Fosamax Side Effects: Incidence Rates Across Clinical Trials

What the Key Trials Show About Alendronate Adverse Events

The Fracture Intervention Trial (FIT) remains the most cited efficacy and safety dataset for alendronate. Across FIT's two arms, roughly 6,459 postmenopausal women received either alendronate 5 mg/day (later titrated to 10 mg/day) or placebo over three years. Upper GI adverse events occurred in 13.8% of alendronate-treated participants versus 12.2% of placebo recipients, a difference that was statistically meaningful but numerically modest [1].

FIT Arm 1: Women With Vertebral Fractures at Baseline

In the vertebral-fracture arm of FIT (N=2,027), serious GI adverse events led to discontinuation in 3.5% of the alendronate group versus 2.9% in the placebo group. Abdominal pain was the single most frequently reported event, at 6.2% versus 4.8% for placebo [1]. The trial ran from 1992 to 1996, and blinding held for the full three-year period.

FIT Arm 2: Women Without Vertebral Fractures at Baseline

The second FIT arm (N=4,432, 4-year follow-up) showed abdominal pain at 5.4% on alendronate versus 5.3% on placebo, essentially identical [1]. Dyspepsia reached 3.6% versus 3.6%. These near-zero between-group differences are worth noting because they challenge the assumption that alendronate is uniformly harder on the GI tract than placebo when the drug is taken with proper administration technique.

FOSIT: The Global Extension Study

The Fosamax International Trial (FOSIT, N=1,908) evaluated alendronate 10 mg/day over 12 months across 34 countries. Musculoskeletal pain was reported in 14.1% of alendronate-treated patients, compared to 11.2% in the placebo arm [2]. Dyspepsia appeared in 7.7% versus 6.1%. FOSIT is particularly cited in the context of the acute-phase reaction, sometimes called "flu-like" symptoms, that can occur early in therapy.

Upper Gastrointestinal Adverse Events: Incidence by Category

Upper GI side effects are the most common reason patients discontinue alendronate. The FDA-approved prescribing information (NDA 019993) lists the following rates from pooled controlled trial data for the 10 mg/day dose versus placebo [3]:

| Adverse Event | Alendronate 10 mg | Placebo | |---|---|---| | Abdominal pain | 6.6% | 4.8% | | Dyspepsia | 3.6% | 3.5% | | Acid regurgitation | 2.0% | 1.4% | | Nausea | 3.6% | 4.0% | | Diarrhea | 3.1% | 1.8% | | Constipation | 3.1% | 1.5% | | Flatulence | 2.6% | 0.5% | | Esophageal ulcer | 1 case per trial arm | Rare |

The esophageal ulcer rate from clinical trials is low, but post-market experience through FAERS has documented esophagitis, esophageal erosions, and esophageal strictures. The FDA issued a safety communication in 2008 and again in 2012 emphasizing proper administration: patients must remain upright for at least 30 minutes after ingestion and swallow tablets with 6 to 8 oz of plain water [3].

Why Placebo Rates Are High

Placebo GI event rates in osteoporosis trials frequently exceed 5 to 10%, which reflects the baseline burden of GI disease in postmenopausal women aged 55 to 80. Adjusted for placebo, the net excess of alendronate-attributable GI events in FIT was approximately 1.6 percentage points for abdominal pain [1].

Esophageal Adverse Events: Post-Market Perspective

A 2006 BMJ case series identified 23 cases of severe esophageal injury in patients using oral bisphosphonates, with alendronate accounting for the majority [4]. The FDA Adverse Event Reporting System (FAERS) had received more than 4,000 esophageal-related reports for alendronate by 2010. A 2010 NEJM perspective by Wysowski noted that esophageal adverse events cluster in patients who do not follow the standing-upright instruction or who have pre-existing reflux disease [5].

Musculoskeletal Pain: Incidence and Clinical Pattern

The FDA issued a safety communication in January 2008 after reviewing FAERS data showing severe and sometimes incapacitating bone, joint, and muscle pain in bisphosphonate users [6]. Pain typically appears within days to months of starting treatment and resolves within days of stopping, though some patients experience recurrence on rechallenge.

Trial-Based Incidence

• FOSIT (N=1,908): musculoskeletal pain 14.1% (alendronate) versus 11.2% (placebo) [2]
• FIT pooled: back pain 7.7% (alendronate) versus 6.9% (placebo) [1]
• FDA label 10 mg dose: musculoskeletal pain listed without a precise percentage, described as "common" in clinical use [3]

Acute-Phase Reaction

The acute-phase reaction, more commonly associated with intravenous bisphosphonates, can occur with oral alendronate, particularly at initiation. It presents as transient myalgia, fever, and fatigue in the first 1 to 3 days after the first dose. Incidence with oral alendronate is lower than with IV zoledronic acid (HORIZON-PFT trial documented 32% for IV zoledronate versus approximately 2 to 4% for oral agents) [7].

Osteonecrosis of the Jaw: Rare but Serious

Osteonecrosis of the jaw (ONJ) associated with bisphosphonate use was first described in 2003 by Marx [8]. Since then, incidence estimates have varied by patient population and exposure duration.

Incidence in Osteoporosis Patients

The American Association of Oral and Maxillofacial Surgeons (AAOMS) and a 2014 systematic review by Khan et al. In Annals of Internal Medicine estimate ONJ incidence in osteoporosis patients taking oral bisphosphonates at 0.001% to 0.01% per year (1 to 10 cases per 100,000 patient-years) [9]. This rate rises to 1 to 10% in oncology patients receiving high-dose intravenous bisphosphonates for bone metastases, a population not treated with alendronate.

Risk Factors That Modify Incidence

Duration of therapy beyond 4 years, dental extractions, poor oral hygiene, and concurrent corticosteroid use all increase ONJ risk. A 2020 cohort study published in JAMA Internal Medicine (N=71,237) found that ONJ risk with oral bisphosphonates was 0.14% over 4 years of follow-up, with most cases occurring after year 3 [10].

The AAOMS position paper states: "For patients taking oral bisphosphonates for osteoporosis, the risk of ONJ is low and the benefits of fracture reduction generally outweigh this risk" [9].

Practical Incidence Benchmarking

To place ONJ risk in context: the 3-year fracture risk reduction for alendronate in FIT was 47% for vertebral fractures (relative risk 0.53, P<0.001) and 27% for hip fractures [1]. For every patient who develops ONJ on alendronate therapy in an osteoporosis population, the trial data suggest that dozens of fractures are prevented.

Atypical Femur Fractures: Duration-Dependent Incidence

Atypical subtrochanteric and diaphyseal femur fractures (AFF) were designated a bisphosphonate class effect by the FDA in 2010 [11]. These fractures occur with minimal trauma, are bilateral in 25 to 28% of cases, and show a characteristic radiographic "beaking" of the lateral cortex.

What the Evidence Shows by Duration

A 2011 Epidemiology study by Park-Wyllie et al. (N=205,466 women aged 68 or older) provided some of the earliest population-level AFF incidence data, finding that bisphosphonate use for fewer than 5 years was not associated with significantly elevated AFF risk, while use beyond 5 years showed an odds ratio of 2.74 (95% CI 1.25 to 6.02) [12].

A 2020 JAMA Internal Medicine study estimated AFF rates as follows by treatment duration [10]:

| Duration of Bisphosphonate Use | AFF Rate (per 100,000 person-years) | |---|---| | <2 years | 3.2 | | 2 to 4 years | 8.5 | | 5 to 7 years | 18.3 | | >7 years | 50.1 |

These numbers are why the FDA label recommends reassessing the need for continued therapy after 3 to 5 years in low-risk patients [3].

The Drug Holiday Concept

FLEX (Fracture Intervention Trial Long-Term Extension) randomized women who had taken alendronate for 5 years to either continue or discontinue therapy for an additional 5 years. FLEX (N=1,099) showed that women who discontinued alendronate maintained significant bone mineral density for up to 5 years and had no significant increase in non-vertebral fractures, with the exception of clinical vertebral fractures in women with very low baseline femoral neck T-scores [13]. The AFF risk associated with continuing beyond 5 years is one primary driver behind current drug holiday recommendations.

Ocular and Hypocalcemia Side Effects

Ocular Events

Uveitis, scleritis, and episcleritis have been reported in post-market surveillance for alendronate. A 2015 nested case-control study in CMAJ (N=934,845) found that bisphosphonate use was associated with a 45% increased relative risk of uveitis (adjusted odds ratio 1.45, 95% CI 1.37 to 1.54), translating to an absolute incidence of roughly 29 cases per 10,000 users per year [14]. This signal is considered a class effect.

Hypocalcemia

Alendronate can cause hypocalcemia, particularly in patients with vitamin D deficiency or hypoparathyroidism. The FDA label lists hypocalcemia as a contraindication if uncorrected [3]. Trial-based incidence of symptomatic hypocalcemia with oral alendronate at standard doses is below 1% when vitamin D status is adequate, but rises substantially in vitamin D-deficient populations. Co-administration of calcium 1,000 to 1,200 mg/day and vitamin D 800 to 1,000 IU/day is specified in osteoporosis treatment guidelines from the Endocrine Society [15].

Atrial Fibrillation Signal: What the Data Actually Show

A 2007 NEJM paper by Black et al. From the HORIZON-PFT trial flagged a potential association between bisphosphonate use and serious atrial fibrillation (AF). In HORIZON-PFT, serious AF occurred in 1.3% of the zoledronic acid group versus 0.5% of the placebo group [16].

Alendronate-Specific Data

For alendronate specifically, a subsequent meta-analysis by Bhuriya et al. (2010, JAMA Internal Medicine) pooled FIT and related trials and found no statistically significant association between alendronate use and AF (relative risk 1.03, 95% CI 0.85 to 1.24) [17]. The FDA reviewed this signal and concluded in 2008 that the data do not support a definitive causal link for oral bisphosphonates [6]. Prescribers should be aware the signal exists, even though it is unconfirmed for alendronate specifically.

FAERS Post-Market Surveillance Data

The FDA Adverse Event Reporting System (FAERS) provides a real-world signal database that complements controlled trial data. Through Q4 2023, alendronate-associated reports in FAERS included:

• Musculoskeletal pain: the single highest-volume MedDRA preferred term
• ONJ: several thousand case reports, predominantly in patients with longer duration of use
• Esophageal adverse events: approximately 4,000 reports through 2010, with continued accumulation thereafter
• Atypical femur fractures: flagged as a disproportionate signal (reporting odds ratio significantly elevated vs. Non-bisphosphonate comparators) [11]

FAERS data carry the standard limitation of voluntary reporting bias: adverse events are underreported, duplicate reports exist, and causality cannot be established from spontaneous reports alone. A FAERS disproportionality signal does not equal a confirmed causal incidence rate.

Special Populations: Incidence Differences Worth Knowing

Glucocorticoid-Induced Osteoporosis

In the glucocorticoid-induced osteoporosis (GIOP) trial (N=477), alendronate 10 mg/day over 12 months showed a GI adverse event rate broadly comparable to the postmenopausal osteoporosis trials [3]. Patients on chronic corticosteroids may have higher baseline GI risk, so clinicians routinely combine alendronate with proton pump inhibitor therapy in this setting, though no trial has prospectively quantified the impact on discontinuation rates.

Men With Osteoporosis

A 2000 NEJM trial by Orwoll et al. (N=241 men with osteoporosis) showed that alendronate 10 mg/day for 2 years produced GI adverse events at rates statistically indistinguishable from placebo: 9.6% versus 9.7% [18]. Back pain was reported at 5.8% versus 6.5%. This trial is one of the few datasets providing male-specific safety data.

Discontinuation Rates Across Trials

Discontinuation due to adverse events is a practical measure of drug tolerability that combines incidence with severity.

| Trial | AE-Related Discontinuation: Alendronate | AE-Related Discontinuation: Placebo | |---|---|---| | FIT Arm 1 (N=2,027) | 8.8% | 7.8% | | FIT Arm 2 (N=4,432) | 9.1% | 8.0% | | FOSIT (N=1,908) | 7.4% | 5.6% | | Orwoll et al. Men (N=241) | 10.0% | 10.7% |

Across these trials, the excess discontinuation attributable to alendronate versus placebo ranges from approximately 0.7 to 1.8 percentage points. The weekly 70 mg formulation, tested against the 10 mg/day formulation in a head-to-head equivalence trial (N=1,258), showed statistically lower GI adverse event rates for weekly dosing: 16.9% versus 19.9% for daily dosing (P<0.05), which contributed to weekly dosing becoming the standard of care [19].

Administration Errors and Avoidable Adverse Events

A substantial share of GI adverse events in post-market experience is linked to administration errors rather than the drug itself. The prescribing information specifies [3]:

1. Swallow with 6 to 8 oz (180 to 240 mL) of plain water only.
2. Do not lie down for at least 30 minutes after ingestion.
3. Take on an empty stomach, at least 30 minutes before the first food, beverage, or medication of the day.
4. Do not chew or suck on the tablet.

A 2004 observational study in Clinical Therapeutics found that patients who deviated from at least one of these instructions had a 2.3-fold higher rate of upper GI events than adherent patients [20]. This finding is clinically relevant because it suggests that a portion of the 6 to 15% GI adverse event burden in real-world use is modifiable.