Monte Carlo–Based Treatment Planning May Increase Chest Wall Pain and Rib Fractures in Lung Stereotactic Body Radiation Therapy (SBRT) Patients Treated According to RTOG 0915 Protocol

Assessment of Monte Carlo algorithm for compliance with RTOG 0915 dosimetric criteria in peripheral lung cancer patients treated with stereotactic body radiotherapy

The purpose of the study was to evaluate Monte Carlo‐generated dose distributions with the X‐ray Voxel Monte Carlo (XVMC) algorithm in the treatment of peripheral lung cancer patients using stereotactic body radiotherapy (SBRT) with non‐protocol dose‐volume normalization and to assess plan outcomes utilizing RTOG 0915 dosimetric compliance criteria. The Radiation Therapy Oncology Group (RTOG) protocols for non‐small cell lung cancer (NSCLC) currently require radiation dose to be calculated using tissue density heterogeneity corrections. Dosimetric criteria of RTOG 0915 were established based on superposition/convolution or heterogeneities corrected pencil beam (PB‐hete) algorithms for dose calculations. Clinically, more accurate Monte Carlo (MC)‐based algorithms are now routinely used for lung stereotactic body radiotherapy (SBRT) dose calculations. Hence, it is important to determine whether MC calculations in the delivery of lung SBRT can achieve RTOG standards. In this report, we evaluate iPlan generated MC plans for peripheral lung cancer patients treated with SBRT using dose‐volume histogram (DVH) normalization to determine if the RTOG 0915 compliance criteria can be met. This study evaluated 20 Stage I‐II NSCLC patients with peripherally located lung tumors, who underwent MC‐based SBRT with heterogeneity correction using X‐ray Voxel Monte Carlo (XVMC) algorithm (Brainlab iPlan version 4.1.2). Total dose of 50 to 54 Gy in 3 to 5 fractions was delivered to the planning target volume (PTV) with at least 95% of the PTV receiving 100% of the prescription dose (V100%≥95%). The internal target volume (ITV) was delineated on maximum intensity projection (MIP) images of 4D CT scans. The PTV included the ITV plus 5 mm uniform margin applied to the ITV. The PTV ranged from 11.1 to 163.0 cc (mean=46.1±38.7 cc). Organs at risk (OARs) including ribs were delineated on mean intensity projection (MeanIP) images of 4D CT scans. Optimal clinical MC SBRT plans were generated using a combination of 3D noncoplanar conformal arcs and nonopposing static beams for the Novalis‐TX linear accelerator consisting of high‐definition multileaf collimators (HD‐MLCs: 2.5 mm leaf width at isocenter) and 6 MV‐SRS (1000 MU/min) beam. All treatment plans were evaluated using the RTOG 0915 high‐ and intermediate‐dose spillage criteria: conformity index (R100%), ratio of 50% isodose volume to the PTV (R50%), maximum dose 2 cm away from PTV in any direction (D2cm), and percent of normal lung receiving 20 Gy V20 or more. Other OAR doses were documented, including the volume of normal lung receiving 5 Gy V5 or more, dose to <0.35 cc of spinal cord, and dose to 1000 cc of total normal lung tissue. The dose to <1 cc, <5 cc, <10 cc of ribs, as well as maximum point dose as a function of PTV, prescription dose, and a 3D distance from the tumor isocenter to the proximity of the rib contour were also examined. The biological effective dose (BED) with α/β ratio of 3 Gy for ribs was analyzed. All 20 patients either fully met or were within the minor deviation dosimetric compliance criteria of RTOG 0915 while using DVH normalization. However, only 5 of the 20 patients fully met all the criteria. Ten of 20 patients had minor deviations in R100% (mean=1.25±0.09), 13 in R50% (mean=4.5±0.6), and 11 in D2cm (mean=61.9±8.5). Lung V20, dose to 1000 cc of normal lung, and dose to <0.35 cc of spinal cord were met in accordance with RTOG criteria in 95%, 100%, and 100%, respectively, with exception of one patient who exhibited the largest PTV (163 cc) and experienced a minor deviation in lung V20 (mean=4.7±3.4%). The 3D distance from the tumor isocenter to the proximal rib contour strongly correlated with maximum rib dose. The average values of BED3Gy for maximum point dose and dose to <1 cc of ribs were higher by a factor of 1.5 using XVMC compared to RTOG 0915 guidelines. The preliminary results for our iPlan XVMC dose analyses indicate that the majority (i.e., 75% of patient population) of our patients had minor deviations when compared to the dosimetric guidelines set by RTOG 0915 protocol. When using an exclusively sophisticated XVMC algorithm and DVH normalization, the RTOG 0915 dosimetric compliance criteria such as R100%, R50%, and D2cm may need to be revised. On average, about 7% for R100%, 13% for R50%, and 14% for D2cm corrections from the mean values were necessary to pass the RTOG 0915 compliance criteria. Another option includes rescaling of the prescription dose. No further adjustment is necessary for OAR dose tolerances including normal lung V20 and total normal lung 1000 cc. Since all the clinical MC plans were generated without compromising the target coverage, rib dose was on the higher side of the protocol guidelines. As expected, larger tumor size and proximity to ribs correlated to higher absolute dose to ribs. These patients will be clinically followed to determine whether delivered MC‐computed dose to PTV and the ribs dose correlate with tumor control and severe chest wall pain and/or rib fractures. In order to establish new specific MC‐based dose parameters, further dosimetric studies with a large cohort of MC lung SBRT patients will need to be conducted.

PACS number(s): 87.55.k