Sodium Fluoride-18 Uptake Overlaps Radium-223 Dichloride Uptake in Osteoblastic Bone Tumors

1201 Objectives: Prostate cancer commonly metastasizes to bone and becomes osteoblastic. Radium-223 dichloride (223RaCl2) alpha therapy has been used to treat patients with metastatic prostate cancer by targeting hydroxyapatite in calcified bone lesions. Therapeutic responses of patients receiving 223RaCl2 have varied and, in some patients previously treated with standard of care therapies, an increase in bone fractures have occurred suggesting greater bone toxicity from these therapies when administered in combination. This variability in therapeutic response may be related to 223RaCl2 dosing, which is adjusted by patient body weight as opposed to total tumor burden and responsiveness of individual lesions. Thus, the bone-seeking PET imaging agent sodium fluoride-18, [18F]NaF, may be useful to select patients for 223RaCl2 therapy and determine appropriate 223RaCl2 doses per patient. Although [18F]NaF PET/CT has shown clinical utility in identifying and monitoring metastatic bone disease, the relationship of [18F]NaF uptake in bone lesions for predicting the uptake of bone-targeted therapies such as 223RaCl2 has not been established. Herein, the in vivo biodistribution of [18F]NaF and 223RaCl2 were evaluated in human osteosarcoma tumor xenograft mouse models (Hu09-H3), and then ex vivo autoradiography and confirmatory histology were performed with the tumors to determine whether the spatial distribution of [18F]NaF and 223RaCl2 coincided with calcified osteoblastic tissue. Methods: In vivo biodistributions were performed using Hu09-H3 xenografts at 1 hour post single or co-injection of [18F]NaF and 223RaCl2 from which percent injected doses per gram of tissue (% ID/g) were determined. [18F]NaF PET/CT and 223RaCl2 SPECT/CT imaging were performed at similar times using the same tumor model. Following biodistribution, tumors were further evaluated by ex vivo autoradiography to determine the regional distribution of [18F]NaF and 223RaCl2 within the same tumor sections (223RaCl2 determined after [18F]NaF decay), which were then validated by histological staining for calcifications (Von Kossa, Alizarin Red). Results: In co-injected Hu09-H3 xenografts, the highest [18F]NaF uptake occurred in target tissues, bone (28.7% ID/g) and tumor (9.7% ID/g) (Table 1). In contrast, 223RaCl2 had 1.5-fold lower target tissue uptake in bone (18.5% ID/g) and tumor (6.3% ID/g) with the highest uptakes occurring in non-target tissues, spleen (40.0% ID/g) and kidney (33.6% ID/g) (Table 1). [18F]NaF uptakes were lower in the blood and other non-target tissues compared to 223RaCl2 indicating faster clearance. The co-injected [18F]NaF and 223RaCl2 biodistribution results were comparable to biodistributions in Hu09-H3 xenograft cohorts injected with the single agents ([18F]NaF only or 223RaCl2 only), indicating the presence of both agents did not alter targeting. The high non-targeted 223RaCl2 uptake in the spleen, kidneys, and intestines are consistent with previous reports [1, 2]. [18F]NaF PET/CT and 223RaCl2 SPECT/CT imaging showed uptake in discrete regions within the tumor and throughout the skeleton with additional 223RaCl2 uptake in the kidneys and spleen, which compared favorably with the biodistribution results. Ex vivo autoradiograms of tumors from co-injected Hu09-H3 xenografts demonstrated [18F]NaF and 223RaCl2 were spatially congruent and selective for hydroxyapatite-containing regions within the tumor; histological staining verified the regional distribution of [18F]NaF and 223RaCl2 in the tumor corresponded with calcified osteoblastic tissue (Figure 1). Conclusions: Uptake of [18F]NaF and 223RaCl2 spatially co-localize in bone-forming tumor tissue and are similar in tumors containing hydroxyapatite. [18F]NaF PET/CT imaging may be useful in predicting uptake of 223RaCl2 in metastatic bone disease and potentially assess 223RaCl2 dosimetry to individual malignant bone lesions.