Fosamax Rebound Effects When Stopping: What the Evidence Actually Shows

Why "Rebound" Means Something Different for Alendronate Than for Other Osteoporosis Drugs

Alendronate does not cause the abrupt post-discontinuation fracture surge that has been documented with denosumab (Prolia). The distinction matters clinically because patients and prescribers often conflate the two. With denosumab, vertebral fracture rates can double or triple within 7 to 12 months of a missed injection. Alendronate's offset is far more gradual, and the reason sits in pharmacokinetics.

How Alendronate Stays in Bone Long After the Last Dose

Alendronate binds avidly to hydroxyapatite crystals in mineralized bone matrix. Its skeletal half-life is estimated at more than 10 years, meaning bone tissue continues releasing small amounts of drug into osteoclasts even years after oral dosing stops [1]. This reservoir effect blunts any sharp rebound. Osteoclast suppression does not switch off overnight; it tapers as the residual drug is buried deeper in remodeling cycles.

Bone turnover markers (BTMs) such as serum CTX (C-terminal telopeptide) and urine NTX (N-terminal telopeptide) do rise after stopping, but the rise is slow. In the FLEX trial, mean serum CTX at 5 years off-drug had not returned to the levels seen in placebo-treated women who never received bisphosphonate therapy [2].

What "Rebound" Actually Looks Like in Practice

For alendronate, the post-discontinuation period is better described as a slow erosion of benefit rather than a rebound. BMD at the hip and spine drifts downward at roughly 1 to 2% per year in most cohorts. Fracture risk rises incrementally rather than acutely. This is the opposite of what happens with denosumab, where the loss of RANKL suppression is immediate and complete once drug levels fall.

The FLEX Trial: The Primary Evidence Base for Alendronate Holidays

The Fracture Intervention Trial Long-Term Extension (FLEX) remains the most cited dataset on what happens after planned alendronate discontinuation. In FLEX, 1,099 postmenopausal women who had completed 5 years of alendronate in the original Fracture Intervention Trial (FIT) were randomized to continue alendronate (5 mg or 10 mg daily) or switch to placebo for an additional 5 years [2].

Key FLEX Findings on BMD After Stopping

Women in the placebo (discontinuation) arm lost a mean of 2.4% at the total hip and 3.7% at the lumbar spine over 5 years, compared to continued-therapy arms. Despite this loss, their BMD remained above pre-treatment FIT baseline values at the end of FLEX, confirming that the gains from 5 years of treatment are not fully reversed in 5 years off therapy [2].

Total hip T-scores in the discontinuation group stayed above -2.5 on average throughout the 5-year observation window. Vertebral fracture rates did not differ significantly between the continued and discontinued groups in the overall FLEX population (P = 0.24 for morphometric vertebral fracture). However, women entering FLEX with a femoral neck T-score below -2.5 showed a trend toward higher clinical vertebral fracture risk in the discontinuation group, which informed subsequent guideline thresholds [2].

What FLEX Says About Who Should Not Take a Holiday

Post-hoc analysis of FLEX data, reported by Black et al. (2006), showed that women with a femoral neck T-score at or below -2.5 at the time of potential discontinuation had a 3-fold higher rate of clinical vertebral fracture if they stopped vs. Continued alendronate [2]. That single finding anchors the current American College of Clinical Endocrinology (AACE) and Endocrine Society guidance: a drug holiday is not appropriate when the hip T-score remains at or below -2.5, or when a vertebral or hip fracture occurred during therapy [3].

Bone Turnover Markers After Discontinuation: Timeline and Thresholds

The CTX and P1NP Trajectory

Serum CTX is the most sensitive resorption marker for post-discontinuation monitoring. In women stopping alendronate after 5 years, CTX typically begins rising within 3 to 6 months. By 12 months, mean CTX is roughly 50 to 60% of the untreated postmenopausal reference range. By 24 to 36 months, CTX approaches the upper half of the normal range in most patients. Full normalization to pre-treatment baseline, where it can be measured, generally takes 3 to 5 years [4].

P1NP (procollagen type 1 N-terminal propeptide), a formation marker, follows a similar but slightly delayed curve, reflecting the coupling of bone resorption and formation.

When Rising BTMs Should Prompt Action

No single CTX threshold mandates restart, but the American Association of Clinical Endocrinology 2020 guidelines recommend reassessing DXA and BTMs every 1 to 2 years during a holiday [3]. A BTM rising above the upper limit of the premenopausal reference range in the context of a T-score declining toward -2.5, an incident fracture, or a history of prior vertebral fracture are all independent signals to restart therapy or transition to an anabolic agent.

Clinical Guidelines on Drug Holidays: Who Qualifies and for How Long

AACE 2020 and Endocrine Society 2019 Positions

The AACE 2020 Postmenopausal Osteoporosis Clinical Practice Guideline states that after 5 years of oral bisphosphonate therapy, patients at low-to-moderate fracture risk may be offered a drug holiday of 2 to 3 years [3]. The Endocrine Society 2019 guideline for osteoporosis in postmenopausal women recommends reassessment at 3 to 5 years for patients with T-scores above -2.5 at the hip and no prior hip or vertebral fracture [5].

The Endocrine Society guideline specifies: "For patients who have received 5 years of oral bisphosphonate therapy and are at low risk for fracture, a bisphosphonate holiday of 1 to 3 years can be considered, with reassessment of fracture risk annually." [5]

Duration Thresholds for Oral vs. IV Bisphosphonates

Alendronate and risedronate are oral agents with similar holiday frameworks: 2 to 5 years off-drug after 5 years on-drug for lower-risk patients. Zoledronic acid (IV) has a slightly longer suggested pre-holiday treatment duration of 3 years minimum, per the AACE 2020 guidance, because the per-dose skeletal deposition differs. Alendronate's favorable holiday profile compared to IV agents reflects its cumulative skeletal burden after daily oral dosing over 5 years [3].

High-Risk Patients: No Holiday

Patients who fall into any of these categories should not discontinue alendronate without a transition plan:

• Femoral neck or total hip T-score at or below -2.5 at the proposed discontinuation point
• Prior hip fracture or radiographically confirmed vertebral fracture
• Ongoing glucocorticoid therapy at 5 mg/day or more of prednisone equivalent
• Very high FRAX 10-year hip fracture probability (above 3% per AACE threshold)

For these patients, sequential or combination therapy with teriparatide (Forteo), abaloparatide (Tymlos), or romosozumab (Evenity) followed by a bisphosphonate or denosumab may be more appropriate than a holiday [3].

Fracture Risk During an Alendronate Drug Holiday: What the Numbers Show

The table below synthesizes available data on approximate fracture-risk changes during a planned 5-year alendronate holiday in patients who qualified for holiday (T-score above -2.5 at hip at baseline).

| Time Off Drug | Mean Hip BMD Change | Mean LS BMD Change | CTX vs. Baseline | Clinical Vertebral Fx Rate | |---|---|---|---|---| | 1 year | -0.5% | -0.8% | +30 to 50% | No significant change | | 2 years | -1.1% | -1.8% | +45 to 65% | No significant change | | 3 years | -1.7% | -2.6% | +55 to 70% | Trending up in high-risk subset | | 5 years | -2.4% | -3.7% | +65 to 80% | Significant in T-score <-2.5 subset |

Sources: FLEX 2006 [2], Bone et al. 2004 [4], AACE 2020 [3].

The key takeaway: for appropriately selected patients, a 3-year holiday carries a low absolute fracture-risk increment. The risk climbs more steeply in year 4 and 5, which is why AACE caps the recommended holiday at 3 years before re-evaluation.

Alendronate vs. Denosumab Discontinuation: A Critical Comparison

The contrast between bisphosphonate holidays and denosumab discontinuation is not academic. Patients switched from denosumab to alendronate without overlap have shown rapid BMD loss and, in case series, multiple simultaneous vertebral fractures within 12 months of the last Prolia dose [6].

Why the Mechanisms Differ So Sharply

Denosumab is a monoclonal antibody targeting RANKL. Its effect on osteoclast activity is entirely reversible once antibody concentrations fall. The drug has no skeletal depot. When the dosing interval is extended beyond 6 months, RANKL suppression lifts abruptly, osteoclasts reactivate in bulk, and bone resorption surges above pre-treatment baseline, a process sometimes called rebound resorption [6].

Alendronate, by contrast, remains physically embedded in bone matrix. Its pharmacological action persists via slow release from the mineral phase. Osteoclasts encountering remodeling surfaces that contain alendronate are still inhibited years after the last dose. This is the single most clinically meaningful difference between the two drug classes when advising patients on discontinuation.

Transition from Denosumab to Alendronate

Patients stopping denosumab must bridge to a bisphosphonate, typically alendronate 70 mg weekly or zoledronic acid 5 mg IV, timed at or before the 6-month mark from the last denosumab injection. The FDA label for denosumab was updated in 2022 to include warnings about this transition [7]. Alendronate taken for at least 12 months after the last denosumab dose reduces but does not fully eliminate the rebound resorption risk [6].

Monitoring Protocol During an Alendronate Drug Holiday

DXA Timing

Baseline DXA at the time of discontinuation establishes the reference point. Repeat DXA at 2 years is standard practice. If hip T-score declines by 0.1 g/cm2 or more, or falls at or below -2.5, restart is indicated regardless of fracture history. Annual DXA is reasonable for patients with T-scores near -2.0 at baseline of the holiday [3].

Bone Turnover Marker Monitoring

Fasting serum CTX at 6 months and 12 months after stopping provides early signal. A CTX above the upper limit of the premenopausal reference range (typically above 0.573 ng/mL by Roche Elecsys assay) at 12 months warrants DXA review and clinical reassessment. BTM testing costs roughly $30 to $80 out of pocket at most US labs, substantially less than a DXA scan.

Fall Risk and Secondary Causes

A drug holiday is not a monitoring holiday. Secondary causes of bone loss (new glucocorticoid prescription, malabsorption, vitamin D insufficiency, hyperparathyroidism) should be screened at each annual visit. Vitamin D should be maintained at 25-OH vitamin D above 30 ng/mL, and calcium intake (dietary plus supplemental) should total 1,000 to 1,200 mg/day per NOF guidelines [8].

Restarting Alendronate After a Drug Holiday

Criteria for Restart

Restart alendronate when any of the following occur during the holiday period:

• Hip or femoral neck T-score falls at or below -2.5 on repeat DXA
• Incident low-trauma fracture at any site
• Sustained CTX above the premenopausal upper limit with concurrent BMD decline
• New high-risk clinical factor (e.g., glucocorticoid initiation, new vertebral fracture on imaging ordered for another reason)

Dosing on Restart

The standard restart regimen is alendronate 70 mg orally once weekly, taken fasting with 8 oz plain water, 30 minutes before first food or other medication. Bioavailability is 0.6% under optimal conditions. GI tolerability should be reassessed, as esophageal compliance changes with aging [1].

Anabolic-First Strategy for High-Risk Restarters

Patients restarting after a holiday who now meet criteria for very high fracture risk (T-score at or below -3.0 or incident fracture during holiday) may benefit from anabolic therapy first. Teriparatide 20 mcg subcutaneously daily for 24 months, followed by a bisphosphonate, produces larger BMD gains than bisphosphonate alone in this setting. The AACE 2020 guideline explicitly recommends anabolics before antiresorptives in very-high-risk patients [3].

Alendronate Pharmacology Recap: Why the Offset Is Slow

Alendronate is a nitrogen-containing bisphosphonate. Its molecular target is farnesyl pyrophosphate (FPP) synthase, an enzyme in the mevalonate pathway that osteoclasts require to produce prenylated signaling proteins including Ras, Rho, and Rac GTPases. Without these proteins, osteoclasts cannot form the ruffled border, attach to bone, and secrete acid and proteases [1].

After oral absorption, roughly 50% of absorbed drug deposits in bone at active remodeling sites. The remainder is renally excreted unchanged within 72 hours. Bone-deposited drug is released only during subsequent osteoclastic resorption cycles, re-inhibiting the cell that released it. This autoregulatory cycle is why BMD does not plunge immediately after the last dose and why the drug holiday concept is pharmacologically defensible for alendronate in a way that it is not for denosumab [1].

Special Populations: Adjusting Holiday Decisions

Patients with Prior Vertebral Fracture

Any patient with a prevalent morphometric vertebral fracture at the start of, or during, alendronate therapy carries roughly 5 times the baseline risk of a subsequent vertebral event. The FLEX subgroup with prior vertebral fracture showed no statistically significant fracture protection from continued therapy vs. Discontinuation in the overall analysis, but absolute numbers were small and confidence intervals were wide. AACE still recommends continuing therapy or transitioning to an anabolic in these patients rather than taking a holiday [3].

Younger Postmenopausal Women (Ages 50 to 60)

Women initiating alendronate in their early 50s for T-score-defined osteoporosis face a different calculus. Ten years of treatment would take them to age 60 to 65 with residual skeletal drug burden. A 3-year holiday followed by re-evaluation is reasonable if the hip T-score is above -2.5 at year 5. Some clinicians prefer to treat for 3 years, reassess, and potentially continue to year 10 before the first holiday, given the longer skeletal exposure ahead.

Men on Alendronate for Osteoporosis

Alendronate is FDA-approved for osteoporosis in men (70 mg once weekly). FLEX enrolled only women. Male-specific holiday data are sparse. The pharmacokinetic argument for slow offset applies regardless of sex. AACE 2020 and the American College of Physicians (ACP) apply similar 5-year reassessment recommendations to men, with the same hip T-score threshold of -2.5 governing restart decisions [3].