Denosumab in Patients With Fibrous Dysplasia Previously Treated With Bisphosphonates

Bas C. J. Majoor; Socrates E. Papapoulos; P. D. Sander Dijkstra; Marta Fiocco; Neveen A. T. Hamdy; Natasha M. Appelman-Dijkstra


J Clin Endocrinol Metab. 2019;104(12):6069-6078. 

In This Article


Data from this case series of 12 patients with mostly polyostotic FD/MAS pretreated with bisphosphonates demonstrated that treatment with denosumab is promising and effective in reducing bone turnover, is relatively safe and well tolerated, and although not quantitively measured, probably also effective in reducing pain, thus representing a potential alternative therapeutic option in patients with FD with inadequate responses to bisphosphonates. Administration of 60 mg of denosumab at six-month intervals was not sufficient to sustain the initial beneficial effect on BTMs, whereas this was achieved in all but one patient with the use of 60 mg of denosumab at three-month intervals.

Although there is as yet no approved medical treatment to control pain and disease activity as reflected by increased BTMs in FD/MAS, various regimens of different bisphosphonates are commonly used in the management of patients with FD/MAS-related bone pain and/or high rates of bone turnover. Several open-label studies, mostly using intravenous pamidronate and olpadronate, have demonstrated reduction of bone turnover and bone pain in adults and children with FD.[7,9–12,14,29,30] However, it is clear from these studies that some patients, and especially those with a higher skeletal disease burden, do not respond properly to treatment with these agents, necessitating alternative treatment options. The only randomized controlled trial using daily oral alendronate in six-month cycles for two years has also failed to demonstrate a beneficial effect compared with placebo,[13] although in this study serum ALP activity did not decrease, despite the decrease in CTX, perhaps suggesting that higher doses or shorter interval schedules may have been required to achieve an adequate biochemical response.[12] Thus, despite the success of bisphosphonate treatment reported in open studies, it appears that some are refractory to treatment with these agents, demonstrating persistence of pain and inadequate suppression of bone turnover markers, necessitating alternative treatment options.

The upregulation of RANKL expression in skeletal lesions as well as the increased circulating levels of the cytokine that were positively correlated with SBS in patients with FD[4,31] provide a strong rationale for further exploring the use of the RANKL inhibitor denosumab in the management of severely affected patients with insufficient responses to bisphosphonates. Further support for the potential beneficial effect of this agent is provided by the findings of improvement in radiographic features as well as arrest of development of new lesions in a transgenic mouse model treated with anti-RANKL antibody.[32] In addition, published case reports of five individual patients with FD treated with different schedules and doses of denosumab[15–18] described early clinical improvement in the form of reduction of pain in four of the five patients, combined with dramatic reduction in biochemical markers of bone turnover.[15–18]

The present case series addresses the issue of denosumab dosing in FD and showed that the dose of denosumab as used in the treatment of osteoporosis (60 mg once every six months) is insufficient to maintain the treatment-induced reduction of BTMs in patients with FD. In contrast, the same dose of 60 mg used at a shorter three-month interval led to a decrease in the levels of BTMs by more than 50% of baseline values leading to their normalization in a substantial number of patients, in some of whom bisphosphonates had failed to do so even after long-term use and high cumulative doses. In our previous study of a larger group of patients with FD/MAS treated with bisphosphonates, we found an association of clinical flares with increased bone turnover markers and abatement of pain symptoms associated in the majority of patients with normalization of these markers in the majority of patients however this has not been the experience in other groups.[12]

The relation between baseline and final values of serum ALP activity further indicated that some patients, particularly those with extensive disease, may require higher doses of denosumab to normalize bone turnover as illustrated in the patient who required 120 mg every three months to achieve normalization of bone turnover markers. However, other patients with high SBS (patients 9 to 12) did respond adequately to the 60 mg three-month regimen. Important for clinical practice is the observation that the decrease in biochemical markers of bone turnover reached usually their nadir values within the first six months of treatment with denosumab and, thus, a decision about changing the dose regimen can be made if necessary early in the course of treatment. The pathophysiology of pain in FD is complex and still ill-understood, although several mechanisms have been suggested to play a role in the process.[33] The relation between BTMs and pain remains to date elusive, although in a previous study from our group including a larger group of patients with FD/MAS treated with bisphosphonates, we did find an association of clinical flares with increased BTMs and abatement of pain symptoms associated with decrease or normalization of these markers after treatment with bisphosphonates in the majority of patients.[12] However, this has not been the experience of other groups.

Of clinical relevance in our case series was the reduction of bone pain in 10 of 12 patients treated with denosumab and even disappearance of pain at some point during treatment in half of the patients. Although pain was not formally assessed using a validated instrument, it was systemically evaluated and recorded by the treating physicians (N.A.T.H., N.M.A.-D.) during the entire period of observation. In 4 of the 5 denosumab-treated patients reported in the literature, reduction of pain was observed early after the start of treatment.[15–18] With the three-month treatment regimen, the beneficial effect on pain persisted for the whole observation period. Earlier, pharmacodynamic studies of denosumab have shown dramatic reductions of serum CTX within days after a subcutaneous injection and there is no reason to believe that the response would be different in patients with fibrous dysplasia, however, this remains to be shown.[34,35] Our protocol with blood sampling every three months including for CTX preclude any conclusion about an association between a decrease of bone resorption and the improvement of FD-related pain bone symptoms as missing the maximum early decrease in this marker.

Response to treatment was shown to be irrespective of disease severity as observed by comparing treatment between patients. For example, patient 2 had a low SBS of 0.58 and P1NP levels that were less than two times increased but severe pain, for which she required treatment with paracetamol, NSAIDs and morphine prior to start of denosumab treatment. Patient 12 had the highest SBS (64.4) and a near 14-fold increase in P1NP (812 ng/mL), but only mild FD-related pain symptoms. Denosumab treatment resulted in normalization of BTMs and reduction in pain and improved function in both patients, which allowed patient 2 to discontinue morphine, and only use paracetamol and occasional NSAIDs. Findings from this relatively small case series of adult patients with FDS/MAS of variable skeletal burden suggest, therefore, that a dose of 60 mg administered every three months is associated with a positive clinical and biochemical outcome in the majority of patients irrespective of SBS, and that there is thus so far no rationale for initially using higher doses of denosumab or shorter intervals of administration of this agent in more severely affected patients.

An important question precluded by the design of this study is the optimal duration of denosumab therapy and further management once a clinical and biochemical remission is achieved and it may be deemed reasonable to discontinue treatment. The dilemma arises from the well-established quickly reversible antiresorptive effect of denosumab (in contrast to the long-term antiresorptive effect of bisphosphonates) observed after discontinuation of treatment of osteoporosis, with transient increases in BTMs above pretreatment levels, also described as rebound phenomenon, that are thought to be the result of rapid, synchronous upregulation of osteoclastogenesis as also reported in the treatment of a child with FD, who was pretreated at some stage with intravenous pamidronate.[15,36] This response is intriguing because it was reported recently that patients treated with bisphosphonates followed by denosumab do not generally show a rebound of bone turnover markers after discontinuation of the latter.[37] Data from our case series show an increase of BTMs at six months but not above pretreatment levels in the six patients initially treated at six-month intervals compared with a sustained decrease in those treated at three-month intervals. Levels of BTMs increased rapidly but not above pretreatment values in the patient who discontinued denosumab (patient 1), and there was no evidence for hypercalcemia. This suggests a different response from the previously reported pediatric case who demonstrated a dramatic rebound in BTMs and severe hypercalcemia and is likely to be the result of treatment with bisphosphonates prior to starting treatment with denosumab.[15] However, our patients were only treated for a relatively short period and without assessment of clinical and biochemical responses after ceasing denosumab apart from patient 1. Whereas this may suggest a different response from that observed in the first reported pediatric case who demonstrated a dramatic rebound in BTMs exceeding pretreatment levels and severe hypercalcemia after discontinuation of denosumab, other factors should be taken in consideration in pediatric cases, such as the higher bone remodeling activity in children and the higher monthly frequency and possible cumulative dose administered over seven months in the child reported.[15] It is also of note that in our case series all patients had received long-term treatment with bisphosphonates prior to starting treatment with denosumab, shown to provide some degree of protection against severe rebounds after discontinuation of treatment in patients with osteoporosis.[37] We were not in a position to evaluate the effect of discontinuation of denosumab in our case series, because all but one patient were still receiving denosumab every three months with sustained clinical and biochemical effects at the time of analysis of the data.

Until additional data become available from randomized controlled trials, caution should be exercised in the treatment of patients with FD/MAS with denosumab. If such treatment is deemed necessary, it should be administered by physicians with expertise in the management of patients with metabolic bone disorders. A point of caution with the use of antiresorptive agents in patients with FD, which equally applies to the use of bisphosphonates or denosumab, is that attention should be given to correction of any vitamin D deficiency and/or FGF-23–related hypophosphatemia that may potentially aggravate the underlying osteomalacia that is a feature of FD/MAS, before starting antiresorptive therapy.

Although data from this case series are promising, a number of safety issues still need to be addressed before this potent antiresorptive could be advocated for widespread use in FD/MAS Until additional data become available from randomized controlled trials, the use of denosumab is thus not recommended outside specialist centers with expertise in the management of patients with metabolic bone disorders, preferably in the context of clinical trials.

We believe that data presented in this case series, together with findings from published case reports provide sufficient background and rationale for the design of a randomized controlled study on the efficacy and tolerability of denosumab in patients with FD/MAS refractory or not tolerating treatment with bisphosphonates and for whom no alternative therapeutic option is currently available.