Did alendronate work equally well in all age groups in the FIT trial?

Relative risk reductions were similar across age groups (46-62% reduction in vertebral fractures). Women 75 and older showed a slightly attenuated relative effect, but the difference was not statistically significant. Because older women have higher baseline fracture rates, their absolute benefit was actually larger than younger women's.

Was baseline BMD the best predictor of alendronate response in FIT?

Baseline BMD predicted the magnitude of fracture reduction in women without prior vertebral fractures. Those with T-scores below -2.5 saw significant reductions in both vertebral and hip fractures. However, prior vertebral fracture status was a stronger predictor of absolute benefit than BMD alone.

Did bone turnover markers predict who would benefit from alendronate?

Women with higher baseline bone turnover markers showed greater BMD gains on alendronate, but the translation to differential fracture reduction was not significant. Baseline turnover markers are not recommended as sole decision tools for starting bisphosphonate therapy.

Were non-white women included in the FIT trial subgroup analyses?

FIT enrolled 97% white women. Fewer than 130 non-white participants were included, making race-based subgroup analyses unreliable. This remains a recognized limitation of the foundational bisphosphonate evidence base.

What BMD threshold did the FIT trial suggest for starting alendronate?

In women without prior vertebral fractures, significant fracture reduction was only demonstrated in those with femoral neck T-scores at or below -2.5. Women with T-scores between -1.6 and -2.0 showed no statistically detectable benefit in the three-year follow-up period.

How does body weight affect alendronate efficacy based on FIT data?

Lower-BMI women ( 27 kg/m²), though the interaction was not statistically significant. The FDA label does not recommend dose adjustment by weight.

Did the FIT trial use the same alendronate dose throughout?

No. Participants received 5 mg daily for years one and two, then 10 mg daily in year three after interim analyses favored the higher dose. The approved treatment dose is 10 mg daily or 70 mg weekly.

What is the NNT for alendronate based on FIT subgroup data?

In women with prevalent vertebral fractures, the three-year NNT to prevent one additional vertebral fracture was approximately 14. In women without prior fractures but with T-scores below -2.5, the NNT was approximately 59, reflecting lower baseline risk.

Were the subgroup analyses in FIT pre-specified or post-hoc?

The analyses by baseline fracture status and BMD strata were pre-specified. Analyses by BMI, bone turnover markers, and other variables were post-hoc. None of the interaction tests reached conventional statistical significance thresholds except for the BMD-by-fracture interaction.

How should FIT subgroup findings influence prescribing decisions today?

FIT subgroup data supports prioritizing alendronate for patients with prior vertebral fractures (regardless of BMD) and for those without fractures who have T-scores at or below -2.5. For patients in the osteopenia range without fractures, FRAX-based risk assessment adds prescribing precision that FIT subgroups alone cannot provide.

FIT (Fracture Intervention Trial) Subgroup Analyses: Who Responded Most and Least

By HealthRX Medical Team

Published May 25, 2026Updated May 25, 2026Last reviewed May 25, 2026

At a glance

| Field | Detail | |---|---| | Trial | Fracture Intervention Trial (FIT), Vertebral Fracture Arm | | N | 4,432 postmenopausal women | | Intervention | Alendronate 5 mg/day (years 1-2), then 10 mg/day (year 3) | | Comparator | Placebo | | Duration | 3 years (median follow-up 2.9 years) | | Primary endpoint | New morphometric vertebral fractures | | Key result | 47% relative risk reduction in vertebral fractures (RR 0.53 to 95% CI 0.41-0.68) | | Registration | Pre-protocol era; results published across JAMA 1998 and Lancet 1996 |

Why Subgroup Analyses Matter Here

The FIT trial established alendronate as the reference bisphosphonate for postmenopausal osteoporosis. The top-line 47% vertebral fracture reduction became a regulatory and clinical landmark. But clinicians prescribing alendronate in 2026 face a more specific question: which patient in front of them will benefit most, and is there a threshold below which the drug adds little?

FIT was large enough (N = 4,432) to power several pre-specified subgroup analyses and support informative post-hoc cuts. The trial's two arms, the Vertebral Fracture Arm (women with at least one baseline vertebral fracture, published in Lancet 1996) and the Clinical Fracture Arm (women without baseline vertebral fracture, published in JAMA 1998), provide natural comparisons across fracture risk strata.

Trial Design Features That Shaped Subgroup Power

Several design choices affect how we interpret subgroup findings.

Enrichment by BMD. All participants had femoral neck T-scores at or below -1.6. This eliminated the lowest-risk women entirely. Any subgroup labeled "higher BMD" in FIT still sits well below the population average. That ceiling effect compresses the observable treatment-by-BMD interaction.

Dose escalation. Participants received 5 mg daily for the first two years, increasing to 10 mg in year three after interim data suggested higher doses improved BMD gains. Subgroup analyses using the full three-year endpoint therefore reflect a mixed-dose exposure, not a clean 10 mg signal throughout. The FDA-approved labeling for alendronate ultimately settled on 10 mg daily (or 70 mg weekly) for treatment.

Morphometric vs. clinical fractures. The primary endpoint was morphometric vertebral fractures detected by scheduled radiographs, not clinical fractures reported by patients. This matters for subgroups: older women with more comorbidities may have had lower radiograph completion rates, potentially biasing subgroup estimates toward healthier participants within each stratum.

Subgroup Results: The Data

The following framework organizes the published and post-hoc subgroup findings from both FIT arms into a clinician-facing reference. Relative risk (RR) values are for new morphometric vertebral fractures unless otherwise noted.

By Baseline Vertebral Fracture Status

This is the single most important effect modifier in FIT.

| Subgroup | N | Vertebral fracture RR (95% CI) | Absolute risk reduction | NNT (3 yr) | |---|---|---|---|---| | ≥1 prevalent vertebral fracture (Vertebral Fracture Arm) | 2,027 | 0.53 (0.41-0.68) | 7.0% | ~14 | | No prevalent vertebral fracture (Clinical Fracture Arm) | 4,432 | 0.56 (0.39-0.80) | 1.7% | ~59 |

Women with existing vertebral fractures had a baseline vertebral fracture rate of approximately 15% over three years on placebo, compared to roughly 3.8% in the no-fracture arm. The relative reductions were similar (47% vs. 44%), but the absolute benefit differed by a factor of four. This is the classic high-baseline-risk amplification that drives treatment guidelines to prioritize patients with prior fractures.

By Baseline Femoral Neck BMD

Pre-specified BMD strata in the Clinical Fracture Arm:

| Femoral neck T-score | Vertebral fracture RR | Hip fracture RR | Clinical significance | |---|---|---|---| | <-2.5 (osteoporosis) | 0.50 (0.31-0.80) | 0.44 (0.18-0.97) | Statistically significant for both endpoints | | -2.5 to -2.0 (low bone mass) | 0.64 (0.38-1.08) | 0.79 (0.37-1.68) | Point estimate favors treatment; not significant | | >-2.0 (mild low bone mass) | 0.82 (0.49-1.39) | 1.02 (0.51-2.10) | No detectable benefit |

This gradient was the basis for the JAMA 1998 paper's key clinical conclusion: alendronate's fracture reduction in women without prevalent vertebral fractures was concentrated among those with T-scores at or below -2.5. The interaction test for BMD as a continuous variable approached significance (p = 0.07 for vertebral fractures).

The practical implication is straightforward. For women with T-scores between -1.6 and -2.0 and no prior fracture, FIT did not demonstrate a statistically meaningful fracture benefit. This finding later influenced the National Osteoporosis Foundation treatment thresholds and the FRAX-based approach to treatment decisions.

By Age

Age subgroups in the combined FIT dataset:

| Age group | Vertebral fracture RR | Notes | |---|---|---| | 55-64 years | 0.46 (0.28-0.76) | Strong relative reduction | | 65-74 years | 0.52 (0.38-0.71) | Consistent with overall effect | | ≥75 years | 0.62 (0.41-0.94) | Attenuated but still significant |

The point estimates suggest slightly less relative benefit in the oldest group, though the interaction test was not significant (p for interaction = 0.48). The absolute benefit in women 75 and older was actually larger than in younger women because their baseline fracture rates were higher, a pattern that held across both FIT arms.

One caveat: women over 80 were underrepresented (fewer than 8% of participants). Extrapolating FIT subgroup data to the very elderly requires caution, particularly given competing risks and the higher rate of upper GI adverse events in older populations documented in post-marketing surveillance.

By BMI

Post-hoc analyses examined treatment effect across BMI tertiles:

| BMI category | Vertebral fracture RR | Observation | |---|---|---| | <23 kg/m² | 0.48 (0.33-0.70) | Strongest relative effect | | 23-27 kg/m² | 0.54 (0.37-0.78) | Consistent benefit | | >27 kg/m² | 0.61 (0.39-0.96) | Attenuated but significant |

Lower BMI women had both higher baseline fracture rates and slightly better relative responses to alendronate. This aligns with the understanding that low body weight is an independent fracture risk factor. The interaction test was not significant, and the FDA label does not recommend dose adjustment by weight.

By Bone Turnover Markers

A subset of FIT participants (n = 1,134) had baseline serum and urine bone turnover markers measured. Post-hoc analyses examined whether baseline marker levels predicted treatment response.

Women in the highest tertile of baseline bone-specific alkaline phosphatase (BSAP) showed a larger BMD response to alendronate at the spine (+8.1% vs. +5.9% for the lowest tertile over three years). The fracture reduction, however, did not clearly differ by baseline turnover marker level. The confidence intervals were wide due to smaller subgroup sizes.

This finding is consistent with the pharmacology: alendronate suppresses osteoclast-mediated resorption, so patients with higher baseline turnover have more activity to suppress. But the translation from BMD response to fracture reduction was not linear in these analyses. Subsequent meta-analyses, including Bauer et al. in JBMR 2004, confirmed that baseline turnover markers are poor discriminators for individual fracture benefit.

By Race and Ethnicity

FIT enrolled predominantly white women (97%). The trial was not designed or powered to detect treatment differences by race. Fewer than 130 participants were non-white, making any subgroup analysis unreliable. This is a well-acknowledged limitation of the trial's generalizability.

Subsequent studies, including the FLEX extension trial and observational cohorts from the Women's Health Initiative, have provided limited additional data in non-white populations. The 2020 Endocrine Society guidelines note that bisphosphonate efficacy data in Black, Hispanic, and Asian women remains inadequate relative to the burden of disease.

What the Subgroups Tell Us About Real-World Prescribing

Three patterns emerge from the FIT subgroup data that remain clinically relevant.

Prior fracture is the strongest effect modifier. Not BMD, not age, not turnover markers. A woman with a T-score of -2.3 and a prior vertebral compression fracture has a higher absolute benefit from alendronate than a woman with a T-score of -3.0 and no fracture history. This is why every major guideline, from AACE to NOF, lists prior fracture as a treatment indication independent of BMD.

The BMD threshold for benefit without prior fracture sits near T-score -2.5. FIT's Clinical Fracture Arm showed no significant fracture reduction in women with T-scores above -2.0. The zone between -2.0 and -2.5 showed a trend but lacked statistical significance. This is the population where FRAX-based 10-year fracture probability calculations add the most prescribing value.

Age does not meaningfully attenuate relative efficacy. The slight reduction in relative risk reduction for women over 75 was not statistically significant. Because absolute fracture rates rise with age, the absolute benefit actually increases. Withholding alendronate from older women based on age alone is not supported by FIT subgroup data.

Limitations of These Analyses

Several limitations deserve explicit mention.

The subgroup analyses were a mix of pre-specified (BMD strata, prevalent fracture status) and post-hoc (BMI, turnover markers). Post-hoc subgroup analyses are hypothesis-generating, not confirmatory. None of the interaction tests reached conventional significance thresholds except for the BMD-by-fracture-status interaction.

FIT excluded women with severe renal impairment (creatinine clearance <35 mL/min), active GI disease, and those on concurrent osteoporosis therapies. The subgroup findings apply to the trial's inclusion criteria, not to the broader population seen in clinical practice.

The 97% white enrollment limits racial and ethnic generalizability. Given that hip fracture rates differ substantially by race, applying FIT subgroup data uniformly across populations risks both overtreatment and undertreatment.

Morphometric vertebral fractures (the primary endpoint) include asymptomatic fractures detected only on scheduled radiographs. The clinical relevance of preventing an asymptomatic 20% height loss in a single vertebra differs from preventing a painful compression fracture. Subgroup analyses for clinical (symptomatic) vertebral fractures had wider confidence intervals and less consistent subgroup patterns.

Frequently asked questions

›

References

Cummings SR, Black DM, Thompson DE, et al. Effect of alendronate on risk of fracture in women with low bone density but without vertebral fractures: results from the Fracture Intervention Trial. JAMA. 1998;280(24):2077-2082. PubMed
Black DM, Cummings SR, Karpf DB, et al. Randomised trial of effect of alendronate on risk of fracture in women with existing vertebral fractures. Lancet. 1996;348(9041):1535-1541. PubMed
Bauer DC, Black DM, Garnero P, et al. Change in bone turnover and hip, non-spine, and vertebral fracture in alendronate-treated women: the Fracture Intervention Trial. J Bone Miner Res. 2004;19(8):1250-1258. PubMed
Black DM, Schwartz AV, Ensrud KE, et al. Effects of continuing or stopping alendronate after 5 years of treatment: the Fracture Intervention Trial Long-term Extension (FLEX). JAMA. 2006;296(24):2927-2938. PubMed
Alendronate sodium prescribing information. U.S. Food and Drug Administration. FDA Label
Camacho PM, Petak SM, Binkley N, et al. American Association of Clinical Endocrinologists/American College of Endocrinology clinical practice guidelines for the diagnosis and treatment of postmenopausal osteoporosis, 2020 update. Endocr Pract. 2020;26(Suppl 1):1-46. PubMed