Step-down Therapy After Long-term Osteoporosis Treatment

Effective maintenance of bone mineral density (BMD) after discontinuing denosumab (Prolia) remains a major clinical challenge. Although denosumab is highly effective in increasing BMD, patients frequently discontinue therapy due to issues such as adherence, comorbidities, or personal circumstances. This interruption can lead to rapid and profound bone loss, substantially increasing fracture risk. Current clinical consensus recommends follow-on bisphosphonate therapy-most commonly alendronate (Fosamax) or zoledronic acid (Aclasta)-after the pharmacologic effect of denosumab wears off. However, evidence indicates that in patients treated with denosumab for more than three years, conventional bisphosphonate transition alone is often insufficient to fully prevent post-discontinuation BMD decline. Importantly, most bone loss occurs within the first year after stopping denosumab, highlighting the urgent need for strategies that can suppress the rebound increase in bone turnover during this critical period.

This study aims to investigate whether a dose-reduction (de-escalation) strategy-administering half-dose denosumab (30 mg every six months) combined with sequential zoledronic acid therapy-can better preserve lumbar spine BMD after long-term denosumab treatment. The target population includes (1) postmenopausal women with osteoporosis or osteopenia-associated fractures who have received at least three years of denosumab treatment, and (2) men aged ≥50 years with osteoporosis or osteopenia-associated fractures and ≥3 years of denosumab therapy.

The study uses an open-label, stratified randomized trial design. After enrollment, participants are stratified according to total duration of prior denosumab use: <4 years or ≥4 years. Within each stratum, participants are randomized into two groups, with 22 participants per group.

Control Group (Standard Sequential Therapy):

Participants receive conventional post-denosumab management. A single infusion of zoledronic acid is administered at the expected end of denosumab's pharmacologic effect, followed by a second zoledronic acid infusion one year later (±1 week).

Intervention Group (De-escalation + Bisphosphonate Strategy):

At study entry, participants receive a half-dose denosumab injection (expected duration of effect: six months) concurrently with one dose of zoledronic acid. Six months later (±1 week), participants receive a second half-dose denosumab injection. At twelve months after enrollment (±1 week), a second zoledronic acid infusion is administered.

This investigational regimen is designed to gradually taper denosumab exposure while simultaneously initiating bisphosphonate therapy to stabilize bone turnover. By providing partial RANKL inhibition during the early high-risk period and reinforcing mineralization with zoledronic acid, the study aims to determine whether this combined approach can more effectively mitigate the rebound effect and prevent lumbar spine BMD loss.

The results of this trial will help clarify whether a structured de-escalation plus bisphosphonate strategy offers superior protection compared with standard transition therapy in patients with long-term denosumab exposure.

Inclusion criteria

• Postmenopausal women aged ≥50 and ≤85 years with osteoporosis- or osteopenia-related fractures who have received denosumab (Prolia) continuously for at least 3 years, with no dosing interval exceeding 9 months; and men aged ≥50 and ≤85 years with osteoporosis- or osteopenia-related fractures who have received continuous denosumab (Prolia) therapy for at least 3 years, with dosing intervals not exceeding 9 months.

Exclusion criteria

• Secondary osteoporosis Metabolic bone diseases Active or prior malignancy Ongoing systemic glucocorticoid therapy Current use of hormone replacement therapy Use of any medications that affect bone metabolism Estimated glomerular filtration rate (eGFR) < 50 mL/min/1.73 m² Known hypersensitivity to zoledronic acid Hypocalcemia Any other contraindications to zoledronic acid use