Validation of the Semi-Automatic Quantification of 18 F-Fluoride PET/CT Whole-body Skeletal Tumor Burden.

PURPOSE: To validate a semi-automatic quantification of the skeletal tumor burden on 18 F-Fluoride PET/CT by using manual quantification as a reference. METHODS: We quantified 51 18 F-Fluoride PET/CT exams performed on female breast cancer patients. Clinical information (age, time of disease, presence of visceral metastases, and time to death, progression or bone event) was recorded. The total volume of fluoride-avid skeletal metastases (FTV10) and the total activity of 18 F-Fluoride-avid metastases (TLF10) were calculated manually (mTLF10 and mFTV10, respectively) and semi-automatically (saTLF10 and saFTV10, respectively). RESULTS: Manual and semi-automatic metrics were highly correlated (p<0.0001; CI 0.9300 to 0.9769). On multivariable analysis, saTLF10 measures were significantly correlated to OS ( P = 0.0001) and progression-free survival ( P = 0.0006). Approximate times for calculating skeletal tumor burden (semi-automatic vs manual) were, respectively: 30s vs 321s in patients with less than 5 metastases; 120s vs 640s in patients with 5-10 metastases and 240s vs 1207s in patients with more than 10 lesions. CONCLUSION: Semi-automatic quantification of whole-body 18 F-Fluoride PET/CT skeletal tumor burden can replace manual quantification in breast cancer patients and is a strong independent prognostic imaging biomarker.