CORR Insights®: Is Treatment with Denosumab Associated with Local Recurrence in Patients with Giant Cell Tumor of Bone Treated with Curettage? A Systematic Review

Where Are We Now?

Giant cell tumor of bone is one of only a handful of primary benign bone tumors that may behave aggressively. Local recurrence of giant cell tumor of bone increases the risk of further pain, pathologic fracture, metastases, and even malignant degeneration [9]. Hence, local control of a giant cell tumor is important in order to minimize associated morbidity.

Until 2013, the primary treatment to achieve local control of giant cell tumors of bone was extended intralesional curettage with an adjuvant and either cementing or bone grafting [5]. Adjuvant treatment options before and since then have included bone cement, phenol, ethanol, argon beam, hydrogen peroxide, and liquid nitrogen [5]. Other alternatives still used in selected patients include en bloc resection in expendable bone or, in case of extensive bone destruction, radiotherapy for tumors deemed to be inoperable, as well as embolization, although more often as an adjuvant approach.

The FDA originally approved denosumab as a RANK-L inhibitor to treat postmenopausal women with osteoporosis who are at a high risk of having a fracture [3, 4]. In 2013, denosumab was approved for the treatment of giant cell tumor of bone in unresectable tumors and those in which resection would result in “considerable morbidity” [1, 2]. Approval was based on two single-arm clinical trials with 304 adult or skeletally mature adolescents who received at least one dose.

The FDA-approved giant cell tumor of bone dosing is 120 mg every 4 weeks with additional doses of 120 mg subcutaneously on Days 8 and 15 of the first month [1]. The initial impression was that denosumab might replace surgery as the standard treatment of giant cell tumor of bone. However, questions remain regarding the appropriate use of denosumab in giant cell tumors of bone. What mode of treatment (preoperative, postoperative, both, or in lieu of surgery) is best? What is the best treatment duration? Should the duration differ according to the mode of treatment? What is its efficacy in diminishing local control and/or metastases? Despite the passing of 7 years since FDA approval, these questions linger. In large part, as illustrated nicely by their comprehensive review of previous studies, Tsukamoto et al. [10] showed that there are no randomized controlled clinical trials, and available studies suffer from selection bias and are inadequate to answer any of these questions definitively. These findings underscore where we’re at but not where we need to go or how to get there.

Where Do We Need To Go?

The answers to the questions of the best denosumab therapy duration, efficacy of local and systemic control, and long-term effects may differ completely based on the treatment setting (both the treatment mode and tumor location). Some giant cell tumor of bone locations, such as the sacrum, have high morbidity with operative intervention, and for these, denosumab may be effective to chronically suppress the disease in lieu of surgery. However, more information is still needed as to the duration, efficacy, and long-term risks. In expendable bone, such as the proximal fibula, distal ulna, iliac wing, or rib, where resection of bone is highly likely to provide permanent local control of the tumor, there is no role for denosumab (nor is it FDA-approved for that indication). Between these extremes, the precise role of denosumab in giant cell tumor of bone remains largely a mystery—should it play more of an adjuvant role in tumors of the proximal femur and distal radius, where local recurrence is more frequent and associated with distant spread [5, 7, 11]? What, if any, is the role of denosumab in giant cell tumors of bone of the most common long-bone locations—distal femur, proximal tibia, and proximal humerus—which have historically been treated with extended intralesional curettage and a local adjuvant, with a modern local recurrence rate of 20% to 34% [7, 9]? Denosumab is not a panacea, and we need to decide what subpopulations of patients with giant cell tumors of bone will potentially benefit the most from a deviation in the standard of care. Further, if there is any potential for an increased risk of local recurrence with denosumab (as suggested but not proven in the current paper [10]) or other complications, a study of denosumab in clinical trials should be considered cautiously. It should first be studied in instances where extended intralesional curettage leads to higher recurrence rates (for example, in the proximal femur and distal radius) or where en bloc resection results in high morbidity (for example, in the sacrum and spine) [6, 13]. In other words, we should step back from the perspective that denosumab has a role in every giant cell tumor of bone and consider more-focused Phase II studies, isolating cases in which the potential benefit-to-risk ratio is the highest.

How Do We Get There?

I believe the most promising next step would be to conduct Phase II randomized discontinuation clinical trials to evaluate the ideal duration [14]. In randomized discontinuation clinical trials, all patients would be treated with denosumab during a run-in period. Thereafter, patients who respond to treatment would be randomly assigned to continue denosumab or switch to a placebo. Poor responders identified during the run-in phase would be excluded from randomization [14]. These could be applied in giant cell tumor of bone subgroups such as the sacrum and spine, where the potential morbidity of surgery is great, or in the proximal femur and distal radius, where local recurrence is higher. For other locations where the standard of care (extended intralesional curettage with a local adjuvant) has been successful, if denosumab is going to be studied, it needs to be evaluated in closely monitored studies with frequent data monitoring and appropriate standard of care controls. Of course, the problem with this approach is the diminished statistical power to show differences between treatments. However, we can see in the current article [10] what happened in the seven included studies. Selection bias, with the tendency to use denosumab only for giant cell tumors of bone that are the most difficult to treat surgically, was the result.

The best way denosumab studies could be accomplished is through a multi-institutional, independent, standardized evaluation. The currently available orthopaedic oncology database developed by Miller [12] and the excellent research infrastructure developed by Schneider et al. [8] in the recently completed PARITY study and the ongoing SAFETY study provide hope that denosumab might play a role in the treatment of giant cell tumors of bone.