Single-isocenter multiple-target stereotactic radiosurgery for multiple brain metastases: dosimetric evaluation of two automated treatment planning systems

Abstract Purpose Automated treatment planning systems are available for linear accelerator (linac)-based single-isocenter multi-target (SIMT) stereotactic radiosurgery (SRS) of brain metastases. In this study, we compared plan quality between Brainlab Elements Multiple Brain Metastases (Elements MBM) software which utilizes dynamic conformal arc therapy (DCAT) and Varian HyperArc (HA) software using a volumetric modulated arc therapy (VMAT) technique. Patients and methods Between July 2018 and April 2021, 36 consecutive patients ≥ 18 years old with 367 metastases who received SIMT SRS at UPMC Hillman Cancer San Pietro Hospital, Rome, were retrospectively evaluated. SRS plans were created using the commercial software Elements MBM SRS (Version 1.5 and 2.0). Median cumulative gross tumor volume (GTV) and planning tumor volume (PTV) were 1.33 cm 3 and 3.42 cm 3 , respectively. All patients were replanned using HA automated software. Extracted dosimetric parameters included mean dose (D mean ) to the healthy brain, volumes of the healthy brain receiving more than 5, 8,10, and 12 Gy (V 5Gy , V 8Gy , V 10Gy and V 12Gy ), and doses to hippocampi. Results Both techniques resulted in high-quality treatment plans, although Element MBM DCAT plans performed significantly better than HA VMAT plans, especially in cases of more than 10 lesions). Median V 12Gy was 13.6 (range, 1.87–45.9) cm 3 for DCAT plans and 18.5 (2.2–62,3) cm 3 for VMAT plans ( p < 0.0001), respectively. Similarly, V 10Gy , V 8Gy , V 5Gy ( p < 0.0001) and median dose to the normal brain ( p = 0.0001) were favorable for DCAT plans. Conclusions Both Elements MBM and HA systems were able to generate high-quality plans in patients with up to 25 brain metastases. DCAT plans performed better in terms of normal brain sparing, especially in patients with more than ten lesions and limited total tumor volume.