Assessment of skeletal tumour burden on 18F-NaF PET/CT using a new quantitative method.

PURPOSE: The purpose of this study was to test a method of quantifying skeletal tumour burden with F-NaF PET/CT.

PATIENTS AND METHODS: We retrospectively reviewed the charts of 117 patients who underwent F-NaF PET/CT for the detection of bone metastases, 68 women and 49 men, 16-82 years old (mean±SD: 62.9±10.7 years). Mean standardized uptake values (SUVmean) were measured in five anatomic sites to evaluate normal skeleton activity. The influence of sex and age was investigated. Skeletal tumour burden was calculated in 69 exams positive for bone metastases using volumetric data and SUVmean values. Intraobserver and interobserver reproducibility was tested. In 10 patients with breast cancer, skeletal tumour burden in pretreatment and post-treatment F-NaF PET/CT was compared with tumour marker and clinical evolution.

RESULTS: The range of normal skeleton SUVmean for the 410 volume of interests analysed was 2.2-5.9 (mean±SD: 4.4±1.5). A threshold of 10 was chosen to exclude F-NaF normal skeleton uptake. An inverse relationship was found between normal skeleton SUVmean and age (r=-0.237; P=0.032). Our results show excellent intraobserver and interobserver reproducibility, with intraclass correlation values of 0.995 and 0.997, respectively. The percentage change in the skeletal tumour burden in response to therapy shows a moderate direct correlation with the percentage variation of the tumour marker (r=0.668; P=0.035).

CONCLUSION: The methodology that we used to quantify skeletal tumour burden is easy to perform, highly reproducible and allows for the evaluation of bone tumour response to therapy in a subgroup of breast cancer patients. The possibility of skeletal tumour burden quantification is another advantage of F-NaF PET/CT over the visual and subjective interpretation of bone scintigraphy.