Body weight predicts denosumab pharmacodynamic duration: a real-world longitudinal cohort study.

Abstract

Denosumab is administered as a fixed 60-mg subcutaneous injection every 6 months, regardless of body weight. In a retrospective cohort of 2123 patients with osteoporosis, the GAM-predicted duration of CTX reduction after each injection was about two months shorter at 80 kg than at 40 kg, but still exceeded the 6-month interval across the evaluated weight range. These findings support body-weight-aware CTX monitoring in heavier patients, not a change to the approved dosing interval. Denosumab is administered as a fixed 60-mg subcutaneous injection every 6 months, but the duration of bone turnover reduction may vary with body weight. We investigated whether body weight affects the duration of bone turnover marker reduction and whether this association could inform body-weight-aware monitoring. We retrospectively analyzed data from 2123 adults with osteoporosis treated with denosumab between 2020 and 2025. Serial serum measurements of beta-isomerized C-terminal telopeptide of type I collagen (CTX) and total procollagen type I N-terminal propeptide (P1NP) were modeled using generalized additive models and random forests to estimate the time from injection until prespecified operational thresholds (CTX > 0.28 ng/mL; P1NP > 35 ng/mL) were crossed. Predicted threshold-crossing times were summarized by body weight. The cohort contributed 4257 CTX and 4477 P1NP measurements. For CTX, the generalized additive model predicted threshold-crossing times that decreased from 262 days (8.7 months; 95% CI: 236-296) at 40 kg to 196 days (6.5 months; 95% CI: 180-238) at 80 kg, corresponding to a slope of - 1.60 days/kg (95% CI: - 2.03 to - 1.17). Random forest predictions were directionally consistent with those from the generalized additive model (Pearson r = 0.917 for CTX). For P1NP, the association followed an inverted U-shaped pattern, with the longest predicted interval at intermediate body weights; P1NP was therefore less informative for interval estimation. Sensitivity analyses using alternative thresholds, one observation per patient per dosing cycle, and adjustment for parathyroid hormone (PTH) all preserved the weight-stratified gradient. Body weight is an independent determinant of model-predicted denosumab pharmacodynamic duration. These findings, based on surrogate markers, support body-weight-aware CTX monitoring, particularly in heavier patients. They do not justify changing the approved 6-month dosing interval. Prospective studies incorporating fracture endpoints are needed before clinical implementation.