Effects of setup margin and biology tracking zone variations on dosimetric plan quality and treatment feasibility in biology-guided radiotherapy for bone metastasis.

Abstract

e15107 Background: In biology-guided radiotherapy (BgRT), real-time positron emission tomography (PET) signals are used to guide radiation beams to F-18 fluorodeoxyglucose (FDG)-avid lesions. This could enable precise targeting of the tumor with reduced setup margins for the planning target volume (PTV) when the tumor moves within a biology tracking zone (BTZ). However, since FDG concentration is mapped to beam fluence in BgRT plan optimization, PTV and BTZ geometrical variations profoundly impact dose distribution and key FDG metrics for target tracking. This study is aimed at evaluating effects of setup margin and BTZ variations on dosimetric plan quality and BgRT feasibility. Methods: Treatment planning data for five patients with non-superficial bone lesions was retrieved. Each patient had a CT simulation followed by a PET scan on the BgRT machine with 15 mCi FDG. The gross target volume (GTV) was defined as the FDG-avid volume. In the standard BgRT plan, the PTV was created by adding a 5-mm setup margin to the GTV and the BTZ as a 5-mm margin expansion from the PTV. Ten more treatment plans were created, first by reducing the GTV-to-PTV margin from 5 to 0 mm in a 1-mm step size while keeping the BTZ size constant, and then by shrinking the BTZ size along with the PTV by keeping a 5-mm PTV-to-BTZ margin. The same plan optimization template was applied to all treatment plans for each case. Calculated dosimetric parameters include the conformity index (CI), CI80 and CI50 (ratio of the 80% and 50% isodose volume to the PTV volume, respectively), and homogeneity index (HI). The activity concentration (AC) and normalized target signal (NTS) were calculated based on FDG concentration within the PTV and surrounding the BTZ. Results: With an average GTV volume of 6.6±3.8 cm 3 and prescribed dose ranging from 10 to 27 Gy in one to three fractions, the standard BgRT treatment plans had an average CI, CI80, CI50, and HI of 1.18±0.10 (range: 1.07 – 1.29), 2.57±0.44 (range: 2.04 – 3.26), 7.26±1.08 (range: 6.29 – 9.11), and 1.35±0.08 (range: 1.24 – 1.43), respectively; the average AC and NTS was 16.85±14.80 kBq/ml and 9.73±5.45, respectively. When the GTV-to-PTV margin was reduced while keeping a constant BTZ, CI, CI80, and CI50 all increased monotonically while the AC and NTS remained constant. At zero GTV-to-PTV margin, the average CI was 1.72±0.70 (range: 1.14 – 2.76). When the PTV-to-BTZ margin was kept constant with decreasing GTV-to-PTV margins, the AC decreased monotonically with mixed results for the NTS. The AC fell below the BgRT feasibility threshold for one case when the GTV-to-PTV margin was below 3 mm. Conclusions: While a smaller GTV-to-PTV margin led to a reduced absolute irradiated volume, less conformal target dose and slower dose drop-off were observed in BgRT treatment plans. A reduction in BTZ size could affect BgRT feasibility for lesions with a low FDG concentration.