Can change in FRAX score be used to "treat to target"? A population-based cohort study

It is unknown how responsive the Fracture Risk Assessment (FRAX) tool is to osteoporosis treatment (OTX) or whether it can serve as a target for “goal-directed” treatment. We studied 11,049 untreated women aged ≥ 50 years undergoing baseline and follow-up DXA examinations in Manitoba, Canada. We identified clinical risk factors, intervening OTX based on medication possession ratios (MPR), and incident fractures. FRAX scores for major osteoporotic and hip fractures were computed for each scan using the most current (updated) FRAX inputs. Over 4 years, median FRAX scores showed an increase of 1.1% for major fractures and 0.3% for hip fractures, including women highly adherent to OTX (0.6% and 0.1% increases). Few (2.2%) highly adherent women had a decrease in major fracture probability exceeding 4%, whereas 9.0% had a decrease in hip fracture probability exceeding 1%. Compared with untreated women, OTX was associated with a higher dose-dependent likelihood of attenuating the expected increase in major fracture risk: adjusted odds ratios (aOR) 2.3 (95% confidence interval [CI] 1.8–2.9) for MPR <0.50; 7.3 (95% CI 5.6–9.6) for MPR 0.50–0.79; and 12.0 (95% CI 9.5–15.2) for MPR ≥0.80. In the 4 years after the second DXA scan, 620 (6%) women had major fractures (152 hip fractures). FRAX scores were strongly predictive of incident major fractures (adjusted hazard ratios [aHR] per SD increase in FRAX 1.8, 95% CI 1.7–1.9) and hip fractures (aHR per SD 4.5, 95% CI 3.7–5.7); however, change in FRAX score was not independently associated with major fracture (p = 0.8) or hip fracture (p = 0.3). In conclusion, FRAX scores slowly increased over time, and this increase was attenuated but not prevented by treatment. Few women had meaningful reductions in FRAX scores, and change in FRAX score did not independently predict incident fracture, suggesting that FRAX with BMD is not responsive enough to be used as a target for goal-directed treatment.