Real-world rebound pattern of serum collagen type I C-telopeptide after denosumab discontinuation and zoledronate rescue in postmenopausal osteoporosis and cancer treatment-induced bone loss.

Stopping denosumab causes a temporary rebound in bone resorption, even when zoledronate is administered afterwards. We analyzed 65 women with postmenopausal osteoporosis or cancer treatment-induced bone loss who received a single zoledronate infusion 6 months after their last denosumab dose. Blood levels of the bone turnover marker CTX were modeled over time, showing a nonlinear increase peaking between 6 and 9 months and partially declining by 12 months. The rebound pattern was comparable in both groups, and only one vertebral fracture occurred. These findings show that the rebound is only partially mitigated by zoledronate.

backgroundDenosumab (Dmab) discontinuation triggers a rebound in bone resorption that may be only partially mitigated by zoledronate (ZOL). However, the temporal pattern of this rebound under real-world conditions remains poorly defined. We aimed to model continuous serum C-terminal telopeptide of type I collagen (s-CTX-I) trajectories after Dmab withdrawal and ZOL rescue and to compare rebound profiles between postmenopausal osteoporosis (PMO) and cancer treatment-induced bone loss (CTIBL).

methodsSixty-five women who received a single 5-mg ZOL infusion 6 months after Dmab discontinuation were retrospectively analyzed (30 PMO, 35 CTIBL). s-CTX-I measurements obtained from -30 to +360 days relative to ZOL were modeled with linear mixed-effects regression using natural cubic splines (knots 0, 180 days), adjusting for age and Dmab duration.

resultss-CTX-I increased nonlinearly, peaking at 6-9 months post-ZOL and reaching a partial plateau by 12 months. The spline model fits the data substantially better than a linear specification (ΔAIC > 20). No interaction was found between time and indication (PMO vs CTIBL; p > 0.3), indicating comparable rebound trajectories across groups. Peak s-CTX-I values remained below the upper reference limit for premenopausal women. Rebound intensity (Δs-CTX-I) was not associated with percentage bone mineral density (BMD) gain at the lumbar spine, femoral neck, or total hip during Dmab therapy. Only one vertebral fracture occurred during follow-up.

conclusionsIn real-world practice, bone resorption rebounds after Dmab discontinuation in a reproducible, nonlinear pattern despite ZOL rescue, peaking around 6-9 months. Although peak s-CTX-I values remained within the premenopausal reference range, rebound magnitude was unrelated to prior densitometric response, supporting a biologically driven reactivation of remodeling. The similarity between PMO and CTIBL suggests a shared biological mechanism and supports individualized, turnover-guided monitoring rather than fixed-timing strategies.