Bryan Johnson Longevity Protocol: How a Regular Patient Gets Access

The short answer: Blueprint is not a fixed drug stack

Blueprint is Bryan Johnson's personal, frequently revised health experiment. His public materials combine nutrition, exercise, sleep, measurements, prescription medicines, supplements, and procedures. That makes an old snapshot easy to mistake for his current regimen.

The most important correction is rapamycin. Johnson wrote that he stopped it in September 2024 after observing higher cholesterol, blood sugar, and resting heart rate, along with intermittent infections. His current protocol, dated January 23, 2026, does not include rapamycin in the prescription list. It also does not list testosterone or DHEA. Pages that still describe those three as his current core medicines are outdated.

Johnson's current page lists:

• Armour Thyroid and levothyroxine, which he attributes to hypothyroidism diagnosed at age 21
• oral minoxidil
• metformin in six-week on/off cycles
• tadalafil
• empagliflozin (Jardiance)
• acarbose
• evolocumab (Repatha)

The page itself gives inconsistent acarbose totals in two sections, so this article does not present one of them as a settled current dose. That is a useful reminder: a first-person protocol page documents what one person reports, but it is not a prescribing standard.

What the measurements can and cannot prove

Johnson reports unusually favorable values for body composition, blood pressure, glucose control, sleep, fitness, and biological-age measures. These are his team's reported results, not outcomes from a controlled trial. A single-person experiment cannot separate the effects of diet, training, weight change, sleep, medicines, supplements, testing, and chance.

One tool Johnson discusses, DunedinPACE, is a real research measure. Its validation paper describes a DNA-methylation biomarker designed to estimate the pace of aging and tested it across multiple cohorts. That validates the measurement approach; it does not independently validate Johnson's regimen or prove that a lower score predicts a longer life for one individual. DunedinPACE validation study [1]

The same distinction applies to the 12 hallmarks of aging. They are a useful framework for organizing aging biology, not a clinical checklist showing that every drug aimed at AMPK or mTOR improves human longevity. Hallmarks of aging: An expanding universe [2]

Evidence check on the most searched Blueprint drugs

Metformin

Metformin is established treatment for type 2 diabetes. Its use solely to slow aging is off-label, meaning FDA has not determined that the drug is safe and effective for that specific use. The central longevity question remains unanswered. FDA explanation of off-label prescribing [11]

The proposed TAME program was designed to test whether metformin could delay a composite of age-related diseases in roughly 3,000 adults ages 65 to 79. The published TAME Biomarkers Workgroup paper describes a framework for a future geroscience trial; it is not a completed efficacy trial. AFAR's current TAME page likewise describes the planned program and should not be rewritten as proof that thousands of participants have completed treatment. TAME Biomarkers Workgroup [3]

Human exercise studies also make the story less one-sided. In adults 65 and older, the MASTERS randomized trial found that metformin blunted some gains in lean mass and thigh muscle during progressive resistance training. Another randomized study found that metformin attenuated improvements in insulin sensitivity and aerobic capacity during exercise training in older adults. These findings do not make metformin broadly harmful; they show why a drug with a valid metabolic indication should not automatically be treated as a universal healthy-aging add-on. MASTERS trial [4] Aerobic-training trial [5]

For condition-specific information, see metformin off-label uses and adult monitoring.

Rapamycin

Rapamycin is the clearest example of the animal-to-human evidence gap. The National Institute on Aging's Interventions Testing Program has repeatedly found lifespan extension with rapamycin in genetically heterogeneous mice. The program is explicitly a mouse-testing program, not a source of human treatment recommendations. NIA Interventions Testing Program [6]

The PEARL study was a randomized, placebo-controlled trial of weekly rapamycin in healthy older adults. Its ClinicalTrials.gov record reports 129 actual participants and completion in December 2023, with no results posted in the record. That is meaningful human safety-and-feasibility research, but it cannot yet establish that rapamycin slows aging or extends life. PEARL study record [7]

Rapamycin remains FDA-approved for specific transplant-related uses, not longevity. Its current label includes clinically important risks and drug interactions, including infection, lipid abnormalities, and strong CYP3A4/P-gp inhibitors or inducers. DailyMed sirolimus label [12] Johnson's decision to stop it reinforces that a public figure's stack is not static and that individual biomarker responses can change the decision.

Acarbose

Acarbose is approved to improve glucose control in people with type 2 diabetes. The NIA mouse program found a larger median-lifespan effect in male than female mice. That finding is scientifically interesting but does not show that acarbose extends lifespan in healthy men or women. A patient with diabetes may have a conventional reason to discuss acarbose; a healthy person seeking longevity does not inherit an indication from a mouse study or from Johnson's personal use.

Testosterone

Testosterone is not on Johnson's January 2026 public prescription list. More broadly, testosterone treatment is evidence-based for appropriately evaluated hypogonadism, not as a general anti-aging intervention.

The TRAVERSE trial enrolled 5,246 men with symptoms, two fasting testosterone measurements below 300 ng/dL, and existing or elevated cardiovascular risk. Testosterone was noninferior to placebo for major cardiovascular events over the study period, but atrial fibrillation, acute kidney injury, and pulmonary embolism occurred more often in the testosterone group. Those results apply to the trial population; they are not evidence that men with normal testosterone benefit from treatment. TRAVERSE trial [8]

See testosterone for hypogonadism for the diagnostic pathway rather than treating a celebrity protocol as the indication.

NMN and other supplements

The original version of this page cited the wrong PubMed record and overstated an NMN result. The correct small randomized trial enrolled 30 healthy adults: 15 received 250 mg/day of NMN and 15 received placebo for 12 weeks. Blood NAD+ increased, and the study reported no obvious adverse effects during that short period. It did not demonstrate slower aging, disease prevention, or longer life. NMN randomized trial [9]

That pattern is common in supplement research: a compound may change a biomarker without proving a patient-centered outcome. Product quality, interactions, and the total burden of a large supplement stack also cannot be inferred from one small trial.

Omega-3 evidence is population-specific

REDUCE-IT is often summarized as generic evidence for high-dose fish oil, but the trial tested prescription icosapent ethyl in statin-treated patients who had elevated triglycerides and either established cardiovascular disease or diabetes plus risk factors. It reduced ischemic events in that defined high-risk population; it does not establish the same benefit for healthy adults taking an arbitrary over-the-counter EPA/DHA supplement. REDUCE-IT [10]

How a regular patient actually gets access

The practical route is not to ask for a "Blueprint stack." It is to identify the health problem being treated and choose an evidence-based intervention for that problem.

1. Start with the goal, not the celebrity medicine

A clinician needs to know whether the goal is diabetes prevention, lipid lowering, treatment of hypothyroidism, erectile dysfunction, hair loss, documented hypogonadism, or something else. The appropriate evaluation and follow-up are different for each. "Longevity" alone does not make metformin, acarbose, empagliflozin, thyroid hormone, testosterone, or rapamycin medically interchangeable.

2. Use measurements that can change a decision

Useful inputs often include blood pressure, weight trend, family history, current medicines and supplements, smoking status, sleep, exercise, and standard preventive-care history. Targeted labs then follow the clinical question: for example, HbA1c and kidney function for a metabolic evaluation, a lipid panel for cardiovascular risk, TSH and free T4 for suspected thyroid disease, or two morning testosterone measurements when symptoms suggest hypogonadism.

This is different from ordering every available biomarker. A test adds value when the result can alter a decision and has a credible interpretation.

3. Match access and monitoring to the medicine

• Metformin, acarbose, and Jardiance: normally prescribed for defined metabolic indications. Kidney function and the patient's other conditions affect whether they are appropriate.
• Repatha: used for specific lipid disorders and cardiovascular-risk situations, generally after a clinician reviews LDL cholesterol, prior therapy, and coverage criteria.
• Thyroid hormone: replaces deficient thyroid hormone; pushing levels beyond the appropriate range is not a proven longevity strategy.
• Testosterone: requires a diagnosis-based evaluation rather than a single screening value or an age-management target.
• Rapamycin: longevity use remains off-label and investigational, with no established human lifespan benefit.

Telehealth can be a way to complete a legitimate evaluation, but availability is not universal. The prescriber must be licensed for the patient's location, and the clinical, laboratory, pharmacy, and follow-up requirements depend on the drug and the patient's circumstances.

4. Build the foundation before adding experimental layers

The least speculative parts of Blueprint are also the least exclusive: regular exercise, adequate sleep, a nutrient-dense eating pattern, avoidance of tobacco, appropriate vaccination and screening, and treatment of established cardiovascular and metabolic risks. These interventions have outcome data that a 100-item personal stack does not.

A better way to use Blueprint

Blueprint is most useful as a case study in measurement and iteration. It is less useful as a shopping list.

Ask four questions about any claimed component:

1. Is the claim about a biomarker, symptoms, disease events, or lifespan? These are not equivalent outcomes.
2. Was the evidence from cells, mice, an observational study, or a randomized human trial? Evidence should not silently jump species or study designs.
3. Does the study population resemble the person considering treatment? REDUCE-IT and TRAVERSE answer questions about defined clinical populations, not healthy adults in general.
4. Is the item still in Johnson's current protocol? Rapamycin shows why date-stamping matters.

That approach preserves what is genuinely valuable about Johnson's project: transparent revision in response to data, without turning one person's experiment into a universal prescription.