Knowledge-Based Planning for Identifying High-Risk Stereotactic Ablative Radiation Therapy Treatment Plans for Lung Tumors Larger Than 5 cm

PURPOSE: Stereotactic ablative body radiotherapy (SABR) for lung tumors ≥5cm, can be associated with more toxicity. We investigated the relationship between dosimetry and toxicity and used a knowledge-based planning solution to retrospectively perform individualized treatment plan quality assurance (QA) with the aim of identifying where planning could have been improved. MATERIAL AND METHODS: Prior retrospective analysis of 53 patients with primary or recurrent non-small cell lung cancer ≥5cm, treated with 5 or 8-fraction volumetric modulated arc therapy SABR, between 2008-2014, showed 30% ≥grade (G) 3 toxicity. During this period, several improvements were made to departmental planning protocols. RapidPlan was used to compare dosimetry of patients with/without G≥3 toxicity and a model comprising plans from patients without toxicity and compliant with the current planning protocol, was used to quality assure the plans from patients who had toxicity. RESULTS: 16/53 patients had G≥3 toxicity, including 10 radiation pneumonitis (RP), 3 lung hemorrhage (LH) (1 also had RP) and one airway stenosis/atelectasis. RP was again shown to be significantly correlated with contralateral and total-lung V5 and mean lung dose. The 4 highest contralateral-lung doses belonged to patients with RP. 5/10 clinical plans in patients with RP had a contralateral-lung mean dose up to 2.5x higher than the knowledge-based plan. For 2/3 patients with LH and one with airway stenosis/atelectasis, the clinical plans had the highest proximal bronchial tree doses, which was also higher than in plans from the model. In 8 patients with G≥3 toxicity, clinical plans had similar dosimetry as the predictions from the model. CONCLUSION: A "no-toxicity" RapidPlan model identified the potential for dosimetric improvement in nearly 50% of historical treatment plans from patients with G≥3 toxicity after SABR for lung tumors≥5cm. Model-based QA may be useful for benchmarking treatment planning protocols in routine practice and in clinical studies.