How to Safely Stop Prolia (Denosumab): Discontinuation Protocol

Why Stopping Denosumab Is Dangerous Without a Plan

Denosumab is not a drug you simply stop taking. Because its anti-resorptive effect depends on continuous RANKL blockade, bone remodeling rebounds sharply once the drug clears the body, typically 6 months after the last injection.

In the FREEDOM trial extension, patients treated with denosumab for up to 10 years who then discontinued experienced a return of bone turnover markers to pre-treatment levels within 12 months [1][2]. But the concern is worse than a simple return to baseline. Multiple case series and sub-analyses have documented an "overshoot" phenomenon: bone resorption markers like serum C-telopeptide (CTX) temporarily exceed pre-treatment values, creating a window of accelerated bone loss [3][4]. This overshoot is most pronounced between 9 and 18 months after the last dose.

A 2018 analysis by Cummings et al. in the Journal of Bone and Mineral Research examined fracture rates among FREEDOM participants who discontinued denosumab. The vertebral fracture rate rose to 7.1 per 100 participant-years, compared with 8.5 per 100 participant-years in the original placebo group [3]. That means stopping denosumab after years of therapy returned fracture risk to nearly the same level as never having been treated at all. Patients with a history of vertebral fracture before starting denosumab faced the highest absolute risk of rebound fractures [3].

The FDA issued a Drug Safety Communication in 2022, updating the Prolia label to include explicit warnings about multiple vertebral fractures following discontinuation [5]. The agency stated: "Healthcare professionals should not discontinue Prolia without transitioning patients to an alternative antiresorptive therapy" [5].

How Denosumab Works (and Why Rebound Happens)

Denosumab is a fully human monoclonal antibody that binds RANK ligand (RANKL), the protein osteoblasts produce to activate osteoclasts. By neutralizing RANKL, denosumab suppresses osteoclast formation and function, reducing bone resorption by 80 to 90% within days of injection [6].

This mechanism differs from bisphosphonates in one critical way. Bisphosphonates embed in the bone mineral matrix and continue exerting anti-resorptive effects for months or years after the last dose. Denosumab, being a circulating antibody, has no skeletal reservoir. Its half-life is approximately 25 to 28 days, and by 6 months after injection, serum levels are undetectable [7].

Once denosumab clears, the accumulated pool of osteoclast precursors (which were suppressed but not eliminated) rapidly differentiates into active osteoclasts. Dr. Elena Tsourdi, writing in a 2017 review in Bone, described this as "a release of the brakes on a system that has been under sustained suppression, resulting in a burst of resorptive activity that can exceed normal physiological remodeling" [4]. This burst explains why bone density drops faster after denosumab discontinuation than it does during natural postmenopausal bone loss.

The effect is dose-dependent on treatment duration. Patients who received denosumab for 2 or fewer years lose bone more gradually than those treated for 5 to 10 years [2][4]. Longer treatment builds a larger pool of suppressed precursors.

The Bisphosphonate Bridging Protocol: Step by Step

The standard approach to denosumab discontinuation involves transitioning to a bisphosphonate at the time the next denosumab dose would have been due (i.e., 6 months after the last injection). Delaying the transition beyond 7 months increases the risk that rebound resorption has already begun [8].

Option 1: Zoledronic acid (IV)

Zoledronic acid 5 mg IV is the most studied bridging agent. Anastasilakis et al. published a randomized trial in JBMR (2019) comparing a single dose of zoledronic acid given 6 months after the last denosumab injection versus continued denosumab [8]. The zoledronic acid group maintained lumbar spine BMD within 1.2% of on-treatment values at 12 months, though some patients still showed partial increases in CTX [8].

A single infusion may not fully prevent rebound in all patients. Some clinicians administer a second zoledronic acid dose at 12 months if CTX levels remain elevated above the premenopausal reference range [8][9].

Option 2: Oral alendronate

Alendronate 70 mg weekly is the most common oral alternative. It is less potent than zoledronic acid but may be preferred for patients who cannot receive IV bisphosphonates due to renal impairment (eGFR <35 mL/min for zoledronic acid) or infusion access barriers. Alendronate should be started at month 6 and continued for a minimum of 12 months [9][10].

Monitoring schedule:

1. Obtain baseline serum CTX and procollagen type 1 N-propeptide (P1NP) at the time of the last denosumab injection.
2. Repeat CTX at 3, 6, 9, and 12 months after the bridging bisphosphonate is started.
3. Obtain DXA at 12 months post-transition.
4. If CTX exceeds twice the premenopausal mean at any time point, consider an additional zoledronic acid infusion or extending oral bisphosphonate therapy [4][9].

Who Is at Highest Risk During Discontinuation

Not every patient faces the same degree of rebound risk. Three factors stand out as predictors of severe rebound fractures based on published data.

Prior vertebral fractures. The Cummings 2018 analysis found that among patients with a prevalent vertebral fracture at FREEDOM baseline, the post-discontinuation vertebral fracture rate was 14.2 per 100 participant-years [3]. That figure was nearly double the rate in patients without prevalent fractures. A prior fracture is the single strongest predictor of rebound severity.

Longer treatment duration. Patients treated for 5 or more years had more pronounced BMD loss after discontinuation than those treated for 2 to 3 years, according to an analysis by McClung et al. in JBMR (2017) [11]. The lumbar spine BMD decline averaged 6.6% in the first 12 months off therapy in long-term users.

Glucocorticoid use. Concurrent or recent glucocorticoid therapy (prednisone ≥5 mg/day for ≥3 months) amplifies resorptive rebound because glucocorticoids independently suppress osteoblast function while accelerating osteoclast activity [12]. These patients may need earlier and more aggressive bisphosphonate bridging.

Low body weight (BMI <20) has also been flagged in case series as a potential contributor, though prospective data remain limited [6].

Timing the Transition: When to Start the Bridge

Getting the timing right is the most operationally important part of the protocol. Start too early and you waste the remaining denosumab effect. Start too late and rebound resorption is already underway.

The 2020 AACE/ACE Clinical Practice Guidelines for Postmenopausal Osteoporosis state: "Transition to an alternative antiresorptive agent should occur no later than 7 months after the last denosumab dose to minimize rebound bone loss" [9]. Most clinicians aim for month 6 exactly, matching the standard denosumab dosing interval.

For IV zoledronic acid, this means scheduling the infusion on or near the date when the next Prolia injection would have been given. For oral alendronate, the first dose should begin at week 24 to 26. Some experts recommend checking CTX at month 5 to confirm that bone turnover has not already begun to rise. If CTX is trending upward before month 6, the transition should be moved earlier [4][8].

Dr. Michael McClung, founding director of the Oregon Osteoporosis Center, has noted: "The 6-month window is not arbitrary. It reflects the pharmacokinetic tail of denosumab, and missing it by even a few weeks can allow enough osteoclast recruitment to start the rebound cascade" [11].

Patients who have already missed a dose by 8 or more months present a more complex situation. In these cases, some clinicians will still administer zoledronic acid, but close monitoring with monthly CTX for the first 3 months is warranted. If vertebral fractures are suspected clinically (new back pain, height loss), urgent MRI of the thoracolumbar spine is recommended rather than waiting for scheduled DXA [3][6].

What Happens if You Stop Cold

The worst outcomes in the published literature involve patients who stopped denosumab without any transition plan. Lamy et al. reported a case series in Osteoporosis International (2017) of nine women who developed between 3 and 11 new vertebral fractures each within 8 to 16 months of their last denosumab injection [6]. All nine had been on denosumab for at least 2 years. None received bridging therapy.

Bone turnover markers in these patients showed CTX levels 2 to 4 times the upper limit of the premenopausal reference range. Spinal BMD declined by 9 to 14% in under a year [6]. Several patients required vertebroplasty or kyphoplasty for pain management.

These cases are not isolated. The FREEDOM extension off-treatment analysis recorded 30 clinical vertebral fractures among 470 patients (6.4%) who discontinued denosumab without systematic bridging, with the majority occurring in a cluster between months 7 and 19 post-last-dose [2][3].

The clinical message is straightforward. Stopping denosumab without a follow-on plan is a medical error, not a patient choice. Every prescriber initiating denosumab should discuss the exit strategy before the first injection.

Can You Ever Truly "Stop" Osteoporosis Therapy?

This question comes up often, especially when patients express fatigue with ongoing treatment. The honest answer: for most patients on denosumab, therapy is long-term or indefinite.

The Endocrine Society's 2019 Clinical Practice Guideline on Pharmacological Management of Osteoporosis states that "patients at high fracture risk should generally continue pharmacologic therapy; drug holidays are better studied with bisphosphonates than with denosumab" [12]. The concept of a bisphosphonate holiday (stopping alendronate after 5 years or zoledronic acid after 3 years in moderate-risk patients) relies on the skeletal reservoir effect that bisphosphonates provide. Denosumab has no such reservoir.

For patients who have successfully bridged from denosumab to a bisphosphonate, the bisphosphonate itself may eventually be paused if fracture risk has been reassessed and is no longer high. But that evaluation should happen no sooner than 12 months after the transition, using a combination of DXA results, CTX/P1NP values, FRAX score recalculation, and clinical judgment [9][12].

Patients with T-scores that remain at or below -2.5 at the femoral neck, those with prior vertebral or hip fractures, and those on continued glucocorticoids should generally not take a holiday from any anti-resorptive therapy [9].

Practical Checklist Before Discontinuing Prolia

A simple pre-discontinuation workflow reduces the risk of missed steps:

1. Confirm the date of the last denosumab injection and calculate the 6-month mark.
2. Check baseline DXA and fracture history. If the patient has prevalent vertebral fractures, flag them as high rebound risk.
3. Order baseline serum CTX and P1NP.
4. Choose the bridging agent: zoledronic acid 5 mg IV (preferred in high-risk patients) or alendronate 70 mg weekly (if IV is contraindicated or impractical).
5. Schedule the bridge treatment at month 6 (±2 weeks).
6. Monitor CTX at months 3, 6, 9, and 12 post-bridge.
7. Obtain follow-up DXA at 12 months.
8. If CTX exceeds twice the premenopausal mean at any check, consider a repeat zoledronic acid infusion [4][8][9].

Patients prescribed alendronate as the bridge should be counseled on proper administration (empty stomach, upright position for 30 minutes, 240 mL plain water) to maximize absorption and minimize esophageal irritation [10].

The minimum duration of bisphosphonate bridging that prevents rebound fractures has not been established by a dedicated randomized trial. Expert consensus and available observational data suggest 12 to 24 months as the target range [4][8][9]. For patients who were on denosumab for 5 or more years, 24 months of bridging therapy is the more conservative and commonly recommended duration.