Acuros ® dose verification of ultrasmall lung lesions with EBT‐XD film in a homogeneous and heterogeneous anthropomorphic phantom setup

Purpose Modern type 'c' dose calculation algorithms like Acuros(R)can predict dose for lung tumors larger than approximately 4 cm(3)with a relative uncertainty up to 5%. However, increasingly better tumor diagnostics are leading to the detection of very small early-stage lung tumors that can be treated with stereotactic body radiotherapy (SBRT) for inoperable patients. This raises the question whether dose algorithms like Acuros(R)can still accurately predict dose within 5% for challenging conditions involving small treatment fields. Current recommendations for Quality Assurance (QA) and dose verification in SBRT treatments are to use phantoms that are as realistic as possible to the clinical situation, although water-equivalent phantoms are still largely used for dose verification. In this work we aim to demonstrate that existing dose verification methods are inadequate for accurate dose verification in very small lung tumors treated with SBRT. Method The homogeneous PTW Octavius4D phantom with the Octavius 1000 SRS detector ("Octavius4D phantom") and the heterogeneous CIRS Dynamic Thorax phantom ('CIRS phantom') were used for dose measurements. The CIRS phantom contained different lung-equivalent film-holding cylindrical phantom inserts ("film inserts") with water-equivalent spherical targets with diameters 0.5, 0.75, 1, 2, and 3 cm. Plans were calculated for 6 and 10 MV for each spherical target in the CIRS phantom, resulting in 14 treatment plans. The plans were delivered to both Octavius4D and CIRS phantom to compare measured dose in a commonly used homogeneous and more realistic heterogeneous phantom setup. In addition, treatment plans of seven clinical lung cancer patients with lung tumors below approximately 1.0 cm(3)were irradiated in the heterogeneous CIRS phantom. The actual tumor size within the clinical treatment plans determined the choice of the spherical target size, such that both measurement geometry and clinical target volumes match as closely as possible. The Acuros(R)dose algorithm (version 15.5.11) was used for all dose calculations reporting dose-to-medium using a 0.1-cm-grid size. Results The measurement discrepancies in the homogeneous Octavius4D phantom for the fourteen treatment plans were within 1.5%. Dose discrepancies between measurement and treatment planning systems (TPS) for the heterogeneous CIRS phantom increased for both 6 and 10 MV with decreasing target diameters up to 23.7 +/- 1.0% for 6 MV and 8.8 +/- 1.1% for 10 MV for the smallest target of 0.5 cm in diameter with a 2-mm-CTV-PTV margin. For the seven clinical plans this trend of increasing dose difference with decreasing tumor size is less pronounced although the smallest tumors show the largest differences between measurement and TPS up to 16.6 +/- 0.9%. Conclusion Current verification methods using homogenous phantoms are not adequate for lung tumors with diameters below approximately 0.75 cm. The current Acuros(R)dose calculation algorithm underestimates dose in very small lung tumors. Dose verification of small lung tumors should be performed in an anthropomorphic lung phantom incorporating a water-equivalent target that matches clinical tumor size as closely as possible.