Purpose To identify the optimal dose parameters predictive for local/lobar control after stereotactic ablative radiation therapy (SABR) in early-stage non-small cell lung cancer (NSCLC). Methods and Materials This study encompassed a total of 1092 patients (1200 lesions) with NSCLC of clinical stage T1-T2 N0M0 who were treated with SABR of 50 Gy in 4 fractions or 70 Gy in 10 fractions, depending on tumor location/size, using computed tomography-based heterogeneity corrections and a convolution superposition calculation algorithm. Patients were monitored by chest CT or positron emission tomography/CT and/or biopsy after SABR. Factors predicting local/lobar recurrence (LR) were determined by competing risk multivariate analysis. Continuous variables were divided into 2 subgroups at cutoff values identified by receiver operating characteristic curves. Results At a median follow-up time of 31.7 months (interquartile range, 14.8-51.3 months), the 5-year time to local recurrence within the same lobe and overall survival rates were 93.8% and 44.8%, respectively. Total cumulative number of patients experiencing LR was 40 (3.7%), occurring at a median time of 14.4 months (range, 4.8-46 months). Using multivariate competing risk analysis, independent predictive factors for LR after SABR were minimum biologically effective dose (BED10) to 95% of planning target volume (PTVD95 BED10) ≤86 Gy (corresponding to PTV D95 physics dose of 42 Gy in 4 fractions or 55 Gy in 10 fractions) and gross tumor volume ≥8.3 cm3. The PTVmean BED10 was highly correlated with PTVD95 BED10. In univariate analysis, a cutoff of 130 Gy for PTVmean BED10 (corresponding to PTVmean physics dose of 55 Gy in 4 fractions or 75 Gy in 10 fractions) was also significantly associated with LR. Conclusions In addition to gross tumor volume, higher radiation dose delivered to the PTV predicts for better local/lobar control. We recommend that both PTVD95 BED10 >86 Gy and PTVmean BED10 >130 Gy should be considered for SABR plan optimization.
|Original language||English (US)|
|Number of pages||10|
|Journal||International Journal of Radiation Oncology Biology Physics|
|State||Published - Jul 15 2016|
All Science Journal Classification (ASJC) codes
- Radiology Nuclear Medicine and imaging
- Cancer Research