Denosumab Rebound on Discontinuation: What Happens When You Stop and How to Prevent It

Why Denosumab Is Different From Every Other Osteoporosis Drug

Denosumab works in a fundamentally different way from bisphosphonates. It does not bind to bone mineral. Instead, it blocks RANK ligand (RANKL), the signaling protein that activates osteoclasts. Twice-yearly subcutaneous injections of 60 mg suppress osteoclast activity for roughly six months. When the drug clears the body, RANKL activity rebounds sharply, osteoclasts flood bone resorption sites, and bone mineral density (BMD) drops fast. Bisphosphonates, by contrast, bind hydroxyapatite and keep suppressing resorption for years after the last dose.

This reversibility is biologically elegant during treatment. After treatment, it is a clinical hazard.

The RANKL Rebound Mechanism

When denosumab is present, RANKL is neutralized and osteoclast formation is suppressed. Once serum levels fall below the therapeutic threshold (approximately 6 months after injection), RANKL rises above pre-treatment levels in a compensatory overshoot. A 2018 paper in the Journal of Bone and Mineral Research documented that serum C-terminal telopeptide (CTX), a marker of bone resorption, exceeded pre-treatment values within 3 to 6 months of the last dose and remained elevated for up to 12 months.

This is not simple loss of drug effect. The rebound exceeds the baseline rate, meaning bone is lost faster than it was being lost before treatment ever started.

How Much BMD Is Lost

In the FREEDOM extension trial (N=4,550), patients who discontinued denosumab after up to 10 years of therapy lost all accumulated lumbar spine BMD gains within approximately 12 months. Hip BMD returned to baseline within 18 to 24 months. The magnitude of loss was roughly 6 to 8 percentage points at the lumbar spine in the first year, a rate 3 to 4 times faster than the natural bone loss seen in untreated postmenopausal women.

Patients who had received denosumab for longer durations lost more absolute BMD in the rebound phase, because they had more to lose.

The Rebound Fracture Signal: What the Data Show

The most alarming consequence of the denosumab rebound is not just lower BMD. It is a cluster of multiple simultaneous vertebral fractures occurring within 12 to 18 months of stopping.

Incidence Estimates

A 2019 systematic review and meta-analysis published in Osteoporosis International pooled data from 11 studies and found that the incidence of vertebral fractures after denosumab discontinuation ranged from 1.4% to 7.1%, with multiple simultaneous fractures (defined as three or more vertebrae) occurring in up to 60% of fracture cases. That pattern is not seen at that frequency with any other osteoporosis drug discontinuation.

A separate Danish registry cohort published in the Journal of Bone and Mineral Research (2020) followed 1,462 women who stopped denosumab and found a vertebral fracture hazard ratio of 1.98 (95% CI: 1.35-2.91) compared with women who continued treatment, after adjusting for baseline fracture risk.

Who Is at Highest Risk

The patients most likely to experience rebound fractures share several characteristics:

• Prior vertebral fracture before or during treatment
• Lumbar spine T-score below -2.5 at baseline
• Duration of denosumab therapy exceeding 2 years (longer treatment, sharper rebound)
• Age above 70 at discontinuation
• No sequential therapy initiated after stopping

The 2022 American Society for Bone and Mineral Research (ASBMR) task force position paper states directly: "All patients discontinuing denosumab should receive sequential antiresorptive therapy to mitigate bone loss and fracture risk." That recommendation is grade B, based on consistent observational data and mechanistic plausibility.

The Only Proven Prevention Strategy: Sequential Bisphosphonate Therapy

Stopping denosumab without a transition plan is now considered below the standard of care at most major academic centers. The transition window is narrow and the timing matters.

Zoledronic Acid: The First-Choice Option

Zoledronic acid (Reclast) 5 mg IV, given once approximately 6 months after the last denosumab injection, is the most studied transition agent. A prospective study by Anastasilakis et al. (2019) in the Journal of Bone and Mineral Research showed that a single dose of zoledronic acid given 6 months post-denosumab preserved lumbar spine BMD at 12 months (mean change: -0.4%, not statistically significant) compared with a 5.6% loss in the untreated control arm (P<0.001).

A second zoledronic acid dose at 12 months is recommended for patients with very low BMD (T-score <-2.5) or prior vertebral fracture, based on the 2022 ASBMR position paper.

Timing is everything. Giving zoledronic acid too early (before 5 months post-injection) risks blunting the drug's uptake onto bone because osteoclast activity, the mechanism zoledronic acid exploits to bind bone surface, is still suppressed. Giving it too late (past 7 months) leaves a window in which unprotected rebound resorption occurs.

Oral Alendronate: The Alternative for Patients Who Cannot Receive IV Therapy

Alendronate 70 mg once weekly, started 6 months after the last denosumab injection, is the second-line option. A 2020 randomized controlled trial published in Bone assigned 63 postmenopausal women to alendronate or placebo after denosumab discontinuation and found that alendronate significantly attenuated BMD loss at the lumbar spine (-1.3% vs. -5.0% at 12 months; P<0.01) and at the total hip (-0.9% vs. -3.4%; P<0.05). Alendronate does not fully prevent rebound-phase BMD loss, but it reduces it substantially.

Some patients with esophageal conditions, low creatinine clearance, or prior IV infusion reactions are better served by alendronate. Renal function should be checked before zoledronic acid: the FDA prescribing information contraindicates zoledronic acid in patients with creatinine clearance <35 mL/min.

How Long to Continue the Sequential Bisphosphonate

The HealthRX clinical team uses the following decision framework for duration of sequential bisphosphonate therapy after denosumab:

• Low residual risk (T-score above -2.0 at time of denosumab stop, no prior fracture): zoledronic acid once, then reassess BMD at 24 months.
• Moderate residual risk (T-score -2.0 to -2.5, no prior vertebral fracture): zoledronic acid at 6 months and again at 18 months, then reassess.
• High residual risk (T-score <-2.5 or any prior vertebral fracture): two doses of zoledronic acid plus oral alendronate continuation to 2 years minimum, followed by a full fracture risk reassessment using FRAX.

This framework is consistent with the stratified approach described in the 2022 ASBMR position statement but adds explicit T-score thresholds to make clinical decision points concrete.

Osteonecrosis of the Jaw: Real Risk, Real Rarity

Osteonecrosis of the jaw (ONJ) is exposed, necrotic bone in the jaw that fails to heal over eight weeks in a patient taking or recently treated with antiresorptive therapy. It is a legitimate concern with denosumab, but the absolute risk in osteoporosis patients is very low.

Incidence in Osteoporosis vs. Oncology Doses

Denosumab is used at two completely different doses for two different indications. Prolia (60 mg every 6 months) is used for osteoporosis. Xgeva (120 mg monthly) is used in oncology for bone metastases. ONJ rates differ sharply between the two:

• Osteoporosis dose (Prolia): 0 to 0.09% per patient-year, based on pooled FREEDOM and extension trial data reviewed by the FDA MedWatch program
• Oncology dose (Xgeva): 1 to 2% per year, per a 2014 review in the Journal of the National Cancer Institute

The FDA issued a drug safety communication in 2013 confirming ONJ as a labeled risk for both formulations. The label requires that prescribers advise patients to maintain good oral hygiene and avoid invasive dental procedures without discussion with their prescribing physician.

Risk Factors for ONJ

Patients at higher ONJ risk include those with:

• Active dental infection or periodontal disease
• Recent tooth extraction or implant placement
• Concurrent corticosteroid therapy
• Diabetes or smoking history
• Duration of antiresorptive therapy beyond 3 to 4 years

The standard recommendation from the American Association of Oral and Maxillofacial Surgeons is to complete all elective invasive dental work before starting denosumab when possible, and to time non-urgent extractions to coincide with a period of lower antiresorptive activity. For denosumab, that window is months 4 to 6 after the last injection, when drug levels are falling.

Management If ONJ Occurs

Conservative management is first-line: antimicrobial mouth rinses (chlorhexidine 0.12%), pain control, and avoidance of further trauma. Surgery is reserved for stage III disease (exposed bone with pathological fracture, fistula, or extension to inferior border of mandible). The 2022 AAOMS position paper on medication-related ONJ provides staging criteria and management algorithms.

Atypical Femoral Fractures: Low Absolute Risk, Long-Term Signal

Atypical femoral fractures (AFFs) are stress fractures with a distinctive radiographic appearance: transverse orientation, cortical thickening at the fracture site, and minimal or no trauma. They occur in the subtrochanteric or diaphyseal femur, outside the typical osteoporotic fracture location.

How Common Are They With Denosumab?

The 2014 ASBMR task force report on AFFs estimated the background rate at approximately 2 to 3 per 100,000 person-years in the general population. With bisphosphonate use beyond 5 years, this rises to 100 per 100,000 person-years. Data for denosumab are less mature, but a 2020 analysis of the FREEDOM extension (N=4,550, up to 10 years of treatment) identified 12 AFF cases across 34,422 patient-years of exposure, giving a rate of approximately 35 per 100,000 patient-years for long-term denosumab users.

That rate is elevated compared with the background population rate, but lower than long-term bisphosphonate rates in the same studies.

Symptoms and Diagnosis

Patients with an impending AFF often report prodromal thigh pain for weeks or months before fracture completion. Any antiresorptive patient reporting new groin or thigh pain should have AP and lateral femur X-rays ordered promptly. A 2014 NEJM case series showed that 70% of complete AFFs had prodromal symptoms that went unrecognized for a median of 4 months.

MRI or bone scan is recommended when X-ray is negative but clinical suspicion remains high, per ASBMR AFF task force guidelines.

Management of AFFs During Denosumab Therapy

Complete AFFs require surgical fixation with an intramedullary nail. Incomplete AFFs (stress reactions without cortical breach) are managed with protected weight-bearing and, critically, stopping or pausing the antiresorptive agent. Teriparatide (Forteo) 20 mcg/day subcutaneously has been used to accelerate healing in incomplete AFFs; a 2012 case series in the Journal of Bone and Joint Surgery documented radiographic healing in 8 of 11 patients treated with teriparatide after bisphosphonate-associated AFFs, a finding that informs (though does not directly prove) its use in denosumab cases.

If denosumab is stopped for an AFF, the rebound discontinuation protocol described above applies. Stopping without a sequential bisphosphonate exposes the patient to rebound vertebral fracture risk on top of an existing femoral injury.

When Stopping Denosumab Is Appropriate

Not every patient should remain on denosumab indefinitely. Clinically reasonable reasons to discontinue include:

• Significant improvement in fracture risk (T-score above -2.0 and no prevalent fractures after 5 to 10 years of treatment)
• New contraindication: hypocalcemia, severe renal impairment, pregnancy
• Serious adverse event directly attributed to denosumab
• Patient preference after shared decision-making about risk

The Endocrine Society 2019 clinical practice guideline on osteoporosis notes that treatment holidays are well established for bisphosphonates but "not appropriate for denosumab" because of the rebound phenomenon. This is a critical distinction. A bisphosphonate holiday after 5 years is evidence-based. A denosumab holiday without a transition plan is not.

The Drug Holiday Misconception

Patients and even some clinicians transfer the concept of the bisphosphonate drug holiday to denosumab. The two drugs are not interchangeable in this context. Alendronate's half-life in bone is approximately 10 years. Denosumab's half-life is 26 days, with bone effects wearing off fully in 6 months. Skipping a denosumab injection by even a few weeks increases fracture risk, as demonstrated in a 2016 cohort study in Osteoporosis International that showed a 64% higher odds of vertebral fracture with late injections (greater than 7.5 months apart) compared with on-schedule injections.

Monitoring During and After Denosumab Therapy

During Treatment

• BMD by DXA at baseline and every 1 to 2 years
• Serum calcium before each injection (hypocalcemia is a contraindication)
• Serum 25-hydroxyvitamin D: maintain above 30 ng/mL to reduce hypocalcemia risk
• Dental evaluation before initiating therapy; annually thereafter for high-risk patients
• Femur X-ray if any new thigh or groin pain develops

After Stopping

• Bone turnover markers (CTX, P1NP) at 3 and 6 months post-last injection to confirm rebound
• DXA at 12 months post-discontinuation to assess BMD preservation with sequential therapy
• Vertebral fracture assessment (VFA) or thoracolumbar X-ray at 12 to 18 months, especially in patients with prior vertebral fractures, per ISCD 2019 guidelines

A rising CTX above the upper limit of the reference range at month 3 post-denosumab, before the transition bisphosphonate has fully taken effect, is not a treatment failure. It is expected physiology. The question is whether CTX returns to the premenopausal reference range by months 9 to 12 after the sequential agent is started.