2232 - Feasibility and Initial Clinical Experience with Online Adaptive MRI-Guided SBRT for Central Lung Tumors

02:30pm - 04:00pm PT

Hall F

Screen: 20

POSTER

Presenter(s)

Teo Stanescu, PhD - Princess Margaret Cancer Centre, Toronto, ON

T. Stanescu^1,2, A. Shessel³, O. Semeniuk⁴, H. T. Wong⁵, and A. J. Hope⁶; ¹University of Toronto, Toronto, ON, Canada, ²Medical Physics, Princess Margaret Cancer Centre, University Health Network, Toronto, ON, Canada, ³Radiation Medicine Program, Princess Margaret Cancer Centre, Toronto, ON, Canada, ⁴Rhode Island Hospital, Providence, RI, ⁵Princess Margaret Cancer Center, Toronto, ON, Canada, ⁶Princess Margaret Cancer Centre, University Health Network, Toronto, ON, Canada

Purpose/Objective(s): To establish clinical feasibility and provide preliminary data on MR-guided online adaptive SBRT for the treatment of central lung malignancies using a 1.5T MR-Linac.

Materials/Methods:

Six patients with central lung tumors underwent SBRT delivering 35 or 50 Gy in 5 fractions on a 1.5T MR-Linac. Online adaptive MRI-guided radiotherapy was implemented, utilizing 3D T2-weighted image data acquired during both the exhale phase and time-averaged scans optimized for lung targets. To ensure treatment robustness, patient setup and intra-fraction motion were quantified through repeat 3D scans, 2D fast imaging, and an automated dose-of-the-day (DOTD) methodology. The DOTD analysis, based on deformable image registration (DIR), projected the pre-treatment adapt-to-shape (ATS) plan onto during-beam-on (BON) images, generating a dosimetric report on treatment delivery quality. Finally, dose accumulation was performed via DIR to evaluate the overall effectiveness of MR-guided treatment. Clinical goals were evaluated based on several scenarios: (a) reference plan propagated to all ATS datasets to simulate conventional IGRT, (b) each ATS plan propagated to remaining fractions, and (c) sum of all ATS plans on an arbitrary fraction.

Results: MR image data facilitated GTV delineation and real-time patient setup monitoring throughout all treatment fractions. The mean GTV volume and motion were 10.3 cc (4.1-15.1 cc) and 4.8 mm (0 – 11 mm), respectively. The inter-fraction GTV volumes varied within 3% per case. Overall, online adaptation provided uncompromised target coverage and significantly improved OAR sparing. The DOTD analysis revealed minimal dosimetric deviations, with GTV and OAR metrics consistently remaining within 10 cGy per fraction of the planned clinical goals.

Conclusion: This initial experience demonstrates the viability of online adaptive MRI-guided stereotactic body radiotherapy for central lung malignancies. The technique offers the potential for enhanced OAR sparing, suggesting a promising avenue for improving treatment tolerance in this challenging patient population.