Noninvasive cardiac radioablation for ventricular tachycardia: dosimetric comparison between linear accelerator- and robotic CyberKnife-based radiosurgery systems

Background Few dosimetric comparisons have been published between linear accelerator (LA)-based systems and CyberKnife (CK)-based robotic radiosurgery systems for cardiac radio-ablation in ventricular tachycardia. This study aimed to compare the dosimetry of noninvasive cardiac radio-ablation deliverable on LA with that on CK.Methods Thirteen patients who underwent noninvasive cardiac radio-ablation by LA were included. The prescribed dose was 25 Gy in 1 fraction, and the average planning target volume was 49.8 +/- 31.0 cm(3) (range, 14.4-93.7 cm(3)). CK plans were generated for comparison.Results Both the CK and LA plans accomplished appropriate dose coverage and normal tissue sparing. Compared with the LA plans, the CK plans achieved significantly lower gradient indices (3.12 +/- 0.71 vs. 3.48 +/- 0.55, p = 0.031) and gradient measures (1.00 +/- 0.29 cm vs. 1.17 +/- 0.29 cm, p < 0.001). They had similar equivalent conformity indices (CK vs. LA: 0.84 +/- 0.08 vs. 0.87 +/- 0.07, p = 0.093) and maximum doses 2 cm from the planning target volume (PTV) in any direction (CK vs. LA: 50.8 +/- 9.9% vs. 53.1 +/- 5.3%, p = 0.423). The dosimetric advantages of CK were more prominent in patients with a PTV of <= 50 cm(3) or a spherical PTV. In patients with a PTV of > 50 cm(3) or a non-spherical PTV, the LA and CK plans were similar regarding dosimetric parameters. CK plans involved more beams (232.2 +/- 110.8 beams vs. 10.0 +/- 1.7 arcs) and longer treatment times (119.2 +/- 43.3 min vs. 22.4 +/- 1.6 min, p = 0.007).Conclusions Both CK and LA are ideal modalities for noninvasive cardiac radio-ablation. Upfront treatment should be considered based on clinical intent.