READER RESOURCES: THE APOCALYPSE ALMANAC: Hidden cures in our dystopian age. Check out the “Cure Cancer in Your Kitchen” chapter. FULLSCRIPT SUPPLEMENTS: top quality and economical. AFFILIATE STORE: I offer competitively priced products HERE that I have personally tested and used.
Summary
· The old general practitioner’s 2 cc shot of cyanocobalamin contained 2,000 mcg. In one Kentucky practice, 10 percent of patients received regular injections.
· Deficiency is common after age 60, and in 28 percent of patients with neurologic damage, the blood count is normal. Giving folate first hides the anemia while the nerves keep dying.
· The strongest evidence for a pharmacologic effect is in ALS: 50 mg of methylcobalamin injected twice weekly slowed decline by 43 percent in early-stage disease, and Japan approved it in 2024.
· In Parkinson’s, low B12 predicts twice the rate of motor decline and a 5-fold risk of dementia, but no trial of high-dose treatment has been completed.
· Cyanocobalamin injection is an FDA-approved generic, sold in 30 mL multi-dose vials at about $1 per 1,000 mcg dose. Methylcobalamin injection is available in the United States only as a compounded product.
· Foreign ampoules are small because Japan licensed 500 mcg for neuropathy in the 1980s, and the rest of the world followed suit. Mexico sells 10,000 mcg.
· Oral B12 at 1,000 to 2,000 mcg a day corrects deficiency as effectively as injections. What the pill does not do is deliver the 50 mg pulses used for ALS.
The shot every old doctor gave
For half a century, the family doctor kept a 30 mL vial of red liquid in the refrigerator and drew up 2 cc for anyone who complained of being tired. The standard strength was 1,000 mcg per mL, so the shot was 2,000 mcg, about 800 times the daily requirement. Patients reported a burst of energy for days afterward and returned monthly for the next shot. In 1989, Lawhorne and Ringdahl reviewed 1,222 charts in a rural Kentucky family practice and found 120 patients, 10 percent of the practice, on regular cyanocobalamin injections. Four of them, 3 percent, met any accepted criterion for needing the vitamin. Of the 48 interviewed, the average duration of shots was 9.9 years, the effect was rated 2.9 on a 4-point scale, and 38 percent said that if the doctor stopped, they would find one who would not.
The controlled trials were less kind to the shot than the patients were. Ellis and Nasser ran a double-blind crossover in 1973 on tired patients with normal blood counts, giving 5 mg of hydroxocobalamin twice weekly for 2 weeks. General well-being and happiness scores improved on the vitamin; energy, sleep, and appetite did not. In 1978, Tin-May-Than and colleagues gave 31 subjects 1 mg three times a week for 6 weeks and measured oxygen uptake, grip, jumping, and tapping; nothing changed. Hvas and colleagues in Denmark took 140 people with elevated methylmalonic acid, the most sensitive marker of tissue deficiency, gave 1 mg weekly for a month against placebo, and found the metabolic markers normalized while quality of life, cognition, and mood barely moved. Whatever the old doctors were treating, the objective instruments did not detect it.
The shot did one thing that the studies confirm. It treated the deficient patient who had not been tested, and in the 1950s and 1960s that was a lot of people. The rest of this essay sorts the uses into those with evidence, those with a plausible mechanism, and those with a story.
Who is deficient and how it shows
Up to 15 percent of people over 65 are deficient, and up to 20 percent of older adults have atrophic gastritis, which prevents them from freeing the vitamin from food protein. Pernicious anemia, the autoimmune loss of intrinsic factor, is the classic cause. The list now includes gastric bypass, ileal disease, vegan diets, metformin, acid-suppressing drugs, and nitrous oxide, which oxidizes cobalt and inactivates the vitamin in a single exposure. Britain’s drug regulator warned in 2022 that up to 1 in 10 metformin users become deficient, and the American label followed with advice to check levels periodically.
The textbook picture is megaloblastic anemia with a smooth tongue. The dangerous picture is neurologic: numb feet, loss of vibration and position sense, an unsteady gait, spasticity, memory loss, depression, and psychosis.
Subacute combined degeneration is the destruction of the posterior and lateral columns of the spinal cord, and it is permanent if treatment is delayed. In 1988, Lindenbaum and colleagues reported 141 patients with neuropsychiatric deficiency in the New England Journal of Medicine. Forty of them, 28 percent, had a normal hematocrit and normal red cell size. Every one who was treated improved. A normal blood count rules out nothing, and the serum B12 level misses cases that methylmalonic acid and homocysteine detect.
The FDA label regimen for pernicious anemia is outdated: 100 mcg daily for a week, then every other day, then every 3 or 4 days, then 100 mcg monthly for life. Nobody uses it. The regimen in practice is 1,000 mcg daily for a week, then weekly for a month, then monthly. Britain uses hydroxocobalamin, which binds plasma proteins and stays in the body longer: 1 mg every other day until improvement stops when nerves are involved, then 1 mg every 2 months; 1 mg three times a week for 2 weeks, then every 2 or 3 months when they are not.
Folate and the trap
Folate deficiency produces the same megaloblastic anemia because folate and B12 act at the same step in DNA synthesis. Causes include alcohol, poor diet, malabsorption, pregnancy, hemolysis, dialysis, and drugs that block folate metabolism: methotrexate, phenytoin, sulfasalazine, and trimethoprim. Since 1998, American grain has been fortified with folic acid at 140 mcg per 100 grams, and neural tube defects have fallen from 10.7 to 7.0 per 10,000 births, about 1,300 fewer affected pregnancies per year. Treatment is folic acid 1 to 5 mg daily, or folinic acid when methotrexate is the cause.
The trap is the order of treatment. Folic acid at doses above 0.1 mg per day corrects the anemia of B12 deficiency, and the FDA label says so in plain words: hematologic remission. The neurologic damage proceeds. A doctor who sees macrocytosis and reaches for folate without checking B12 has erased the one sign that would have led him to the spinal cord. Give B12 first, or give both.
Methylfolate has a separate use. Papakostas and colleagues added L-methylfolate to antidepressants that had failed. At 7.5 mg it did nothing. At 15 mg a day, 32 percent responded against 15 percent on placebo, a number needed to treat of 6. The dose is 40 times the dietary requirement, which is the point: at that dose it is a drug.
Cyanocobalamin versus methylcobalamin, in brief
Alternative practitioners make much of the difference between the forms, but for almost everyone it may not matter. The four forms of the vitamin differ by a single group attached to the cobalt atom: cyanide, hydroxyl, methyl, or adenosyl. Inside the cell, a protein called MMACHC strips that group off whichever form arrives, and the cell builds its own coenzymes from the bare cobalamin. Obeid, Fedosov, and Nexo reviewed the biochemistry in 2015 and concluded that the coenzyme forms “are not likely to be superior.”
A 1,000 mcg shot of cyanocobalamin contains 20 mcg of cyanide, less than in a cigarette, and the liver converts it to thiocyanate. (Of course, if you are getting 20 to 30 times this via doses like those described in this article, it might be the equivalent of a pack of cigarettes.) The one human comparison, in 42 vegans, favored cyanocobalamin. The two exceptions are early Leber’s hereditary optic neuropathy and dialysis patients, who clear thiocyanate poorly; those patients should get hydroxocobalamin or methylcobalamin.
Three additional problems exist with methylcobalamin. First, no methylcobalamin injection is FDA-approved in the United States. Every vial comes from a compounding pharmacy, and the substance is on the FDA’s 503A bulk list in Category 1, meaning it is permitted while the agency evaluates it and has not been approved. A compounded product lacks manufacturer’s stability data, and methylcobalamin degrades in light; losses of 48 to 76 percent in multivitamin formulations have been measured, compared with 6 to 21 percent for cyanocobalamin.
Second, methylcobalamin costs 5 to 10 times more for the same microgram dose. Third, acne. Doses above 5 to 10 mg a week are associated with acneiform eruptions, more often in women than men, and Kang and colleagues showed in 2015 that B12 shifts the skin bacterium Propionibacterium acnes toward producing inflammatory porphyrins. Any form does this at high doses, but the high-dose products sold to the public are almost all methyl.
One fact cuts the other way, and it is the most important in this essay. The only indication with a positive randomized trial and regulatory approval for ALS is methylcobalamin. No one has tested cyanocobalamin at 50 mg, and no one will, because it has no patent. The methyl form’s claim to superiority in deficiency is marketing. Its claim in ALS is data.
Nerves and brain
The story in ALS began in Japan, where Eisai has sold Methycobal (methylcobalamin) for peripheral neuropathy since 1984. Kaji and colleagues randomized 373 patients to placebo, 25 mg, or 50 mg via intramuscular injection twice weekly for 3.5 years. The trial missed its endpoints, but patients enrolled within a year of symptom onset who received 50 mg lived or remained off the ventilator a median of 1,197 days, compared with 570 days on placebo.
Eisai ran a second trial in that population. JETALS, published in JAMA Neurology in 2022, gave 130 early patients 50 mg twice weekly or placebo for 16 weeks. The functional rating scale fell 2.66 points on the drug and 4.63 on placebo, a 43 percent slowing, with 90 percent of both groups on riluzole. Side effects were no different from placebo. Japan approved Rozebalamin (mecobalamin) 25 mg vials for ALS in September 2024. The dose is 50,000 mcg, 25 times the old doctor’s shot, administered twice a week.
Parkinson’s is a different situation: strong correlation, no trial. Christine and colleagues at UCSF went back to stored blood from 680 untreated early patients in the DATATOP study. Those in the lowest B12 tertile lost walking ability at 1.53 units per year, compared with 0.77 in the highest, twice the rate. Those with high homocysteine lost 1.96 points per year on the Mini-Mental State Examination, while the others held steady. Patients whose B12 rose during the study, usually from a multivitamin, gained 10 disability points over follow-up, compared with 14.4 for those whose levels stayed low. The same group confirmed the finding in spinal fluid in 2020. McCarter and colleagues at Mayo found that patients with B12 below 587 ng/L at diagnosis had 5 times the risk of dementia, and that each 100 ng/L increase cut the hazard by 69 percent.
Levodopa consumes methyl groups and raises homocysteine, so the mechanism is plausible. 55 percent of Parkinson’s patients have peripheral neuropathy, compared with 9 percent of matched controls. Yet the one observational look at supplement users, 1,741 patients in the PPMI cohort, found only a trend toward less neuropathy at 3 years, and an Emory trial of supplementation was terminated without results. No randomized trial of pharmacologic-dose B12 in Parkinson’s has been completed. The ALS result makes one worth running.
Cognition has a completed trial. VITACOG gave 168 people over 70 with mild cognitive impairment folic acid 0.8 mg, B12 0.5 mg, and B6 20 mg daily, or placebo, for 2 years, and measured brain shrinkage by MRI. Atrophy slowed by 30 percent overall and by 53 percent in those with homocysteine above 13. Douaud and colleagues then showed that the protected tissue was the gray matter that Alzheimer’s attacks first: in the high-homocysteine half, 5.2 percent was lost on placebo and 0.6 percent on the vitamins, a 7-fold difference. That was oral, at 500 mcg a day.
Peripheral nerves are the traditional target. Japanese and Indian physicians give methylcobalamin 500 mcg by mouth three times daily or 500 mcg by injection three times weekly for diabetic neuropathy. A 2021 trial found that 5 intravenous doses of 500 mcg, added to amitriptyline, cut pain scores by 63 percent, compared with 31 percent for amitriptyline alone. For Bell’s palsy, Jalaludin randomized 60 patients in 1995 to prednisone, methylcobalamin 500 mcg injected three times a week, or both. Recovery took 9.6 weeks on the steroid alone and about 2 weeks in the 2 groups given the vitamin. The trial was open-label and has never been repeated, which is the usual fate of a positive result for a generic. In multiple sclerosis, Kira and colleagues gave 6 patients with chronic progressive disease 60 mg by mouth daily for 6 months; visual and brainstem evoked potentials improved, but motor disability did not.
Fatigue, again
Chronic fatigue syndrome is the one fatigue diagnosis with serious defenders. Regland and colleagues in Sweden studied 38 women with myalgic encephalomyelitis who had been receiving B12 injections for 6 months to 20 years. Among the 15 good responders, injections averaged every 3.8 days; 93 percent used methylcobalamin, and they took 6.7 mg of folic acid daily alongside it, 4 times the dose of the mild responders. The mild responders were largely on opioids, duloxetine, or pregabalin. It is a case series, and it says what the old doctors said: frequent injections, high folate, and no sedating drugs. Nobody has funded the trial.
The doses in use
Deficiency replacement is 1,000 mcg by intramuscular injection daily for a week, then weekly for a month, then monthly, or 1 mg of hydroxocobalamin every 2 or 3 months under the British scheme. Japanese neuropathy dosing is 500 mcg three times a week by injection or 1,500 mcg a day by mouth. VITACOG used 500 mcg a day by mouth with folate and B6. In the Swedish series, chronic fatigue patients injected 1 to 5 mg every 3 to 6 days. ALS dosing is 50 mg twice a week by intramuscular injection, indefinitely. The old doctor’s 2,000 mcg monthly is a deficiency dose given to people who, for the most part, were not deficient, and the ALS dose is 100 times larger per injection and 8 times more frequent.
Why the foreign ampoules are so small
500 mcg three times a weekThe injectable B12 that arrives from India is a 1 mL ampule of 500 mcg methylcobalamin, and the reason is regulatory history. Eisai’s Methycobal, launched in Japan in 1984, was licensed for peripheral neuropathy at 500 mcg three times a week, a dose chosen for a nerve indication and never meant to replace stores. Indian generics copy the originator, so the world’s cheapest methylcobalamin comes in the Japanese dose.
Britain’s Neo-Cytamen (hydroxocobalamin) is 1,000 mcg per ampule because that is the British deficiency regimen. German pharmacies sell 1,000 mcg cyanocobalamin ampules without a prescription, 10 for about 6 euros. Mexico is the exception: Bedoyecta Tri and Tribedoce are each 10,000 mcg of hydroxocobalamin with 100 mg of thiamine and 50 mg of pyridoxine in a 2 mL ampule, sold in 5-packs for about $6 to $22, and the “50,000” on the Tribedoce box is a unit label rather than a microgram count. The small ampule is a licensing artifact, and the buyer who wants 1,000 mcg from a 500 mcg ampule uses 2.
The prices are low everywhere. Methycobal 500 mcg from an Indian pharmacy is about $1.30 an ampoule, and Indian cyanocobalamin 1,000 mcg ampoules list at 35 rupees, about 40 cents. Japan’s national health insurance pays 92 yen, about 65 cents, per ampoule.
What it costs in America and where it comes from
Cyanocobalamin injection USP is an FDA-approved generic, made by Hikma, American Regent, Somerset, Vitruvias, and others, and it has never been a compounding-pharmacy product. It comes in 1 mL single-dose vials and in 10 mL and 30 mL multi-dose vials preserved with 1.5 percent benzyl alcohol. GoodRx prices in 2026 are $12.92 for the 10 mL vial and $29.13 for the 30 mL vial, which is $0.97 per 1,000 mcg dose, against about $2 a dose in single vials. A doctor’s office or medical spa charges $20 to $75 for the same injection. A patient with a prescription and a 30 mL vial spends less on a year of monthly shots than on one visit.
Methylcobalamin injection is compounded only by 503A pharmacies such as Empower, Olympia, and Hallandale, available in strengths from 1 to 20 mg per mL. A 10 mL vial of 1 mg per mL costs $40 to $120. Telehealth companies sell subscriptions for the same product at $39 to $79 per month. Hydroxocobalamin injection has one American manufacturer, Actavis, and has been in shortage since 2024 due to a lack of raw material. Nascobal (cyanocobalamin) nasal spray, 500 mcg once a week, costs $174 per month with a coupon and $600 without, for a dose that a 5-cent tablet delivers.
Cautions
The label has 5 warnings. Anaphylactic shock and death have been reported after parenteral B12, though rarely, and cobalt-sensitive patients may react. Severe megaloblastic anemia treated intensively can cause hypokalemia as the marrow suddenly builds cells, and sudden death has followed; check potassium in the first week. Early Leber’s disease progressed to optic atrophy on cyanocobalamin. The vehicle contains aluminum, which accumulates in kidney failure with prolonged use. Folic acid above 0.1 mg a day masks the anemia and leaves the neurologic damage untreated. Treatment also unmasks polycythemia vera in the occasional patient whose anemia was hiding it.
Two large trials raise concerns about long-term high doses in specific groups. In the DIVINe trial, 238 diabetics with kidney disease received folic acid 2.5 mg, B6 25 mg, and B12 1 mg daily for 3 years; their kidney function declined faster than on placebo, 16.5 versus 10.7 mL/min, and vascular events were more frequent. In the VITAL cohort of 77,118 people, men who averaged more than 55 mcg of B12 a day from supplements for 10 years had twice the lung cancer risk, and male smokers had 3.7 times; women had no excess. Neither finding applies to a monthly shot, and both were oral. The homocysteine-lowering trials in heart disease, HOPE-2, NORVIT, and SEARCH, lowered homocysteine but prevented nothing, which is why homocysteine reduction alone is no reason to inject anyone.
A high B12 level in someone taking no supplement is a warning, not a comfort. Arendt and colleagues followed 333,667 Danes and found that a level above 800 pmol/L predicted 6 times the expected number of cancer diagnoses in the following year, especially liver, blood, and smoking-related tumors. The vitamin did not cause them; a sick liver dumps its stores. An unexplained high level is a reason to look for the cause.
What the pill does
About 1 percent of an oral dose crosses the gut by passive diffusion without intrinsic factor. This means 1,000 mcg a day delivers 10 mcg to a patient with pernicious anemia, 4 times his daily need. Kuzminski and colleagues proved it in 1998: 2,000 mcg a day by mouth beat 1,000 mcg by injection on every marker at 4 months, with serum B12 of 1,005 against 325 pg/mL and lower methylmalonic acid. The Cochrane review found the routes equivalent and the pill cheaper. Sublingual tablets absorb no better than swallowed ones. For deficiency, 1,000 to 2,000 mcg a day by mouth is the answer for anyone who will take it, and a 160-count bottle costs $20.
The pill cannot replicate the injected pharmacology. A 50 mg oral dose delivers 500 mcg, a tenth of what the ALS regimen delivers in a single shot, and no oral trial in ALS exists. The VITACOG brain result came from 500 mcg a day by mouth, so the cognitive indication is oral. The neuropathy and fatigue regimens are injected in practice because that is how they were studied, and the Swedish fatigue patients who switched to tablets relapsed. The division is this: deficiency and cognition, tablets; ALS, injection; everything else, the injection has the evidence, such as it is, and the tablet has never been tried.
How to buy the large dose is behind the paywall at the very bottom.
Synthesis
The old doctors were half right. They gave a deficiency dose to a population that was 5 to 15 percent deficient without anyone checking, and the ones who improved were the ones who needed it. The tired patient with normal blood values did not gain measurable energy in any controlled trial, and the burst he felt has never been caught by an instrument. The vitamin’s pharmacology begins at doses the old doctors never used: 50 mg twice a week in ALS, where it is the only intervention, besides riluzole, to slow the disease in a trial that met its endpoint, and 500 mcg a day with folate in the aging brain, where it cut atrophy by half in those with high homocysteine.
Parkinson’s has the correlation and lacks the trial, and I expect it to respond to the same regimen. Neuropathy, Bell’s palsy, and chronic fatigue have small positive studies that nobody has repeated because the drug costs a dollar. For deficiency, the pill works; for the pharmacologic uses, the injection is the only thing that has been tested. Buy the 30 mL vial of cyanocobalamin, give 1,000 mcg wherever 1,000 mcg is the dose, use hydroxocobalamin or methylcobalamin in the dialysis patient and the Leber’s patient, and use compounded methylcobalamin at 50 mg twice a week for ALS, where the trial proved it, and for Parkinson’s, where the biology says to try. Alzheimer’s is closely related and may benefit as well.
Selected references
Lawhorne L, Ringdahl D. Cyanocobalamin injections among primary care patients who do not have the diagnosis of vitamin B12 deficiency. JAMA, 1989.
Lindenbaum J, et al. Neuropsychiatric disorders caused by cobalamin deficiency in the absence of anemia or macrocytosis. New England Journal of Medicine, 1988.
Kuzminski AM, et al. Effective treatment of cobalamin deficiency with oral cobalamin. Blood, 1998.
Wang H, et al. Oral vitamin B12 versus intramuscular vitamin B12 for vitamin B12 deficiency. Cochrane Database of Systematic Reviews, 2018.
Obeid R, Fedosov SN, Nexo E. Cobalamin coenzyme forms are not likely to be superior to cyano- and hydroxyl-cobalamin in prevention or treatment of cobalamin deficiency. Molecular Nutrition and Food Research, 2015.
Kaji R, et al. Ultra-high-dose methylcobalamin in amyotrophic lateral sclerosis: a long-term phase II/III randomised controlled study. Journal of Neurology, Neurosurgery and Psychiatry, 2019.
Oki R, et al. Efficacy and safety of ultrahigh-dose methylcobalamin in early-stage amyotrophic lateral sclerosis: a randomized clinical trial. JAMA Neurology, 2022.
Eisai. Rozebalamin for injection 25 mg (mecobalamin) approved in Japan for amyotrophic lateral sclerosis. September 2024.
Christine CW, et al. Vitamin B12 and homocysteine levels predict different outcomes in early Parkinson’s disease. Movement Disorders, 2018.
McCarter SJ, et al. Higher vitamin B12 level at Parkinson’s disease diagnosis is associated with lower risk of future dementia. Parkinsonism and Related Disorders, 2020.
Smith AD, et al. Homocysteine-lowering by B vitamins slows the rate of accelerated brain atrophy in mild cognitive impairment. PLoS ONE, 2010.
Douaud G, et al. Preventing Alzheimer’s disease-related gray matter atrophy by B-vitamin treatment. PNAS, 2013.
Jalaludin MA. Methylcobalamin treatment of Bell’s palsy. Methods and Findings in Experimental and Clinical Pharmacology, 1995 (summarized in American Family Physician, 2008).
Regland B, et al. Response to vitamin B12 and folic acid in myalgic encephalomyelitis and fibromyalgia. PLoS ONE, 2015.
Papakostas GI, et al. L-methylfolate as adjunctive therapy for SSRI-resistant major depression. American Journal of Psychiatry, 2012.
House AA, et al. Effect of B-vitamin therapy on progression of diabetic nephropathy. JAMA, 2010.
Brasky TM, et al. Long-term supplemental one-carbon metabolism-related vitamin B use in relation to lung cancer risk in the VITAL cohort. Journal of Clinical Oncology, 2017.
Arendt JFB, et al. Elevated plasma vitamin B12 levels as a marker for cancer. Journal of the National Cancer Institute, 2013.
Kang D, et al. Vitamin B12 modulates the transcriptome of the skin microbiota in acne pathogenesis. Science Translational Medicine, 2015.
Cyanocobalamin injection, USP, prescribing information. FDA, 2021.
Cyanocobalamin prices. GoodRx, 2026.
Methycobal injection, Japanese prescribing information. KEGG Medicus.
Rozebalamin for injection 25 mg, Japanese prescribing information. KEGG Medicus.
Eisai. Rozebalamin for injection 25 mg (mecobalamin) launched in Japan. November 2024.
FDA personal importation policy, frequently asked questions. FDA.
Tribedoce solucion inyectable, prescribing information. PLM Mexico.
WeCare Pharmacy injectable price sheet. 2023.
Methylcobalamin and light. Compounders’ Lab, 2018.