Use of paired FDG and 18F-Fluoride PET to predict response to therapy in patients with bone dominant metastases from breast cancer

Objectives Bony lesions are common sites of breast cancer metastasis, but they can be difficult to assess for response to therapy with conventional imaging (CT and bone scan). Previous studies have shown that 18F-fluoride PET improves bone metastasis detection compared to bone scans and FDG-PET was predictive of time-to-progression (TTP) and risk of skeletal related events (SRE) in a retrospective setting. We prospectively evaluated the combination of 18F-fluoride PET and FDG-PET to assess response to therapy in patients starting a new therapy for bone dominant metastatic breast cancer.

Methods Patients with metastatic bone lesions were imaged with paired 18F-fluoride PET and FDG-PET studies prior to starting new therapy (baseline) and again at approximately 4 months after start of therapy. 18F-fluoride and FDG static images from skull base to mid-thighs were analyzed quantitatively using the maximum standardized uptake value (SUVmax) of up to 5 lesions with the most prominent 18F-fluoride lesion used in the statistical analysis. Tumor locations were confirmed by CT and regions-of-interest (ROIs) were drawn on both 18F-fluoride and FDG images. SUVs on the 18F-fluoride lesions were corrected with the mean uptake in representative normal bone. Choice of systemic therapy was at the discretion of the treating physician. The clinical endpoints (TTP and time to SRE) were determined by clinical follow-up data exclusive of the PET images.

Results 26 patients (mean age 56, range 33-91) completed all four PET scans. The majority (65%) had primary ductal cancers. The mean time from initial breast cancer diagnosis to discovery of bone metastases was 79 months (range 0-440). The average time between the first 18F-fluoride and FDG-PET scans was 10 days (range 1-32) and between the second set of scans was 22 days (range 1-75). The average time between sets of scans was 123 days (range 62-244) for 18F-fluoride and 128 days (range 48-217) for FDG. With a median of 37 months follow-up, 23/26 patients died from their disease (88%) (median overall survival (OS) was 1130 days (range 185-3892); 24/26 (92%) patients had documented disease progression (median TTP 186 days (range 77-3892); and 16/26 (62%) had an SRE (median time to SRE 981 days (range 0-2632). In univariate analysis, neither initial nor change in 18F-fluoride or FDG were prognostic for TTP, SRE or OS. In multivariable analysis, after adjusting for presence of an elevated tumor marker and baseline FDGmax, percent change in FDG SUVmax was associated with OS. Specifically, a 34% percentage reduction in FDGmax uptake at the second scan was associated with a 40% reduction in the risk of death. In a similar way, an increase of 6.3 units in baseline FDG SUV was associated with more than a doubling of risk of death (p=.05 percent change and .005 baseline SUVmax). In TTP and SRE this model had similar results- percentage change in FDGmax from baseline and FDGmax were significantly associated with the endpoints (TTP p=.004,.03, respectively; SRE p=.0005,.0004: (Age was included in the SRE model)). The percentage change in 18F-fluoride or 18F-fluoride SUVmax at baseline did not add any prognostic information to this model.

Conclusions Changes in FDG SUV over the course of treatment were a robust predictor of TTP, SRE and OS, while changes in 18F-fluoride PET did not predict response. These results indicate a role for FDG-PET, in assessing bone dominant breast cancer response to therapy and support a role for PET in future clinical trials. Funding support 5R01CA124573-05, Komen SAC130060, SFI 11/PI/1027