$30–45
Monthly cost, compounded, cash-pay
vs $6,900+/mo for a biologic treating similar symptoms
350+
Human RCT subjects, 3 conditions
Fibromyalgia, Crohn's, MS — real controlled trials exist, not "no evidence"
1/10–1/33
Of the FDA-approved dose
LDN is 1.5–4.5mg; the approved tablet is 50mg
0
FDA-approved LDN products
Must be specially compounded — no commercial low-dose tablet exists
Two Doses, Two Very Different Stories
Naltrexone is one molecule prescribed at wildly different doses for wildly different reasons. Keeping the doses straight is the whole ballgame.
#1 — THE APPROVED DRUG
50mg · FDA 1984 / 1994
Full-dose oral naltrexone (ReVia, generic) completely blocks opioid receptors and is FDA-approved for opioid and alcohol use disorder. An extended-release injectable version (Vivitrol) followed in 2010. Nobody disputes this use — it's unremarkable, mainstream addiction medicine.
#2 — THE OFF-LABEL DOSE
1.5–4.5mg · used since the 1980s
At roughly a tenth to a thirty-third of the approved dose, LDN is prescribed off-label for fibromyalgia, Crohn's disease, multiple sclerosis, and chronic pain, on a different and only partly-understood mechanism (brief receptor blockade appears to trigger a rebound in the body's own endorphins and calm inflammatory glial cells). Real randomized trials exist — some positive, some negative. Both are shown below.
#3 — THE EMERGING FRONTIER
Cancer adjunct · case-report stage
A small number of physicians have published peer-reviewed case series reporting striking outcomes in advanced-cancer patients using LDN alongside other therapies. Mechanistically plausible and genuinely early. One completed placebo-controlled RCT in high-grade glioma (Duke 2022, n=110, PMID 35001215) tested quality of life and fatigue — not tumor response or survival — and found no significant QoL benefit. No completed Phase 3 efficacy RCT for cancer survival or tumor response was found in PubMed or ClinicalTrials.gov. Full cancer-focused write-up: LDN in Cancer.
Where the Controlled Evidence Actually StandsPubMed
Four conditions, four different pictures. None of them is "zero trials" — and none of them is "proven." Read the actual mix.
What the Controlled Trials Actually FoundPubMed
Fibromyalgia pain, 2013 (positive)
28.8% vs 18.0%
Younger et al., 4.5mg, n=31 women, randomized double-blind crossover: pain reduction on LDN vs placebo (P=0.016). 32% met responder criteria on LDN vs 11% on placebo (P=0.05).
Fibromyalgia pain, 2023 (negative)
No significant difference
The Danish FINAL trial, 6mg once daily, n=99 women, 12 weeks: did not beat placebo on the primary pain endpoint. A secondary signal on memory/cognition was noted for further study.
Crohn's remission, 2011
30% vs 18%
Smith et al., 4.5mg, n=40 adults, 12-week RCT: clinical remission rate on LDN vs placebo, plus improved endoscopic and histologic mucosal-healing scores — the trial's most-cited finding.
MS quality of life, 2010 (positive)
Improved QoL
Cree et al., 4.5mg, n=80, single-center double-masked crossover: significant improvement in mental health, pain, and self-reported cognitive function domains.
Every Completed Human Trial We FoundPubMedInternational
| Study | Condition | Type | n | Outcome |
|---|---|---|---|---|
| Younger 2013 (Arthritis & Rheumatism) | Fibromyalgia | RCT, crossover | 31 | Positive — pain, mood, satisfaction |
| Due Bruun-Plesner 2023/24 (Lancet Rheumatology) — Denmark | Fibromyalgia | RCT, parallel | 99 | Negative — primary pain endpoint |
| Nazir et al. 2025 meta-analysis (5 pooled RCTs) | Fibromyalgia | Meta-analysis | ~275 | Positive on pooled pain (SMD −0.61, P=0.02), moderate–high risk of bias |
| Smith 2011 (Digestive Diseases & Sciences) | Crohn's, adult | RCT | 40 | Positive — remission & mucosal healing |
| Smith 2013 pediatric pilot | Crohn's, pediatric | Pilot RCT | 14–17 | Safety/tolerability only — not powered for efficacy |
| Parker/Cochrane 2018 review | Crohn's, all ages | Cochrane systematic review | 46 (2 RCTs pooled) | "Insufficient evidence," GRADE Low |
| Cree 2010 (Annals of Neurology) — US | Multiple Sclerosis | RCT, crossover | 80 | Positive — quality of life |
| Sharafaddinzadeh 2010 — Iran | Multiple Sclerosis | RCT, parallel | 96 | Negative on primary QoL outcome — international, underpowered per authors |
| Berkson et al. 2009 (case series) | Pancreatic & other advanced cancer | Published case reports | 3 new (7 total across two papers) | Striking individual outcomes, uncontrolled, not a trial |
Reality check. Across fibromyalgia, Crohn's, and MS, at least 350 people have been through completed, published, placebo-controlled or Cochrane-reviewed randomized trials of LDN — this is a real, if thin, evidence base, not "nothing." But it is genuinely split: one positive and one negative trial in both fibromyalgia and MS, and a Cochrane-graded-Low verdict in Crohn's. For cancer, one completed placebo-controlled RCT exists (Duke high-grade glioma, n=110, PMID 35001215) — it measured quality of life and fatigue, not tumor response or survival, and found no significant QoL benefit vs placebo. No completed Phase 3 efficacy RCT for cancer survival or tumor response was found in PubMed or ClinicalTrials.gov. That is a funding/design gap, not proof LDN was tested for cancer outcomes and failed. See the dedicated page: LDN in Cancer.
Follow the Money — and the Missing Persecution StoryFollow the money
This is a different shape of story than a suppressed cheap drug like ivermectin. Read it precisely — overstating it would be as dishonest as ignoring it.
The plain drug is unpatentable
1984
Naltrexone's original patents expired decades ago. Anyone can manufacture plain generic tablets — which is exactly why no company has an incentive to fund a $50-100M trial testing a specific low dose of it.
Same molecule, patented reformulation, gets funded
$1,738 vs $35
Vivitrol (patented extended-release injectable naltrexone) and Contrave (patented naltrexone+bupropion combo, exclusivity through 2030) both had fully funded Phase 3 programs and FDA approval — because there was intellectual property to protect. Plain low-dose oral naltrexone has none.
Corporations are trying to enclose it anyway
2016–present
At least three companies (Immune Therapeutics, LDN Pharma Limited, Statera Biopharma) have filed method-of-use patents on low-dose naltrexone for autoimmune/inflammatory disease and signed "strategic rights agreements" since 2016. None has completed a registrational Phase 3 trial for any LDN indication as of this writing.
The real gatekeeper is the insurer, not the FDA
Investigational
Most US commercial insurers and Medicare Part D still classify compounded LDN as "investigational" for fibromyalgia, Crohn's, MS, and pain — so patients pay the $30-45/month in cash. That denial, not an FDA crackdown, is the actual daily barrier.
State this precisely. Unlike ivermectin, we found no FDA warning letters, no medical-board disciplinary actions, and no platform takedowns targeting doctors or advocates for their views on LDN. This is not a censorship story. The one enforcement action we located — an FDA untitled letter to Healthy Choice Compounding Pharmacy (Elmsford, NY, 2022) — cited a naltrexone 4.5mg capsule whose measured potency differed from its label, a manufacturing-quality problem at one compounding pharmacy, not a finding about whether LDN works. (We could not retrieve the FDA's exact verbatim wording live for this page — the letter is described here from corroborating secondary summaries; verify directly at the source before quoting it elsewhere.) The honest story is a funding-incentive gap: a 60-year-old off-patent molecule at an unpatentable low dose has no one positioned to pay for the trial that would settle any of the open questions above — not a scientific verdict, and not a persecution.
Cost — Why the Money Story Looks Like ThisMarket data
Honest note: LDN is not uniquely cheap compared to other generic pain drugs like pregabalin or duloxetine — those are off-patent too. The real cost contrast that matters is against the biologics used for the same category of autoimmune/inflammatory symptoms, where a single patented product runs 150–300× the price of a month of LDN.
Side Effects & Who Should Be CarefulPubMedFDA label
LDN's side-effect profile is genuinely mild at these doses. One interaction is not mild at all — read it before anything else here.
Common side effects
Vivid dreams
The best-documented LDN side effect: vivid dreams/nightmares (RR 2.41 vs placebo in the 2025 meta-analysis), plus transient insomnia, headache, and nausea in the first 1-2 weeks — usually mild and often fades. That's why it's typically dosed at bedtime.
⛔ The one thing that matters
Blocks opioids
Naltrexone, even at 1.5-4.5mg, is a full opioid-receptor antagonist. Anyone taking opioid pain medication or on opioid-agonist therapy (methadone, buprenorphine) can be thrown into acute, severe precipitated withdrawal — and it will block any opioid painkiller given afterward, including in an ER or during surgery. Do not start LDN while on opioids, and tell every treating clinician you take it before any planned procedure. This is not medical advice — talk to your doctor before starting or stopping.
The liver question — precisely
Removed 2013
The old hepatotoxicity boxed warning on full-dose naltrexone was tied to early studies using doses up to 300mg/day (6× the approved 50mg dose). The FDA removed that boxed warning from the label in July 2013. At LDN's 1.5-4.5mg — a tenth to a thirty-third of even the approved dose — there is no meaningful hepatotoxicity signal in the published trials. It remains contraindicated in acute hepatitis or liver failure; discuss existing liver disease with your prescriber.
Who should be careful or avoid it: anyone currently using opioid pain medication or opioid-agonist maintenance therapy (absolute — requires a medically supervised taper first); acute hepatitis or liver failure; anyone with planned surgery requiring opioid-based pain control (stop LDN in advance under your prescriber's guidance, typically about 72 hours, and confirm with them directly); pregnancy (limited formal safety data, though the largest reported clinical case series — a fertility specialist's cohort — is reassuring specifically for that population; discuss individually with your doctor).
PubMed vs. the Doctors Who Actually Prescribed ItNamed clinicians
Weigh these against the trial table above. Every credential here was independently checked before printing.
Dr. Bernard Bihari MD, Harvard 1957 · board certified, American Board of Psychiatry & Neurology, 1970 · former NYC Addiction Services Commissioner
The pioneer. In the mid-1980s, treating HIV/AIDS patients in his NYC internal-medicine practice, he observed improved immune markers and fewer opportunistic infections in patients given naltrexone at 1.5-4.5mg, and became the central figure who spread the practice to other doctors. Real limitation, stated plainly: Bihari never published his full HIV patient series in an independent peer-reviewed journal — his findings survive as conference abstracts, interviews, and secondhand advocacy-organization accounts, not an audited trial. He has zero remaining financial interest (he died in 2010) — no profit motive shaped his account. → Archival interview, Honest Medicine (2011)
Dr. Jill P. Smith MD · Professor of Medicine, Gastroenterology, Penn State College of Medicine
Both the researcher and the treating clinician — she ran the 2011 adult Crohn's RCT (n=40) herself and continues to prescribe it to her IBD patients. On record that the endoscopic mucosal-healing improvement was a real, biologically meaningful signal even though the trial was small and single-center. Pairs directly against the Cochrane review's "insufficient evidence" verdict on the same data. → Smith 2011, PMID 21380937
Dr. Burton Berkson MD (Autonomous University-affiliated program, Mexico), PhD Biological Sciences (University of Illinois) · postgraduate Internal Medicine/Pathology, Case Western-affiliated hospitals · Integrative Medical Center of New Mexico
Published peer-reviewed case series combining oral LDN (4.5mg) with IV alpha-lipoic acid in patients with advanced pancreatic cancer who had refused or were ineligible for conventional treatment. In his 2009 report of 3 new cases: one patient with liver metastases survived 39 months on the protocol; two others showed no evidence of disease on PET scan after 4-5 months of treatment. Limitation stated with the finding: uncontrolled case reports in single digits of patients, no control arm, and LDN's specific contribution cannot be separated from the alpha-lipoic acid, diet, and other elements of his protocol — the authors themselves concluded only that the protocol "warrants clinical trial," not that it was proven. → Berkson 2009, PMID 20042414
Dr. Phil Boyle MB BCh NUI (National University of Ireland, Galway, 1992) · MICGP, MRCGP · founder, NeoFertility clinic, Dublin
Has prescribed LDN since 2002 for fertility patients with recurrent miscarriage and failed IVF, reporting over 2,000 pregnancies carried to 37 weeks while on it, among roughly 4,000 couples helped overall. Financial interest, disclosed: LDN prescribing is part of the paid Restorative Reproductive Medicine treatment packages at his own commercial clinic — weigh his uncontrolled, self-reported outcomes accordingly. → Interview, 2014
Jarred Younger, PhD PhD, Experimental Psychology (not a physician) · Associate Professor, University of Alabama at Birmingham · NIH/DoD-funded
Ran the positive 2013 fibromyalgia crossover trial and studies the neuroinflammation/microglia mechanism LDN is thought to work through. Credential stated honestly: a PhD researcher, not an MD — his argument stands on the trial data and mechanism work, not on a medical license. → Interview, LDN Research Trust (2020)
Dr. Sean Mackey MD, PhD · Chief, Stanford Division of Pain Medicine · co-author, Younger et al. 2013
The measured, institutional voice — he helped run the original positive trial and gives the balanced insider's read: a real early signal, small trials, a genuinely mixed replication record (the 2023 Danish trial was negative), and an honest "we don't yet know who benefits," while noting the favorable risk/cost profile makes an individual trial reasonable to discuss with a doctor. → "Hype, Hope, or Homerun?", Peter Attia MD channel (2025)
Real-world use (anecdotal, labeled as such): beyond the four conditions with completed RCTs, patient communities (LDN Research Trust, patient forums) report broad off-label use for chronic fatigue/ME, Hashimoto's thyroiditis, endometriosis, PTSD, and mood symptoms. This is lived experience, not trial data — given fair weight, never blended with the controlled evidence above.
The Cancer Frontier — Real Mechanism, Case-Report StagePreclinical / case reports
A genuine, early avenue — not a cure claim, and not dismissible either.
Plausible mechanism
OGF–OGFr axis
Naltrexone increases circulating levels of the endogenous opioid met-enkephalin (opioid growth factor, OGF), which binds the OGF receptor and inhibits proliferation in several tumor cell-line and animal models — decades of mechanistic work, mostly out of Penn State (Zagon & McLaughlin).
Why it's not in clinics yet
Unfunded
Same profit gap as the rest of this page: no company owns the low-dose molecule, so no one is paying for the Phase 2/3 oncology trials that would test it properly. Emerging, not disproven — the trials simply don't exist yet, in any registry we checked.
The honest read
Promising, early
Real mechanism, real published case reports of striking individual outcomes, growing clinician interest — and a real gap where the controlled human trial should be. Worth watching, not worth treating as settled in either direction.
🧮 Build Your Dose — Home Liquid-Titration Calculator
Most LDN patients get precisely-measured capsules from a compounding pharmacy — that's the preferred, most accurate route where it's accessible and affordable. When a doctor has prescribed LDN but only standard-strength (e.g. 50mg) tablets are available, or a compounding pharmacy isn't accessible, a documented harm-reduction method exists: dissolve a standard tablet in water and measure the dose with an oral syringe. The math below is yours to plug in — never assume your tablet or target dose match anyone else's.
Enter what you actually have
This is not medical advice. Confirm your prescribed target dose and titration schedule with your own doctor or pharmacist before using this method.
1. Your tablet strength
25 mg
50 mg
mg
2. Water to dissolve it in
50 mL (gives 1mg/mL if 50mg tablet)
100 mL
mL
3. Your target dose (bedtime, as prescribed)
1.5 mg
3.0 mg
4.5 mg
mg
Method: drop the whole tablet into distilled water in a clean, amber (light-blocking) glass jar. It won't fully dissolve — naltrexone tablets contain insoluble binder, so you'll get a cloudy suspension, not a clear solution.
Every dose: shake well before drawing — the active drug settles unevenly in a suspension. Measure with a calibrated oral syringe or graduated dropper, never a kitchen spoon.
Storage: refrigerate, keep from light, and don't keep a batch longer than about 2 months — make a fresh one after that.
Typical Community & Trial ProtocolsClinic / trial protocol
Standard titration start1.5 mgWeeks 1–2, most common starting dose across published clinic and trial protocols, taken at bedtime
Typical target dose4.5 mgWhere Younger, Smith, and Cree's trials converge; bedtime dosing due to the vivid-dream side effect
Danish FINAL-trial dose6 mgHigher than the typical range — found no fibromyalgia pain benefit at this dose; the dose-response question is unsettled
Critical caveatNo approved labelEvery number above comes from a specific trial or clinic protocol — not an FDA-set dose for any condition. Compounding pharmacies vary in exact strength and formulation; confirm yours.
Regulatory PositionT1
FDA
Approved for addiction only
Approved standard-dose oral naltrexone (1984) and extended-release injectable Vivitrol (2010) for opioid and alcohol use disorder. Has never approved any low-dose (1.5-4.5mg) product or any LDN indication. LDN exists only via general 503A pharmacy-compounding rules, not a drug-specific approval pathway.
EMA / International
Same pattern abroad
European regulators have approved standard-dose naltrexone for addiction; no EU-wide approval exists for any LDN indication. Off-label prescribing is tacitly permitted in several health systems, mirroring the US pattern.
Insurance / CMS
The real gatekeeper
Most US commercial insurers and Medicare Part D classify compounded LDN as "investigational" for fibromyalgia, Crohn's, MS, and chronic pain — so cash-pay is the default reality for patients. This is the actual daily barrier, more than any FDA hostility.
The Bottom Line — In Plain English
What it is: A 1984-approved generic opioid-blocker, prescribed off-label since the 1980s at a tenth to a thirty-third of its approved dose for fibromyalgia, Crohn's, MS, and chronic pain, on the theory that brief receptor blockade triggers a rebound in the body's own endorphins and calms inflammatory glial cells.
Who's using it: Patients whose doctors prescribe it off-label through a compounding pharmacy, guided by a body of physicians going back to Bernard Bihari's original 1980s HIV/AIDS observations — through Jill Smith's Crohn's research, Burton Berkson's cancer case series, and Phil Boyle's fertility practice — set against the mixed, genuinely split randomized-trial record.
What the law and insurance say: Legal and FDA-approved for addiction at 50mg. Off-label and compounding-only at 1.5-4.5mg — no crackdown, no censorship, just no approved product and no insurance coverage for most off-label uses.
What the research shows: A real, thin, genuinely mixed evidence base — one positive and one negative RCT each in fibromyalgia and MS, a Cochrane "insufficient evidence" verdict in Crohn's built on only 46 people, plus one glioma QoL RCT with a null result on its QoL endpoint and no completed Phase 3 efficacy RCT for cancer outcomes. Not "no evidence." Not "proven." Split, and worth reading both halves — including the cancer-specific page.
- The dose matters enormously — 1.5-4.5mg is a completely different clinical story than the approved 50mg.
- Real controlled trials exist on both sides for fibromyalgia and MS; take the negative ones as seriously as the positive ones.
- The Crohn's and cancer signal is real but thin — small numbers, real limitations, honestly stated by the researchers themselves.
- The side-effect profile is genuinely mild — except for the opioid-blocking interaction, which is absolute and requires medical coordination.
- Nobody profits from testing an unpatentable low dose of an old generic — that's a funding gap, not a scientific verdict and not a suppression campaign.
- This is not medical advice. Talk to your doctor before starting, changing, or stopping LDN — especially if you take any opioid medication.