LUSTRE: A Phase III Randomized Trial of Stereotactic Body Radiotherapy (SBRT) vs. Conventionally Hypofractionated Radiotherapy (CRT) for Medically Inoperable Stage I Non-Small Cell Lung Cancer (NSCLC)

Purpose/Objective(s) SBRThas been adopted as standard of care for Stage I medically inoperable NSCLC, yet there is inconclusive data from randomized controlled trials (RCT) regarding efficacy. There remains concern regarding long term disease control and toxicity, particularly in centrally located tumors. The purpose of this trial was to compare SBRT to hypofractionated CRT in central and peripheral NSCLC. Materials/Methods A phase III RCT was conducted in 16 Canadian centers. Patients deemed medically inoperable with either histologically confirmed stage I (≤5cm) NSCLC or a suspicious growing FDG-PET avid lesion were randomized 2:1 to receive either SBRT of 48 Gy/4 fractions (peripheral NSCLC) or 60 Gy/8 fractions (central NSCLC - within 1 cm of mediastinum or 2 cm of the proximal bronchial tree), versus CRT of 60 Gy/15 fractions. Stratification was by tumor size (≤3cm vs >3-5 cm), location: central vs peripheral, and clinical center. All radiation plans were subject to real-time/final review. The primary outcome was local control (LC), defined by the absence of primary tumor or marginal failure. Secondary outcomes included disease-free survival (DFS), overall survival (OS), toxicity, and radiation treatment related death (RTRD). Outcomes were centrally adjudicated, blinded by arm. The primary objective was to detect a LC improvement of SBRT at 3 years from 75% to 87.5% (Hazard Ratio (HR)=0.46) assuming a two-sided α=0.05 and power of 85%, with a planned sample size of 324 patients. Results From May 2014 to January 2020, 233 (154 SBRT, 79 CRT) patients were accrued. Recruitment to the trial closed early due to slow accrual. The mean age of patients was 75 years; 27% had tumors that were centrally located; 49% had biopsy-proven NSCLC; 71% had ≤3cm lesions; and mean tumor diameter was 2.5 cm. With a median follow-up of 36 months, 34 (18 SBRT, 16 CRT) local failures were observed. The 3-year LC rate was 87.6% for SBRT and 81.2% for CRT (HR=0.61, 95% confidence interval (CI):0.31-1.20, p=0.15). The observed treatment effects for DFS were HR=0.83 (95% CI:0.54-1.28, p=0.40), and HR=1.18 (95% CI:0.80-1.76, p=0.40) for OS. Only one patient in each arm experienced grade 3 acute toxicity (no grade 4/5 toxicities were observed). Late grade 3/4 toxicities occurred in 7 patients: SBRT - central 3/45 (6.6%), peripheral 2/109 (1.8%), CRT - central 1/19 (5.2%), peripheral 1/60 (1.6%). One patient who received 60 Gy/8 fractions for a central NSCLC experienced a possible RTRD (hemoptysis). Conclusion This is the largest reported RCT of lung SBRT compared to a contemporary CRT control arm, with mature follow-up and the inclusion of patients with central tumors. There was an observed improvement of LC with SBRT compared to CRT, however, the trial was underpowered to confirm this. No evidence of differences were observed in DFS and OS. Very few patients experienced severe late toxicities, including those with central tumors. This study confirms the efficacy and safety of SBRT for both central and peripheral Stage I NSCLC (NCT01968941).