number comes out. The lab already prints the eGFR on the report. For the liver, there is no equivalent. Child-Pugh is a clinical score, and two of its five components are not lab tests — they are findings from physical examination and neurological assessment.

Component	1 point	2 points	3 points	Where it comes from
Bilirubin	below 2 mg/dL	2 to 3 mg/dL	above 3 mg/dL	Blood test
Albumin	above 3.5 g/dL	2.8 to 3.5 g/dL	below 2.8 g/dL	Blood test
INR	below 1.7	1.7 to 2.2	above 2.2	Blood test
Ascites	absent	mild	moderate	Physical exam / imaging
Encephalopathy	absent	grades 1 and 2	grades 3 and 4	Clinical assessment

The practical consequence of this

Score of 5 to 15 points: 5–6 is class A, 7–9 is class B, 10–15 is class C.
Three of the five components you can order; two, you cannot. Ascites and encephalopathy depend on someone examining the person. There is no "getting your Child-Pugh at the lab" the way you get an eGFR.
That is why the package insert that says "Child-Pugh C" is talking to the doctor, not to you. Whoever tries to classify themselves alone will use only the part that can be measured — and the missing part is precisely the one that indicates advanced disease.

The symmetrical trap: weight training makes the liver test look terrible

On the renal page, the trap was creatinine inflated by muscle mass and creatine. Here it has a bigger, more dramatic sibling.

According to PubMed, a study published in the British Journal of Clinical Pharmacology took 15 healthy men, used to moderate physical activity but not to weight training, and applied one hour of strength training. Five of the eight parameters measured — AST, ALT, LDH, CK and myoglobin — rose significantly (P < 0.01) and remained elevated for at least seven days. The article's title is not cautious: muscular exercise can cause highly pathological liver function tests in healthy men.

One hour of training. Seven days of abnormal tests. In people who had nothing.

The finding that saves the reader: what did NOT rise

The same study records that bilirubin, gamma-GT and alkaline phosphatase remained within the normal range. That is not a detail: it is the criterion that separates muscle from liver.

When AST and ALT rise together with CK and myoglobin, and bilirubin, GGT and alkaline phosphatase stay normal, the likely origin is muscular. When bilirubin and alkaline phosphatase rise, the conversation is a different one — and it is the liver's.

The classic mistake here runs in two directions, and both are costly. One is the scare: a trained person gets tested on Monday after Saturday's workout, sees abnormal AST and ALT and spends weeks thinking they have hepatitis. The other is worse and is the one that matters to this site: a person who really has liver injury from something they took attributes the abnormal test to training and keeps taking it.

Where real liver damage appears in this audience

Here the numbers stop being about gaps and become about counted cases, with outcomes.

According to PubMed, the Drug-Induced Liver Injury Network — a network of eight referral centers in the United States — prospectively followed 839 cases of liver injury from drugs or supplements between 2004 and 2013. Of these, 130 (15.5%) were attributed to herbals and dietary supplements, and the proportion rose from 7% to 20% over the study period (P < 0.001).

Agent	Cases	Profile and outcome
Drugs	709	3% death or transplant
Bodybuilding supplements	45	Prolonged jaundice — median of 91 days — in young men. No deaths and no transplants
Other supplements and herbals	85	Hepatocellular injury, predominantly in middle-aged women. 13% death or transplant — more severe than that of drugs

How to read this table without picking the convenient half

Bodybuilding supplements killed nobody in that series — and that is not reassuring. Ninety-one days of jaundice is the median, not the worst case. Half the men stayed yellow for more than three months.
The supplement that kills the most is not the bodybuilding one. It is the other group, of herbals and wellness products, with 13% death or transplant — four times the rate of prescription drugs. The intuition that "medicine is dangerous, supplements are mild" is inverted in these data.
The proportion nearly tripled in nine years. From 7% to 20% of liver injury cases seen in a referral network. It is not a stable phenomenon: it is a rising curve.

SARMs: the best-documented case, and it belongs to this site

This site already has a SARMs page. LiverTox — the hepatotoxicity database of the National Institute of Diabetes and Digestive and Kidney Diseases, part of the NIH — maintains an entire entry on them. What follows was checked in the text of the entry itself, not in a summary of it.

The first finding dismantles the reassuring reading of the clinical trials. In them, SARMs were described as well tolerated — but at the higher doses, the ones that most increased lean mass, there was elevation of aminotransferases in 5% to 21% of participants, with cases of isolated ALT requiring discontinuation. No case of jaundice appeared in those trials, and LiverTox says why: the duration of therapy was short, patients were monitored regularly, and a small proportion discontinued treatment because of the ALT elevations.

Jaundice appeared later, outside the trials — in people using SARMs on their own for bodybuilding, without medical supervision.

Feature	What is documented
Time to onset	Latency typically 2 to 3 months, ranging from a few weeks to one year
How it starts	Fatigue, loss of appetite, weight loss, abdominal pain and itching, followed by dark urine and jaundice
Tests at the start	Total bilirubin only moderately elevated (4.0 to 8.0 mg/dL), aminotransferases 2 to 5 times the upper limit of normal, and alkaline phosphatase and GGT minimally elevated or normal
How it progresses	Bilirubin rises while aminotransferases fall, and alkaline phosphatase rises little. The test result depends on how much of the disease course had already passed when it was drawn
What the biopsy shows	Moderate to severe canalicular cholestasis, with mild or minimal inflammation and no bile duct loss — the so-called bland cholestasis, the same picture as anabolic steroid jaundice
When bilirubin passes 30 mg/dL	Renal dysfunction with bilirubin casts may arise, sometimes requiring temporary dialysis — self-limited, resolves when bilirubin falls
Outcome	Prolonged, but self-limited. Some patients with prolonged jaundice and disabling symptoms were even referred for liver transplant, but almost all improved spontaneously and transplant was avoided. With long follow-up, complete resolution is expected
Causality classification	LiverTox score B: probable cause of clinically apparent liver injury, with jaundice

The dose in the cases is not the dose in the trials

LiverTox publishes a table that rarely appears anywhere else: the dose used in clinical trials next to the dose used by the people who had liver injury.

Generic name	Also called	Range in clinical trials	Range in liver injury cases
Ligandrol	LGD-4033, VK-5211	0.1 to 2.0 mg	4 to 30 mg
Enobosarm	MK-2866, S-22, GTx-024, Ostarine	0.1 to 3.0 mg	5 to 20 mg
Vosilasarm	RAD-140, Testolone	50 to 150 mg	5 to 30 mg
Andarine	GTx-007, S-4	not reported	25 to 50 mg

What this table shows, and what I will not conclude from it

With Ligandrol and Enobosarm, those who got hurt were using far more than what was tested. Ligandrol was studied between 0.1 and 2.0 mg; in the liver injury cases the range was 4 to 30 mg — the ceiling is fifteen times the maximum trial dose.
With Vosilasarm the relationship appears inverted, with a trial range larger than that of the cases — and the explanation is the opposite of reassuring. For the other SARMs there is a low-dose trial to serve as comparison. For RAD-140 there is none: the only clinical dose the compound has is an oncology dose. It comes from the only human trial ever done with it, a phase 1 first-in-human in 22 postmenopausal women with metastatic breast cancer, heavily pretreated, at 50 mg (6 people), 100 mg (13) and 150 mg (3) per day.
And the liver signal appeared right there. In that trial, under supervision, there was elevation of AST in 59.1% of participants, of ALT in 45.5% and of total bilirubin in 27.3%; grade 3 or 4 events in 16 of the 22 (72.7%). It is not a finding exclusive to the grey market: it is what was seen in the compound's first documented contact with the human body. The SARMs page opens that trial in detail.
None of this licenses the reading that a safe dose exists. The table shows what the people in the reports took, not a threshold below which there is no injury.

Two findings that change management

The first is in the detailed clinical case the entry itself presents, with the week-by-week table of tests. The man stopped the supplement when he became symptomatic. Four days later: bilirubin 7.9 mg/dL and ALT 177 U/L. In the following weeks, with the supplement already discontinued, ALT fell to the normal range — 43, 47, 48 U/L, with a reference limit below 50 — while bilirubin rose to 27.9, then 29.3, and peaked at 41.5 mg/dL.

Anyone following only AST and ALT would have seen tests normalizing in the exact period when the person was getting worse. In a cholestatic injury, the marker that counts is bilirubin — and there is no point waiting for alkaline phosphatase, which LiverTox itself describes as minimally elevated or normal at the start. That case had a latency of five weeks and took four months for the tests to return to normal.

The second is the 2 to 3 month latency, which dismantles the logic of the cycle. Someone who does eight weeks and stops can become jaundiced after stopping — that is what happened in this case — and not make the connection. And someone who reaches the end of the first cycle with no symptoms at all has not received proof of safety: they have received a time interval that has not yet ended.

What LiverTox says about the regulation of this

No SARM is approved for human use by the FDA. The entry lists at least 16 different agents, many with several chemical and trade names.
The FDA has published several clinical alerts about the dangers of SARM use and the lack of proven safety and efficacy, and has acted against black-market producers — but the products remain available.
The label usually says the product is for research only. LiverTox explains the function of that phrase: it is what lets the manufacturer claim the sale does not violate the regulatory approval requirement. It is not a description of the product, it is a maneuver.
What is written on the label may not be reliable, according to the entry itself. And SARMs are banned by the World Anti-Doping Agency, having been detected in competitive athletes, who were suspended.

What to do, in order

Baseline tests before, not after. AST, ALT, alkaline phosphatase, GGT, total and fractionated bilirubin, albumin and INR. Without a prior value, an abnormal test mid-use has nothing to be compared against.
Draw blood with at least one week without heavy strength training. It is the only way for the number to measure liver instead of muscle — the study shows the change persisting for seven days after one session.
Order CK along with it. That is what discriminates: high CK with normal bilirubin, GGT and alkaline phosphatase points to muscle.
Do not monitor transaminases alone. In the cholestatic pattern — the one from SARMs and anabolic steroids — what signals worsening is bilirubin, and it can rise while ALT falls. Alkaline phosphatase starts minimally elevated or normal, so waiting for it delays the diagnosis.
Dark urine, pale stools, itching without skin lesions, yellowed eyes: stop and seek care the same day. With that order of symptoms, the test comes after the visit, not before.
Bring the packaging and the label. In the DILIN and LiverTox series, much of the diagnostic difficulty came from nobody knowing what the person had taken. The product name and the lot are worth more than a description from memory.

What this page is not

It is not a Child-Pugh adjustment table. For the compounds on this site that have a package insert, the published answer is transcribed above and for most it is "no adjustment"; for those without one, there is no number anywhere — as on the renal page.
It is not a list of this site's hepatotoxic compounds. What exists is what is documented: SARMs, bodybuilding supplements and the endothelin antagonist class. Absence of reports for the others is absence of reports, not a certificate.
Does not cover viral hepatitis or drug interactions — which are, in the real world, a far more frequent cause of abnormal liver tests than anything catalogued here. Alcohol and steatosis, which were on this same list, now have their own page: alcohol and steatosis.
Does not replace the Brazilian package insert. The text above is the American label.

References

This page did not come from the secondary source. The text of each package insert was extracted by me from the openFDA label endpoint (api.fda.gov/drug/label.json), fields use_in_specific_populations and boxed_warning, on September 4, 2026, and is transcribed above in literal translation. The supplement liver injury data come from PubMed and LiverTox, a database of the National Institute of Diabetes and Digestive and Kidney Diseases, with DOI and identifier.

← Back to all compounds