Dose-volume-response analysis in stereotactic radiotherapy for early lung cancer.

ABSTRACT

BACKGROUND AND PURPOSE: Japanese and Western approaches to stereotactic ablative radiotherapy (SABR) are considerably different, particularly with respect to dose prescription and reporting, which makes comparisons of Japanese versus European or American results challenging. Using individual patient data, the aim of this study was to analyze the dose-local-control relationship and its impact on survival. MATERIAL AND METHODS: Patients receiving SABR for single-lesion early stage NSCLC in Osaka (OM) or Groningen (GN) were analyzed. Doses were recalculated using state-of-the-art dose calculation algorithms and expressed as biologically effective dose (BED) at PTV margin. Survival, local control (LC), and effect of treatment failure in operable and inoperable patients on survival were analyzed. RESULTS: Between 2006 and 2010, 383 patients were included. The BED at PTV periphery was 102 Gy10 (±21) in GN and 83 Gy10 (±5) in OM. Unadjusted overall survival (OS) was better in OM (72% vs 52%; p<0.001), but GTVs and performance status (PS) were also significantly more favorable in OM. Adjusted for GTV and PS, OS was not different between institutions (HR 0.88; p=0.47). LC was better in GN (93% vs 84%; p<0.05). Local control predicted survival in operable patients Adjusted for GTV and PS, the HR of local failure for OS was 7.5 (2-27; p=0.003) for operable, and 1.1 (0.7-1.9; p=0.6) for inoperable patients. CONCLUSIONS: Sufficient dose is crucial for local control, which was a significant factor for survival for operable patients.