Journal

Microdose Tirzepatide: The Evidence Gap

Do not infer efficacy from standard-dose trials

What a microdose actually is

Microdose programmes prescribe well below the doses studied in the pivotal trials — typically around 1 mg a week against the 5, 10 and 15 mg used in SURMOUNT-1. They are marketed on tolerability and on price, and they are genuinely the cheapest compounded option in the market, reaching $110 to $147 a month.

They are also, as a matter of regulatory mechanics rather than clinical innovation, one of the surviving routes for compounding a drug that has an approved equivalent. Compounding a copy of an available approved product is generally not permitted; compounding something the prescriber documents as clinically different for that patient is. "Personalised dosing" is the mechanism by which that documentation happens.

The dose-response problem

SURMOUNT-1 reported mean weight reductions rising with dose: about 15.0% at 5 mg, 19.5% at 10 mg and 20.9% at 15 mg on the treatment-regimen estimand, against 3.1% with placebo. That is a clean dose-response relationship, and it is the reason the microdose claim does not survive contact with the data.

If the effect increases from 5 to 10 to 15 mg, there is no basis for expecting a 1 mg dose to reproduce a 15 mg result. The trials did not test 1 mg. Nobody has published what it does.

What we do not know, specifically

• What mean weight reduction a 1 mg weekly dose produces over 72 weeks. No trial has measured it.
• Whether the tolerability advantage is real at that dose or simply assumed from the escalation phase.
• Whether a microdose maintains weight already lost, which is a different question from producing loss.
• What happens on stopping, given that SURMOUNT-4 studied withdrawal from full doses.

None of that means microdosing does nothing. It means the evidence is silent, and silence is being filled with figures borrowed from trials of a different dose of a different product.

When a lower dose is a legitimate clinical decision

Prescribers reduce doses for good reasons: side-effect burden, comorbidities, patient preference, or maintenance after a period at a full dose. A clinician choosing a lower dose for a specific patient is practising medicine.

That is different from a programme built around a low dose as its default commercial offer, priced as the entry-level tier, and advertised beside trial figures collected at three to fifteen times the dose.

Questions worth asking

1. What weekly dose does this programme actually deliver, in milligrams?
2. What evidence exists for that dose specifically — not for tirzepatide generally?
3. Is the plan to escalate, and does the price change if I do?
4. If this is maintenance dosing, what was I meant to have done first?

What "microdose" is not standardised to mean

There is no regulatory or clinical definition. Programmes describing themselves as microdose differ in what they actually deliver, and several do not publish a milligram figure at all. Enhance.MD states 1 mg weekly; others describe the concept without the number.

That alone is a reason to ask. A programme unwilling to tell you the weekly dose in milligrams cannot be compared with anything, including itself over time.

The maintenance argument, examined

The strongest case for low-dose therapy is maintenance: someone who reached a target on a full dose, then reduces to hold it. That is a recognised clinical pattern in other chronic conditions.

The evidence gap is that it has not been tested here. SURMOUNT-4 studied continuation at maximum tolerated dose against complete withdrawal. It did not study reduction to a low dose, which is the intermediate option the maintenance argument depends on.

So a provider offering microdosing as maintenance is extrapolating from a trial that tested something else. That may still be reasonable clinical judgement for an individual patient — but it is judgement, not evidence, and it should be described that way.

The regulatory reason the category exists

Compounding a copy of an available approved product is generally not permitted. Compounding something a prescriber documents as clinically different for an individual patient is. That documentation requirement is what "personalised dosing" satisfies.

This explains the timing better than any clinical development does: the microdose category expanded when the shortage listings ended, not when new evidence appeared.

The price picture

Microdose programmes run $110 to $349 a month in our dataset. The cheapest FDA-approved GLP-1 of any kind is $149, and the Medicare Bridge is $50 for eligible enrollees. Two microdose programmes in the dataset cost more than an approved product with published trial data at its studied dose.

Questions readers actually ask