KEY POINTS
- This population-based study included 778 patients with 814 peripheral T1 to T2N0M0 NSCLC lesions treated across British Columbia between 2021 and 2024. Single-fraction SABR was used for 283 lesions and multi-fraction SABR for 531.
- Single-fraction schedules were 30 Gy or 34 Gy in one fraction. Multi-fraction schedules were 48 Gy in four fractions or 54 Gy in three fractions. Median follow-up was 25.8 months.
- Local control was similar. Two-year local failure was 3.5% after single-fraction versus 4.8% after multi-fraction SABR (p=0.11), and single-fraction treatment remained unassociated with increased local failure after multivariable analysis.
- Overall survival was essentially identical: 79.3% versus 79.4% at two years (p=0.62). Median overall survival was 46.4 versus 48.0 months, respectively.
- Toxicity also showed no significant difference. Any grade 2 toxicity occurred in 11.5% versus 12.4%, grade 3 to 5 toxicity in 1.5% versus 3.9%, and grade ≥2 chest-wall toxicity in 5.0% versus 5.0% after single- versus multi-fraction SABR.
- Chest-wall abutment, rather than fractionation schedule, was the only independent predictor of both grade ≥2 chest-wall toxicity and overall grade ≥2 toxicity. Even among tumors directly abutting the chest wall, clinically relevant chest-wall toxicity remained similar between fractionation groups.
- Two possible grade 5 pulmonary events occurred, both after 48 Gy in four fractions in patients with pre-existing interstitial lung disease. Event numbers were too small to compare treatment schedules, but they reinforce the importance of caution when considering SABR in ILD.
- Treatment allocation was not randomized. Single-fraction patients had better performance status, smaller and more frequently T1 tumors, and fewer chest-wall-abutting lesions. Around half of lesions also lacked pathological confirmation, and follow-up remains relatively short for late failure and toxicity.
CLINICAL TAKEAWAY
For selected patients with peripheral early-stage NSCLC, one-fraction SABR appears capable of delivering local control and toxicity outcomes comparable with commonly used three- or four-fraction schedules. The population-based scale makes the finding highly relevant to routine practice, although the retrospective treatment-selection differences prevent a definitive equivalence claim.