Three months of an osteoporosis drug matched twelve on bone density in a 50-woman trial

Romosozumab is a monoclonal antibody that raises bone mineral density through a dual mechanism — it stimulates new bone formation while inhibiting resorption [s1]. It is also expensive, requires monthly injections, and carries cardiovascular concerns that limit its use [s1]. The standard course is twelve months.

The pharmacology contains a clue that the standard course may be longer than it needs to be: the bone-forming effect diminishes after several months [s1]. If most of the anabolic benefit arrives early, the later injections may be delivering little beyond cost and exposure.

The LIDA trial, published on January 29 in The Lancet Diabetes & Endocrinology, tested that directly [s1].

The design

Fifty postmenopausal women at high fracture risk were enrolled at a single US academic centre in a 12-month, open-label, non-inferiority randomised trial [s1].

One group received three months of romosozumab followed by nine months of denosumab. The other received twelve months of romosozumab [s1]. The primary endpoint was percentage change in total hip bone mineral density at 12 months, with a non-inferiority margin of 2% [s1].

The comparison is not "three months versus nine months of nothing." The abbreviated arm switched to an antiresorptive agent for the remaining nine months — a design choice that reflects standard practice, since romosozumab's gains are known to be lost if it is not followed by an antiresorptive. What is being tested is whether the anabolic phase can be shortened, not whether treatment can be.

The result

Mean 12-month change in total hip bone mineral density was 5.7% (SD 3.3) in the three-month romosozumab group and 6.0% (SD 3.2) in the twelve-month group, meeting the non-inferiority criteria [s1]. Adverse events were balanced between groups [s1].

The authors conclude the abbreviated regimen was non-inferior and could potentially expand access to the therapy [s1].

What fifty participants can establish

Very little about safety, and nothing about fractures.

A 50-person trial [s1] is adequately powered only for a densitometric endpoint with low variance, which is precisely what it used. Bone mineral density is a surrogate. It correlates with fracture risk but imperfectly, and the history of osteoporosis therapeutics includes agents that raised density without a proportionate reduction in fractures. This trial does not report fracture outcomes, and could not have been powered to.

"Adverse events were balanced between groups" [s1] in a 50-person trial is a statement with very wide implicit uncertainty. The cardiovascular concerns attached to romosozumab [s1] would require a study orders of magnitude larger to address — and if anything, the hypothesis motivating LIDA is that less exposure means less risk, which this trial cannot confirm.

Single-centre recruitment [s1] limits generalisability in the usual ways. Open-label design means neither participants nor investigators were blinded, though densitometry is a machine-read outcome and comparatively resistant to that.

Twelve months of follow-up [s1] also leaves the durability question open. Whether the two regimens diverge in year two or three — after the initial romosozumab effect and the subsequent denosumab maintenance have both played out — is not something a 12-month trial observes.

Why a small trial on an old question is interesting

Because it is a de-escalation trial, and those are rare.

Most drug trials ask whether a treatment beats a comparator or a placebo. Trials asking whether less of an approved drug performs as well as the licensed amount have no commercial sponsor with an obvious interest in the answer, and they are correspondingly uncommon relative to how often the question arises in practice.

The access argument the authors make is concrete [s1]: cost and monthly injection burden are among the stated limits on romosozumab use, and a regimen requiring three monthly injections instead of twelve changes both. That is a real potential benefit, contingent entirely on whether the finding survives replication in a larger sample with clinical endpoints.

An accompanying commentary in the same issue frames the shortened course as a pragmatic clinical solution [s1] — a characterisation that is appropriate to a hypothesis-generating result and should not be read as a change in standard of care.

What to watch

Whether a larger, multicentre trial with fracture endpoints is undertaken; whether the two regimens diverge beyond 12 months; and whether guideline bodies or payers respond to a single 50-person non-inferiority result, which would be premature.

This article describes a small randomised trial and is informational only. It is not medical advice and does not recommend any drug, dose, or duration of treatment. Decisions about osteoporosis therapy belong with a clinician.

Sources

• [s1] Leder BZ, Ramchand SK, Jordan M, Ryan S, Patnaik A, Lee H, Tsai JN, 3 months vs 12 months of romosozumab for postmenopausal osteoporosis (LIDA): an open-label, non-inferiority, randomised controlled trial, The Lancet Diabetes & Endocrinology, published online 2026-01-29, with linked commentary in the same journal.