This study provides evidence that a switch from the standard 60 mg dose of denosumab to a low dose of 30 mg every 6 months may prevent bone loss and has not been associated with an increased risk of fracture in postmenopausal women at a moderate fracture risk. This dosing regimen has not previously been evaluated in the phase 2 clinical trials.
Cessation of denosumab therapy may be considered following long term therapy in postmenopausal women not at a high fracture risk(11). However it is necessary to initiate other antiresorptive treatment options if denosumab therapy is stopped in order to prevent declines in BMD and an increase in the risk of fracture.
A recent prospective cohort study evaluated the duration of denosumab exposure prior to discontinuing denosumab therapy on the change in BMD following a switch to IV zoledronate and concluded that a duration of greater than 3 years of denosumab was associated with a significant decrease in LS BMD with no significant changes at the FN BMD. (31, 39).
In contrast, in our study, switching from denosumab 60 mg to low dose denosumab 30 mg every 6 months maintained BMD at the hip and radial sites and showed small improvements in BMD at the LS regardless of the duration of previous standard dose denosumab therapy. We are conducting a follow-up study regarding the effectiveness of IV zoledronate in maintaining BMD after 24 months of low dose denosumab therapy. Low dose denosumab may serve as a bridge to transitioning to bisphosphobnate therapy and this requires further evaluation.
Our subgroup analysis suggested similar changes in BMD in women who had been on standard dose denosumab for less than 3 years in comparison to women who had been on standard dose denosumab for 3 or more years. In our study BMD was maintained at all sites independent of prior treatment duration of standard dose denosumab 60 mg every 6 months. Furthermore, it has been described that prior bisphosphonate therapy may attenuate the rebound rise in bone remodeling observed following cessation of denosumab therapy (40). However, in our linear regression analysis prior bisphosphonate use did not impact the effects on BMD following a reduction in the dose of denosumab at any of the skeletal sites.
Older patients in our study had a greater increase in BMD than younger patients. We believe that this may be a reflection of the fact that older individuals may have had a higher rate of bone remodelling and therefore responded more robustly to low dose denosumab therapy.
Although vertebral fractures have been reported as early as 7 months after the last denosumab injection (41, 42), no patients with a moderate fracture risk experienced clinical fragility fractures in our study.
In our study, small rises in ALP were noted in group A (patients on denosumab for less than 3 years prior to the switch). These rises in ALP were not associated with declines in BMD. A statistically significant change was not seen in group B and this may be a reflection of the smaller number of patients enrolled in group B.
Rarely, AFF and ONJ (43, 44) may occur in patients on long term denosumab therapy and concerns of these potential side effects(45) may limit patient acceptance of long term pharmacologic intervention with standard dose denosumab. Offering a low dose option may be more acceptable to patients who have reached treatment targets with standard denosumab therapy and may be effective in preventing further declines in BMD (13).
Denosumab 60 mg every 6 months, suppresses bone turnover until the end of the dosing interval in the majority of patients (46). As denosumab, a potent RANKL inhibitor, prevents differentiation of osteoclast precursor cells into OC, an accumulation of osteoclast precursor cells develops during denosumab therapy, as recently described.(12) These precursor cells differentiate into OC following cessation of denosumab therapy and this may partly explain the reversibility of the effects of denosumab. We hypothesized that low dose of denosumab 30 mg every 6 months, will result in partial suppression of RANKL and would therefore prevent the accumulation of a larger pool of precursor cells.
Recently fission of OC into osteomorphs has been described in the animal model as occurring in the presence of RANKL inhibition.(37) Upon withdrawal of RANKL inhibition, these osteomorphs rapidly fuse to form bone-resorbing OC. Although the process of fission and fusion was shown in an animal model it is possible that a similar process occurs in humans. Bisphosphonates may not be the ideal approach to prevent bone resorption following cessation of denosumab therapy as bisphosphonates do not lower RANKL production and therefore would not prevent the formation of OC from osteomorphs(47, 48). Low dose denosumab however would lower RANKL production and thus would prevent the formation of excess OC from osteomorphs. This could explain why the low dose denosumab was effective in preventing bone loss, wheras bisphosphonates have inconsistently prevented bone loss following denosumab cessation.