5-year survival and safety of stereotactic ablative radiotherapy in unresectable locally advanced non-small cell lung cancer patients unfit for concurrent radio-chemotherapy:focus on patterns of local recurrence from the START-NEW-ERA non-randomized phase II trial.

Purpose: In the early analysis of XXXX phase II trial, SAbR had optimal local control (LC) and promising OS without ≥G3 toxicity in unresectable LA-NSCLC patients unfit for concurrent chemo-radiotherapy. We report the 5-year outcomes with focus on patterns of local recurrence (LR).

Methods and materials: SAbR was delivered by V-MAT to primary tumor (T) and regional nodes (N) based on PET-CT. When thoracic failures occurred, we matched the SAbR planning with PET-CT images to accurately assess the LR location, precisely to determine in-field versus out-field recurrence, looking at the specific isodose line within which the LR occurred.

Results: 50 unresectable LA-NSCLC patients unfit for concurrent ChT-RT were enrolled. Median dose was 45 Gy and 40 Gy in 5 daily fractions to T and N, respectively. After a median follow-up of 72 months (range, 10-108) the 3-, 5-year PFS rates were 26 ± 6%, 26 ± 6%. The 3-, 5-year OS rates were 70 ± 6%, 46 ± 7%. No patients developed ≥G3 late toxicities. The 3-, 5-year LR-FS rates were 64 ± 7%, 64 ± 7%, respectively. Multivariate analysis revealed SCC (p 0.016) and SAbR dose <40 Gy (p 0.044) as significant predictors of LR. 17 patients experienced LR; 14/17 had SCC, 13/17 IIIA or IIIB. Tumors that recurred were all centrally or ultra-centrally located. The LR was higher in patients receiving 35 Gy compared to those receiving at least 40 Gy (p= 0.037).

Conclusions: The 5-year outcomes of XXXX trial confirm robust and sustained benefit in terms of safety and effectiveness of SAbR in LA-NSCLC patients unfit for concurrent chemo-radiotherapy.