Follow-up of Giant Cell Tumors of Bone with Computed Tomography and Dynamic Contrast-enhanced Magnetic Resonance Imaging during Denosumab Treatment

Purpose or Learning Objective: (1) To determine the value of computed tomography (CT) and dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) for monitoring denosumab therapy of giant cell tumors of bone (GCTB) by correlating imaging with pathology; and (2) to optimize denosumab therapy dosage, duration, and interval.

Methods or Background: Patients with GCTB receiving denosumab treatment were monitored with CT and DCE-MRI every 6 months and retrospectively included (2012–2022). Qualitative CT and DCE-MRI imaging and semiquantitative measurements were tracked over time to assess response, regarding lesion volume, density, morphology, and vascularization. Correlation of denosumab effect with the cumulative dose administered was performed. In six patients, histopathologic slides were stained with CD34 (vascularization)/H3.3G34W (neoplastic cells) and analyzed using computerized segmentation. Radiology versus pathology correlations were evaluated (Pearson's correlation/Spearman's rank/Kruskal-Wallis).

Results or Findings: Eleven patients (mean: 34 years of age; seven men) with 69 CT and 125 DCE-MRI studies were evaluated. On CT, good responders showed progressive reossification (+6 HU/month), cortical remodeling, and volume reduction (− 0.13 cm³/month), correlating with woven bone. A significant correlation was observed between the change in density/month/120 mg denosumab and the cumulative dose already administered (ρ = − 0.54/P = 0.001; r = − 0.34/P = 0.40).

Mean signal intensity (SI) on T1-weighted images increased in the first 1.5 years of treatment and decreased afterward. SI on fat-saturated T2-weighted images continuously decreased with the strongest decrease at 0.5 to 1.5 years of treatment (− 0.07 a.u./month). SI on fat-saturated contrast-enhanced T1-weighted images also decreased, with the strongest decrease in the first 0.5 years (− 0.07 a.u./month).

The time-intensity curve (TIC) evolved from a type IV with high first pass/amplitude and steep washout to a type V with high first pass and continuous wash-in in the first half year, and finally to a slow type II. Fewer giant cells (P = 0.04), neoplastic cells (P = 0.03), and blood vessels (P = 0.42) were observed with a TIC type II.

In the first 0.5 years, a significant decrease in wash-in, positive-enhancement-integral, area under time intensity curve, K^trans, K_ep, area under time concentration curve, and a significant increase in time to peak and washout was observed. Afterward, the parameters continued the same trends at a slower rate and stabilized, despite increasing dose interval to 2 to 3 months.

Two patients relapsed when denosumab was stopped for pregnancy/dental surgery with inverse imaging and pathologic findings: shorter time to peak and increased microvessel density (ρ = − 0.857/P = 0.014; r = − 0.827/P = 0.022).

Conclusion: CT and DCE-MRI show good correlation with pathology and allow adequate evaluation of (failed) response to denosumab and tumor recurrence. The first 0.5 years of induction therapy with 120 mg/month are most important and show the most prominent response. After 1.5 years of good response on induction therapy, 120 mg every 2 to 3 months as maintenance therapy seemed sufficient for continued tumoral control.

Publication History

Article published online:
26 May 2023