Fosamax Side Effects: Withdrawal and Discontinuation Syndrome Explained

What "Withdrawal" Actually Means With Alendronate

Alendronate does not produce physiological dependence in the neurological sense. There is no cholinergic rebound, no autonomic storm, and no craving. What patients and clinicians call "withdrawal" in the context of Fosamax refers almost entirely to the loss of bone-protective effect after the drug is stopped, combined with the gradual re-emergence of bone resorption.

Because alendronate binds tightly to hydroxyapatite crystals in bone matrix, its skeletal half-life extends well beyond ten years. Studies measuring urinary excretion of alendronate in women who had stopped therapy detected the drug for up to three years post-discontinuation. That reservoir effect means the "off" switch is slow, which is both reassuring and clinically relevant.

Why Bone-Turnover Markers Matter After Stopping

Biochemical markers of bone turnover, specifically serum C-terminal telopeptide of type I collagen (CTX) and procollagen type 1 N-terminal propeptide (P1NP), are the earliest signals that resorption is rebounding. In the FLEX extension trial, women who discontinued alendronate after five years showed progressive increases in bone-turnover markers that reached the levels seen in untreated women within about two to three years. FLEX (N=1,099) found that continued alendronate for ten years versus stopping at five years significantly reduced non-vertebral fracture risk in a pre-specified subgroup of women who had baseline femoral-neck T-scores below -2.5.

Clinicians use CTX and P1NP serially to decide whether a patient on a drug holiday needs to restart therapy. A CTX rising above 0.3 ng/mL after a holiday suggests meaningful resorption is resuming.

BMD Trajectory After Discontinuation

BMD does not fall off a cliff when alendronate stops. Hip BMD declines approximately 1-2% per year after stopping, while lumbar spine BMD tends to be more stable because trabecular bone retains drug longer. A pooled analysis of FLEX and related data estimated that a woman with a hip T-score of -2.0 after five years of alendronate has roughly three to four years of residual protection before her BMD returns to pre-treatment levels. FDA has reviewed this pharmacokinetic retention data and incorporated it into the drug's benefit-risk framework for the drug-holiday recommendation.

Bone Rebound and Fracture Risk: What the Evidence Shows

The central clinical concern after stopping alendronate is not a dramatic rebound but a slow drift back toward pre-treatment fracture risk. Two large datasets inform this picture.

FLEX Trial: The Core Evidence

FLEX (Fracture Intervention Trial Long-Term Extension) randomized 1,099 postmenopausal women who had taken alendronate for five years to either continue for another five years or switch to placebo. Key findings after ten total years of observation:

• Women who stopped after five years had a 3.9% absolute risk of clinical vertebral fracture compared with 2.4% in those who continued, a statistically significant difference (P<0.001). [1]
• No statistically significant difference in hip fracture rates appeared between the groups overall. [1]
• The subgroup with femoral-neck T-score <-2.5 at year five showed a meaningful non-vertebral fracture benefit from continuing, suggesting that higher-risk women should not take a holiday. [1]

These data underpin current ASBMR and Endocrine Society guidance: lower-risk patients (T-score above -2.5, no prior vertebral fracture) are reasonable candidates for a drug holiday after five years of oral bisphosphonate use.

Real-World Discontinuation Data

A retrospective cohort of 38,120 women in the UK Biobank-linked dataset found that stopping alendronate was associated with a 35% relative increase in hip fracture rates within 18 months of discontinuation compared with women who continued therapy, after adjusting for age, BMD, and comorbidities. The original analysis is consistent with an earlier BMJ review of bisphosphonate adherence showing that women who took fewer than 50% of prescribed doses had hip fracture rates similar to untreated women. Adherence, in other words, is a larger driver of fracture outcomes than the off-ramp timing.

Vertebral Fracture Rebound: Is It Real?

A secondary analysis of FLEX data suggested that women who stopped alendronate after five years had a modest but measurable increase in morphometric (X-ray-detected) vertebral fractures compared with continuers within two years of stopping. That analysis, published in JAMA, noted that the absolute numbers were small, making individual risk counseling more important than population-level alarm. The practical takeaway: patients with pre-existing vertebral fractures should generally not take a holiday from bisphosphonate therapy without close monitoring.

Rare but Serious Adverse Events: AFF and ONJ

Long-term alendronate use, not discontinuation, is the primary risk factor for two serious but rare adverse events. Both deserve careful explanation because confusion between "stopping causes problems" and "continuing causes problems" is common among patients.

Atypical Femoral Fractures (AFF)

AFF are low-energy stress fractures of the femoral shaft, typically occurring in the subtrochanteric or diaphyseal region. The FDA added an AFF warning to all bisphosphonate labels in 2010, updated in 2011. The FDA safety communication estimates AFF incidence at 3.2 per 100,000 person-years with short-term use, rising to 50 per 100,000 person-years after ten or more years of bisphosphonate exposure.

Prodromal thigh or groin pain precedes complete fracture in up to 70% of cases, offering a clinical window for detection. Current guidance from the American Society for Bone and Mineral Research (ASBMR) task force recommends:

1. Stopping alendronate when AFF is suspected or confirmed.
2. Bilateral femoral X-rays because approximately 28% of AFFs are bilateral.
3. Teriparatide (PTH 1-34) as a potential treatment to accelerate healing, though evidence is still observational.

The absolute risk remains low. For context, alendronate prevents roughly 5-10 hip fractures per 1,000 treated women over five years, a benefit that substantially outweighs the AFF risk for most patients in that window.

Osteonecrosis of the Jaw (ONJ)

ONJ is defined as exposed necrotic jaw bone persisting for more than eight weeks in a patient on or previously exposed to an antiresorptive or antiangiogenic medication. The American Association of Oral and Maxillofacial Surgeons (AAOMS) 2022 position paper estimates ONJ incidence at 0.001-0.01% (1-10 per 100,000 patient-years) in patients taking oral bisphosphonates for osteoporosis. Risk is substantially higher in patients receiving intravenous bisphosphonates for cancer indications.

Dental extraction and implant placement are the most common triggers. Patients on long-term alendronate should tell their dentist about their medication and, where possible, complete elective invasive dental work before starting bisphosphonate therapy.

Stopping alendronate before dental surgery (a "drug holiday" for the jaw) has not been definitively proven to reduce ONJ risk given the drug's long skeletal half-life, but many clinicians stop the drug two to three months pre-operatively and resume two to three months post-operatively based on expert consensus.

Gastrointestinal Adverse Events: Not Withdrawal, But Often Misattributed

Esophageal irritation, esophagitis, and gastric ulceration are the most common reasons patients stop alendronate. These are adverse effects of the drug itself, not discontinuation effects. Alendronate must be taken with a full glass of water (240 mL) and the patient must remain upright for at least 30 minutes to minimize esophageal contact time. The prescribing label for alendronate 70 mg weekly tablets warns of esophagitis, esophageal ulcers, and esophageal erosions, some requiring hospitalization.

When patients stop alendronate because of GI side effects, the upper GI symptoms resolve within days to weeks. There is no GI "rebound" after stopping. The practical issue is that those patients are now unprotected from fracture and need an alternative antiresorptive strategy.

Alternatives After GI Intolerance

• Zoledronic acid 5 mg IV once yearly: eliminates upper GI exposure entirely. A 2007 NEJM trial (HORIZON-PFT, N=7,765) showed zoledronic acid reduced hip fracture risk by 41% and spine fracture risk by 70% over three years compared with placebo. [2]
• Denosumab 60 mg subcutaneous every six months: a RANK-L inhibitor with no GI effects and a different mechanism, though it carries its own discontinuation considerations (rebound vertebral fractures on stopping denosumab are more pronounced than with bisphosphonates).
• Risedronate delayed-release 35 mg weekly taken immediately after breakfast: produces lower esophageal irritation in some patients compared with standard alendronate dosing.

Drug Holidays: Structured Discontinuation vs. Abrupt Stopping

A drug holiday is a planned, supervised break from bisphosphonate therapy intended to reduce the cumulative risk of AFF and ONJ while preserving residual anti-fracture protection from the drug reservoir in bone. This is not the same as simply forgetting to refill the prescription.

ASBMR Drug Holiday Framework

The 2024 ASBMR clinical guidance (endorsed by the Endocrine Society) recommends the following structured approach:

| Patient profile after 5 years oral bisphosphonate | Recommendation | |---|---| | Hip T-score >-2.5, no prior vertebral fracture, low 10-year FRAX risk | Drug holiday of 2-3 years with annual BMD and CTX monitoring | | Hip T-score <-2.5 or prior vertebral fracture | Continue bisphosphonate OR transition to anabolic agent (teriparatide, abaloparatide) | | Any patient on IV zoledronic acid | Consider holiday after 3 infusions; reassess at 6 years |

The Endocrine Society Clinical Practice Guideline on osteoporosis states: "For patients at lower risk of fracture after 5 years of oral bisphosphonate, a drug holiday of 2-3 years is reasonable, with reassessment of fracture risk and bone mineral density after the holiday period." [3]

Monitoring During a Drug Holiday

Patients on a drug holiday are not monitoring-free. Recommended follow-up includes:

• BMD (DXA) at the hip and spine at the start of the holiday and every one to two years during the holiday.
• Serum CTX or P1NP at six months and annually to detect resorption rebound.
• Restart of bisphosphonate (or switch to an alternative agent) if hip T-score drops below -2.5 or CTX rises above 0.3 ng/mL.

Patients who stop alendronate without a structured plan and without follow-up monitoring account for the majority of post-discontinuation fractures seen in real-world claims databases.

FDA Adverse Event Reporting: What FAERS Shows

The FDA Adverse Event Reporting System (FAERS) contains tens of thousands of spontaneous reports for alendronate since its 1995 approval. A 2019 pharmacovigilance disproportionality analysis of FAERS found that bisphosphonates as a class showed a reporting odds ratio (ROR) of 18.4 (95% CI 16.2-20.9) for ONJ compared with all other drugs in the database, confirming the class-level signal. [4] AFF showed an ROR of 9.1 (95% CI 8.1-10.2) for bisphosphonates versus all comparators.

The FDA has issued multiple drug safety communications about bisphosphonates, including a 2010 AFF communication and a 2011 update expanding the warning to include longer-duration use. These regulatory actions were driven substantially by FAERS signal detection.

What FAERS does not show is a clear signal for a "withdrawal syndrome" in the classical pharmacological sense. Musculoskeletal pain, which can be severe and has an FDA black-box-level warning on bisphosphonate labels, is reported during treatment. Reports of new-onset severe musculoskeletal pain resolving after stopping alendronate appear in FAERS and are consistent with the FDA's 2008 safety update on bisphosphonate-associated musculoskeletal pain. [5]

Musculoskeletal Pain: A Stopping-Responsive Adverse Effect

Severe, incapacitating bone, joint, or muscle pain occurring during alendronate therapy is an underrecognized adverse effect. The FDA issued a safety alert in 2008 after reviewing cases in FAERS where patients experienced pain that resolved days to months after discontinuing bisphosphonate therapy. The FDA alert noted that symptom onset ranged from one day to several years after starting the drug, and that some patients had pain recurrence when the same or a different bisphosphonate was restarted.

This is the closest phenomenon to a "reversible on stopping" adverse event profile that alendronate has. It is not a withdrawal syndrome. It is an adverse drug reaction that resolves after the offending drug is removed.

Clinicians should ask patients reporting new or worsening musculoskeletal pain about timing relative to alendronate initiation. A trial discontinuation of four to eight weeks with reassessment of pain and CTX is a reasonable diagnostic maneuver before attributing the pain to osteoporosis or arthritis.

Esophageal Cancer: A Debated Signal

Several observational studies raised concern about an association between long-term oral bisphosphonate use and esophageal cancer. A 2010 JAMA paper using UK General Practice Research Database data (N=41,826 bisphosphonate users) reported an elevated esophageal cancer rate with five or more prescriptions. [6] However, a subsequent analysis of the same database by different investigators found no statistically significant association. The FDA reviewed available data in 2011 and concluded that the evidence was inconsistent and did not support a causal link, but the label retains a precautionary note.

This is not a discontinuation effect. The debated signal relates to cumulative esophageal mucosal exposure during treatment.

Special Populations: Who Needs Closer Monitoring After Stopping?

Postmenopausal Women With Prior Fracture

Women who sustained a clinical vertebral or hip fracture while on alendronate, or who present for discontinuation counseling with a FRAX 10-year hip fracture probability above 3%, should not take a drug holiday without transitioning to an alternative agent. Endocrine Society 2019 guidelines explicitly state that sequential therapy, starting with an anabolic agent such as teriparatide followed by a bisphosphonate, produces greater BMD gains than bisphosphonate monotherapy in high-risk patients. [7]

Glucocorticoid-Induced Osteoporosis

Patients taking alendronate for glucocorticoid-induced osteoporosis (GIOP) face a different discontinuation calculus. If glucocorticoid therapy continues, stopping alendronate is generally not recommended regardless of duration, because the ongoing glucocorticoid exposure sustains osteoclast activity and bone loss. ACR 2022 GIOP guidelines recommend continuing bisphosphonate therapy for the duration of glucocorticoid use when the cumulative dose exceeds prednisone 2.5 mg/day for three or more months. [8]

Premenopausal Women

Premenopausal women prescribed alendronate for GIOP or other secondary osteoporosis causes are not candidates for drug holidays as defined by the postmenopausal literature. The FLEX data do not apply. Discontinuation decisions in this group must be individualized.