Analysis of metastatic bone lesions on sequential 18F FDG and 18F NaF scans in advanced genitourinary cancers

Objectives: Bone metastasis is a common manifestation in non-prostate genitourinary (GU) tumors. PET/CT can noninvasively add insight into the process of osseous metastasis and provide information for improved therapeutic strategy development. 18F FDG is a well-established PET agent for detection of malignancy in soft tissues and osteolytic lesions and 18F NaF is a highly sensitive PET radiotracer for bone lesions. It is especially useful for identifying osteoblastic lesions. In this study, the combination of these two imaging approaches is evaluated for detection and monitoring skeletal metastases in non-prostate GU cancers.

Methods: Study patients with GU cancers with bone metastasis were enrolled on NCT02496208, a phase I treatment trial in metastatic GU cancers. All patients were male ranging in age from 20 - 77 years. Baseline studies involved 18F FDG PET/CT imaging followed immediately by 18F NaF PET/CT imaging the same day. Follow-up restaging studies were repeated with sequential imaging at 4-8 weeks after treatment in 7 patients. Bone lesions with focal uptake on either scan were manually contoured and uptake sites were recorded. Uptake values were reported as SUV. CT characteristics for each lesion were noted as blastic, lytic, bone marrow (no obvious anatomical lesion seen on CT to correspond to focal radiotracer activity on PET) or a combination. Data was analyzed by Wilcoxon rank sum test and Spearman correlation coefficient test. Results: 10 patients were imaged in the study. All patients were male ranging in age from 20 - 77 years. Cancer histologies in the patient group were: testicular (2), renal cell (1) and urothelial (7). A total of 107 lesions were identified on baseline NaF imaging, out of which 61 lesions were identified on FDG scans (57%). From the 107 baseline lesions observed on NaF imaging, CT metastasis characterization showed that 74.7% (80/107) were blastic, 4.6% (5/107) were lytic, and 21.5% (22/107) were bone marrow lesions. FDG failed to identify 43.7% (35/80) blastic lesions, 45.4% (10/22) bone marrow and 20% (1/5) lytic lesions identified on both NaF and CT scans. Restaging follow-up studies performed on 7 patients detected a total of 110 lesions on NaF PET/CT, out of which 60 (55%) were identified on FDG imaging. FDG imaging identified 9 additional lesions not visualized on NaF scans. These 9 lesions were all bone marrow lesions in the patient with testicular cancer. 81 follow-up scan metastases were identified as blastic on NaF, out of which FDG failed to identify 56.7% (46/81). Out of a total of 34 bone marrow lesions, NaF failed to identify 26.4% (9/34) which were only visualized in FDG scans. FDG imaging failed to identify 4/34 (11.7%) metastases, which were identified on NaF imaging. 3 lytic lesions were identified on both FDG and NaF images. Blastic lesions were observed most frequently (80/107 at baseline and 81/119 at follow-up), followed by bone marrow lesions (22/107 and 34/119, respectively). A small proportion of lytic lesions were observed (<10 at each timepoint). FDG uptake was highest in bone marrow lesions (SUV range 1.9- 11.6) compared to blastic (SUV range 2.1-6.6) (p<0.001 both timepoints). Conversely, NaF lesion uptake was highest in blastic lesions (SUV range 10.4- 137.1) compared to bone marrow (SUV range 10.3- 96.5) (p=0.0002 both timepoints). In the 38 lesions detected by both tracers at both timepoints, the relative change in uptake for all SUV metrics were highly correlated for NaF and FDG (ρ=0.45-0.58). Conclusion: NaF imaging had increased bone metastasis detection compared to FDG imaging for both blastic and lytic skeletal metastases. However, both modalities failed to identify all bone marrow lesions noted on CT correlative images. A combination of both scanning types may be optimal for early bone marrow metastasis detection, enabling accurate early staging and management of patients. Acknowledgements: This research was supported by the NCI Contract No. HHSN261200800001E.