3138 - Managing Intrafraction Patient Motion Using Robust Planning Method for Same Day MLC-based Spatially Fractionated Radiotherapy of Large Unresectable and Bulky Tumors
Presenter(s)
S. Srinivasan1, R. Miller1,2, J. A. Holmes1, O. Ishaq Jr3, R. C. Zellars4, and D. Pokhrel1; 1Department of Radiation Oncology, Indiana University School of Medicine, Indianapolis, IN, 2Indiana University, Indianapolis, IN, 3Indiana University School of Medcine, Indianapolis, IN, 4Indiana University Department of Radiation Oncology, Indianapolis, IN
Purpose/Objective(s): To evaluate the dosimetric impact on plan quality due to intrafraction patient motion error via robust planning method of CBCT-guided same day 3D MLC-based spatially fractionated radiotherapy (SFRT) for debulking large (= 8 cm) unresectable tumors.
Materials/Methods: In this IRB approved retrospective study, we evaluated the dosimetric impact of intrafraction motion errors, in SFRT plans of 10 patients (4 head and neck, 1 bone, 5 abdominal tumors). First, an extensive in-house testing and validation of the 3D MLC-based SFRT method was completed. Secondly, an independent dose validation of the 15 Gy in 1 fraction, using 6-crossfire MLC-field of 6MV was done with IROC SRS head phantom, resulting in absolute TLDs dose of ±1% and relative films profiles agreement of ±98% with Acuros dose calculation. For intrafraction motion evaluation, SFRT plan for 15 Gy in 1 fraction was generated. Intrafraction patient motion was simulated in eclipse planning system with ±3 mm and ±5 mm isocenter shift in x,y,z directions using plan uncertainty tool, which resulted in 12-plan uncertainty dose metrics and DVH bands for each patient’s SFRT plan. Dosimetric impact was evaluated for peak-to-valley-dose ratio (PVDR) = GTV(D10%) ÷ GTV(D90%), GTV(V7.5Gy), GTV mean dose, and maximum dose to adjacent organ-at-risk (OAR).
Results: In the 10 SFRT patients’ original plans, GTV diameter, PVDR, GTV mean dose, GTV(V7.5Gy) were 10.5 ± 2.7 cm (8.6–17.7 cm), 2.5 ± 0.2 (range 2.3–2.9), 8.6 ± 0.2 Gy (range 8.1–9.0 Gy), and 57.5 ± 3.5 % (range 49.1–68.3 %), respectively. Based on simulated motion errors, the percentage difference in GTV PVDR ranged between -4% and 9% with large changes found in abdominal tumors and relatively smaller targets in head and neck region. Due to simulated motion, insignificant changes (<1%) in GTV mean dose and GTV (V7.5Gy) were seen compared to original plans. The proximal OAR maximum dose varied between -4.1% and 2.7%. In abdominal tumors, small bowel maximum dose was susceptible to motion error up to 2.4% and in head and neck tumors, maximum dose to spinal canal and trachea were affected by the motion errors.
Conclusion: Due to the high dose gradients, in SFRT plans, delivered dose distribution can be affected by patient motion and setup uncertainties. Robust plan optimization allows mitigation of these issues by creating a plan that is less sensitive to these errors. For more accurate treatment delivery of large single dose of 15 Gy or higher SFRT treatment, robust planning for site-specific and more accurate patient setup and immobilization methods including 4D CT similar to SBRT patient setups, are highly recommended for treating CBCT-guided SFRT on the same day of CT simulation.