Daniel M. Arsenault, MD, Maria I. Monterroso, MS, Zhendong Wang, PhD, Jean L. Wright, MD; University of Miami
Purpose and Objective: Advances in radiotherapy technology have led to an increase in stereotactic body radiotherapy (SBRT) treatments. The most recent CyberKnife (CK) tracking algorithm upgrade, Lung-Optimized Treatment (LOT), allows for treatment of lung tumors located anywhere in the lung without the need for fiducial placement by taking orthogonal kV images to locate and track the lesion when it is visible in the camera or taking kV images of the spine if the lesion is not visible. Fiducial-less SBRT can also be delivered using linear accelerator (LINAC)-based systems. Given that the treatment time is substantially greater with the CK-based approach, we performed a dosimetric comparison of these two systems to attempt to quantify any dosimetric advantages that would substantiate the increased treatment time of CK.
Materials and Methods: We retrospectively analyzed 10 lung patients who had been treated with LINAC-based SBRT using volumetric arc therapy (VMAT) and generated CK-based treatment plans using the LOT tracking algorithm in 0-View mode. All plans included an internal tumor volume (ITV). All VMAT plans were generated using the Eclipse (Varian) treatment planning system (TPS) with the AAA dose calculation algorithm, while the CK plans were generated using the Mulitplan (Accuray) TPS with the Monte Carlo dose calculation algorithm. We compared planning target volume (PTV) coverage; conformity index (CI); lung V5, V10, V20, V25, mean lung dose (MLD), 90% IDL/PTV, 80% IDL/PTV, and 70% IDL/PTV; heart V5, V10, and mean heart dose (MHD); and spinal cord maximum and esophagus maximum doses, when applicable. The mean and standard deviations were calculated for all patient parameters, and a paired t-test was used to identify any statistically significant difference in dosimetric parameters between two systems.
Results: Two tumors were located in the left upper lobe, two were located in the right middle lobe, three were located in the left lower lobe one was located in the right upper lobe, and one was located in the right hilum. The mean LINAC lung V5 was 14.43 cc and 23.48 cc in the CK plans (P = .00088). The mean LINAC lung V10 was 8.49 cc and 10.29 cc in the CK plans (P = .04). The mean LINAC MLD was 2.99 Gy and 4.21 Gy in the CK plans (P = .0009). The mean LINAC CI was 1.035 and 1.141 in the CK plans (P = .012). The mean LINAC 90% IDL/PTV was 1.428 and 1.602 in the CK plans (P = .016). There were no other statistically significant differences between dosimetric parameters.
Conclusion: When comparing fiducial-less SBRT of the lung using VMAT and CK, we identified statistically significant dosimetric advantages in the LINAC plans as compared with the CK plans. These results may be due to the field size (FS) limitation in the CK, where the largest available FS is 6 cm in diameter and may be suboptimal when treating larger targets. Given the superior dosimetry and decreased treatment time of VMAT plans, fiducial-less SBRT of the lung may be more advantageous and efficient using a LINAC system than CK with LOT in 0-View mode. Future study with an increased number of patients is needed.