After denosumab, a potent bisphosphonate — the pooled numbers behind a rule you already half-follow

Design

PRISMA 2020 systematic review and random-effects meta-analysis of six studies with GRADE certainty assessment, searched to May 2026

Population

adults with osteoporosis or low bone mass who discontinued denosumab; 683 participants in the primary vertebral fracture analysis

Primary outcome

incident vertebral fracture after denosumab cessation with potent bisphosphonate versus no antiresorptive or a SERM

Effect

incident vertebral fracture RR 0.29 (95% CI 0.14-0.59); multiple vertebral RR 0.08 (0.01-0.43); clinical vertebral RR 0.18 (0.07-0.49); any fracture RR 0.35 (0.10-1.24), not significant

Denosumab cessation causes rebound bone turnover and rapid bone loss, and guidelines already advise following it with a bisphosphonate. This GRADE-assessed review pooled the evidence for the first time: six studies, with 683 participants in the primary analysis, of adults who stopped denosumab and then received alendronate or zoledronic acid, compared with no subsequent antiresorptive or with a selective oestrogen receptor modulator.

The vertebral fracture results are striking. Incident vertebral fractures fell by 71% (RR 0.29, 95% CI 0.14 to 0.59), multiple vertebral fractures by 92% (RR 0.08, 95% CI 0.01 to 0.43), and clinical vertebral fractures by 82% (RR 0.18, 95% CI 0.07 to 0.49). What did not reach significance was any fracture (RR 0.35, 95% CI 0.10 to 1.24) and non-vertebral fracture (RR 0.43, 95% CI 0.07 to 2.64) — which fits the mechanism, since the rebound phenomenon is characteristically vertebral.

Two honest qualifications. The protective effect was driven almost entirely by comparisons against no subsequent therapy; the comparison against a selective oestrogen receptor modulator was too limited to conclude anything, so this does not tell you a bisphosphonate is the best sequential agent, only that something is far better than nothing. And overall certainty was graded low to very low, because most included studies were observational and several outcomes imprecise. That is enough to make sequential bisphosphonate the default rather than a discussion, and not enough to close the question of which agent, at what dose, and when.

• Make a sequential bisphosphonate the default for every patient stopping denosumab, not a case-by-case decision
• Expect the benefit at the vertebrae; non-vertebral and overall fracture reduction were not demonstrated
• Do not use this to argue a bisphosphonate beats a SERM — that comparison was inconclusive
• Zoledronic acid or alendronate were the agents studied; other bisphosphonates were not
• Recognise low to very low certainty and mostly observational studies when quoting the 71% figure

The statistics, in plain English

A risk ratio of 0.08 for multiple vertebral fractures sounds extraordinary, and its interval (0.01 to 0.43) is very wide because the events are few — the effect is real but its size is loosely estimated. The non-significant results for any fracture (0.10 to 1.24) and non-vertebral fracture (0.07 to 2.64) have intervals so wide they exclude almost nothing; these are underpowered comparisons, not evidence of no effect. GRADE certainty of low to very low means further research is likely to change the estimate, chiefly because most of the 683 participants came from observational studies where the patients given a follow-on drug differed from those who were not.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.