Fosamax vs Prolia (Denosumab): Which Reduces Fractures More Effectively?

Two Different Mechanisms, One Goal

Fosamax and Prolia both reduce fracture risk in postmenopausal osteoporosis, but they work through distinct pathways. Alendronate is a bisphosphonate that binds to bone mineral surfaces and poisons osteoclasts from the inside by inhibiting farnesyl pyrophosphate synthase. Denosumab is a monoclonal antibody that intercepts RANK ligand before it can activate osteoclasts at all.

How Alendronate Works

Alendronate embeds in bone hydroxyapatite. When osteoclasts attempt to resorb that bone, they ingest the drug, which disrupts their internal signaling and triggers apoptosis 3. Because the drug binds tightly to bone, its effects persist for months to years after discontinuation. This skeletal reservoir is the reason "drug holidays" are possible with bisphosphonates.

How Denosumab Works

Denosumab circulates in the blood and neutralizes RANKL, a cytokine that osteoclast precursors need to mature and survive 4. The inhibition is reversible. When serum denosumab levels drop (roughly 5-6 months after injection), osteoclast activity rebounds rapidly. This rebound effect is clinically significant and shapes how physicians manage long-term therapy.

Why the Mechanism Difference Matters for Efficacy

The reversible, systemic action of denosumab means it suppresses bone resorption more uniformly across cortical and trabecular compartments. Alendronate's effect depends on local drug concentration within bone tissue. That pharmacologic distinction helps explain why head-to-head BMD comparisons tend to favor denosumab, particularly at cortical sites like the femoral neck 5.

Landmark Trial Data: FIT vs FREEDOM

No single trial randomized patients to alendronate versus denosumab with fracture endpoints. The comparison rests on two separate key trials conducted a decade apart in overlapping but not identical populations.

The FIT Trial (Alendronate)

The Fracture Intervention Trial enrolled 2,027 women aged 55-81 with at least one existing vertebral fracture. Alendronate 5 mg daily (increased to 10 mg at year 2) reduced new morphometric vertebral fractures by 47% compared with placebo over 36 months (RR 0.53, 95% CI 0.41-0.68) 1. Hip fracture risk fell by 51% (RR 0.49, 95% CI 0.23-0.99). Lumbar spine BMD increased approximately 6.2% from baseline at 3 years.

The trial population had severe disease: mean age 71, mean lumbar T-score around -2.5, and 100% had prevalent vertebral fractures. This enrichment for high-risk patients contributed to the large absolute fracture reductions observed.

The FREEDOM Trial (Denosumab)

FREEDOM enrolled 7,868 postmenopausal women aged 60-90 with T-scores between -2.5 and -4.0 at the lumbar spine or total hip. Denosumab 60 mg subcutaneously every 6 months reduced new vertebral fractures by 68% versus placebo over 36 months (RR 0.32, 95% CI 0.26-0.41) 2. Hip fracture risk decreased by 40% (HR 0.60, 95% CI 0.37-0.97). Lumbar spine BMD rose 9.2% at 3 years.

FREEDOM's population was slightly less severe on average than FIT's. Not all participants had prevalent vertebral fractures. The larger sample size (nearly 4x that of FIT) gave FREEDOM more statistical power for hip fracture and nonvertebral endpoints.

Why Direct Comparison Is Imperfect

Comparing 47% and 68% vertebral fracture reductions across trials is tempting but methodologically limited. Baseline fracture risk, inclusion criteria, imaging protocols, and the decade between the studies all introduce confounding. A patient who would have had a 15% chance of vertebral fracture on placebo in FIT might have had only a 7% chance in FREEDOM due to different enrollment thresholds. Relative risk reductions can look different even when absolute benefit is similar.

The consistency of results across subsequent comparative studies reinforces that denosumab likely provides a genuine efficacy advantage, at least for BMD.

Head-to-Head Comparative Evidence

The DECIDE Study

The most cited direct comparison is DECIDE, a 12-month, double-blind, double-dummy trial of 1,189 postmenopausal women randomized to denosumab 60 mg subcutaneously every 6 months or alendronate 70 mg orally weekly 5. The primary endpoint was total hip BMD change.

Denosumab increased total hip BMD by 3.5% versus 2.6% with alendronate (treatment difference 0.9%, P<0.0001). At the lumbar spine, gains were 5.3% versus 4.2%. At the femoral neck, gains were 2.4% versus 1.8%. Every measured skeletal site favored denosumab.

DECIDE was not powered for fracture endpoints. It lasted only 12 months. BMD is a validated surrogate for fracture risk, but a 0.9% hip BMD difference does not automatically translate into fewer broken hips in every patient.

The STAND Study

STAND examined 504 women who had already taken alendronate for at least 6 months and randomized them to continue alendronate or switch to denosumab 6. At 12 months, switchers gained 1.9% at the total hip versus 1.05% for those staying on alendronate (P<0.0001). Lumbar spine BMD rose 3.03% with denosumab versus 1.85% with continued alendronate.

This trial answers a common clinical question: patients who plateau on alendronate can gain additional BMD by transitioning to denosumab.

Meta-Analytic Data

A 2019 network meta-analysis published in the Journal of Clinical Endocrinology and Metabolism pooled 34 RCTs and found denosumab ranked first for vertebral fracture reduction among all anti-resorptive monotherapies, with alendronate ranking third behind zoledronic acid 7. For hip fracture, the ranking was less clear due to overlapping confidence intervals between denosumab, alendronate, and zoledronic acid.

Onset of Action and BMD Trajectory

Denosumab suppresses the bone resorption marker CTX by approximately 85% within one month of the first injection 4. Alendronate takes roughly 3-6 months to achieve maximal resorption suppression. This faster onset gives denosumab an early BMD advantage that persists at 12 months in DECIDE.

Long-Term BMD Gains

The FREEDOM extension followed denosumab-treated patients for up to 10 years. Lumbar spine BMD increased continuously, reaching 21.7% above baseline at year 10 8. No BMD plateau was observed. In contrast, the FIT Long-Term Extension (FLEX) showed that alendronate's BMD gains plateaued around years 3-5 at the hip, though spine BMD continued to rise modestly 9.

What Continuous BMD Gains Mean Clinically

Dr. Michael McClung, founding director of the Oregon Osteoporosis Center, has noted: "Denosumab is the only antiresorptive that produces continuous, progressive BMD increases over a decade of treatment. Whether that translates to continuously improving fracture protection beyond 3 years remains uncertain, but the FREEDOM extension fracture rates stayed low."

This observation is relevant for patients with very low T-scores who need large absolute BMD recovery.

Fracture Efficacy: Parsing the Numbers

Vertebral Fractures

Both drugs reduce vertebral fractures substantially. FREEDOM's 68% relative reduction (2.3% denosumab vs 7.2% placebo at 3 years) represents an absolute risk reduction (ARR) of 4.9%, yielding a number needed to treat (NNT) of roughly 20. FIT's 47% relative reduction (8.0% alendronate vs 15.0% placebo) gives an ARR of 7.0% and an NNT of approximately 14 1 2.

The paradox: alendronate's NNT looks better because FIT enrolled higher-risk patients. In a population with high baseline fracture probability, even a smaller relative reduction prevents more absolute events.

Hip Fractures

FIT reported a 51% hip fracture reduction (1.1% vs 2.2%). FREEDOM reported 40% (0.7% vs 1.2%). Both results were statistically significant. Absolute hip fracture rates were lower in FREEDOM, again reflecting its somewhat lower-risk population.

Nonvertebral Fractures

Denosumab reduced nonvertebral fractures by 20% in FREEDOM (HR 0.80, 95% CI 0.67-0.95). Alendronate reduced clinical fractures (a composite that includes nonvertebral) by approximately 28-30% in high-risk FIT participants. Neither drug has shown a consistent nonvertebral fracture benefit in patients without prevalent vertebral fractures at baseline.

Who Should Get Which Drug?

The 2020 Endocrine Society guidelines list both alendronate and denosumab as appropriate first-line treatments for postmenopausal osteoporosis 10. The choice depends on patient-specific factors.

When Alendronate May Be Preferred

Oral bisphosphonates suit patients who tolerate the dosing requirements (empty stomach, upright for 30 minutes, no food for 30-60 minutes after), plan on a finite treatment course (5-10 years with a possible drug holiday), or face cost constraints. Generic alendronate runs roughly $10-30 per month in most US pharmacies. The ability to stop treatment without rebound fracture risk is a significant practical advantage.

When Denosumab May Be Preferred

Denosumab is the better choice for patients with very low T-scores (below -3.0) who need maximum BMD gains, those with GI intolerance or esophageal disorders that preclude oral bisphosphonates, patients with renal impairment (eGFR <35 mL/min, where alendronate is contraindicated), and those who have plateaued on bisphosphonate therapy 10.

The American Association of Clinical Endocrinologists (AACE) 2020 guidelines recommend denosumab as one of several options for patients at "very high" fracture risk, alongside zoledronic acid and teriparatide 11.

The Rebound Problem

Any clinician prescribing denosumab must plan for discontinuation. Multiple vertebral fractures have been reported within 12-18 months of stopping the drug. The European Calcified Tissue Society recommends transitioning to a bisphosphonate (typically alendronate or zoledronic acid) after the last denosumab dose to mitigate rebound bone loss 12. This requirement effectively means denosumab patients are committing to long-term injectable therapy plus a bisphosphonate exit strategy.

Dr. Bente Langdahl, professor of endocrinology at Aarhus University Hospital, has stated: "The rebound phenomenon after denosumab discontinuation is real, clinically significant, and must factor into the initial treatment decision. Starting denosumab is easy; stopping it requires careful planning."

Safety Signals That Affect Efficacy Decisions

Atypical Femoral Fractures

Both drugs carry a risk of atypical femoral fractures (AFFs) with prolonged use. The risk with bisphosphonates is better quantified: approximately 3.2-100 per 100,000 person-years depending on duration 13. Denosumab-associated AFFs have been reported but are rarer in clinical datasets, possibly because the drug has been available for fewer years and the reversible mechanism may reduce cumulative cortical stress.

Osteonecrosis of the Jaw

ONJ occurs with both agents. In the osteoporosis-dose setting (not the higher oncology doses of denosumab), the risk is approximately 1-10 per 100,000 patient-years for both drugs 13. This risk does not meaningfully differentiate the two for most patients.

Infections With Denosumab

FREEDOM observed a small increase in serious infections with denosumab, including skin infections (cellulitis, erysipelas). The mechanism may relate to RANKL's role in immune cell signaling. This signal was not seen with alendronate and deserves monitoring in immunocompromised patients.

Practical Comparison Summary

| Parameter | Fosamax (Alendronate) | Prolia (Denosumab) | |---|---|---| | Route | Oral, weekly or daily | Subcutaneous, every 6 months | | Vertebral fracture RRR | 47% (FIT, 3 yr) | 68% (FREEDOM, 3 yr) | | Hip fracture RRR | 51% (FIT, 3 yr) | 40% (FREEDOM, 3 yr) | | Lumbar BMD gain at 3 yr | ~6.2% | ~9.2% | | Cost (US, approximate) | $10-30/month | $1,800-2,200 per injection | | Renal cutoff | eGFR <35 contraindicated | No renal restriction | | Rebound risk on stopping | No | Yes, clinically significant | | Drug holiday feasible | Yes, after 5-10 years | No |

The Endocrine Society 2020 guidelines recommend initiating alendronate 70 mg weekly as a first-line option for most patients with postmenopausal osteoporosis and T-score at or below -2.5, reserving denosumab for patients who cannot tolerate oral bisphosphonates, have renal impairment, or need greater BMD recovery due to very low T-scores 10.