F. K. Keane1, B. Y. Yeap2, J. Lin3, Z. Piotrowska3, M. J. Khandekar4, and H. Willers5; 1Department of Radiation Oncology, Massachusetts General Hospital, Boston, MA, 2Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, MA, 3Massachusetts General Hospital Cancer Center, Boston, MA, 4Harvard Medical School, Boston, MA, 5Massachusetts General Hospital, Boston, MA
Purpose/Objective(s): The role of SBRT in the management of metastatic NSCLC has significantly expanded but its role as consolidation therapy in oncogene-driven NSCLC remains undefined. Approximately 50-60% of relapses on tyrosine kinase inhibitor (TKI) therapy in EGFR mutant NSCLC occur in original involved sites (OF). Consolidation SBRT to original sites may prevent OF, reducing the risk of subsequent distant failures (DF). We conducted a single-arm, Phase II trial of SBRT to residual sites of original disease in patients (pts) with oncogene-driven, metastatic NSCLC on first line TKI. Materials/
Methods: We enrolled pts with oncogene-driven, metastatic NSCLC with stable or responding disease to TKI within 12 months of TKI start. SBRT was delivered to residual sites of original disease, with dose and fractionation per protocol. Residual mediastinal or hilar nodal sites were excluded due to potential toxicity of SBRT to ultracentral sites. Pts held TKI during RT. The main hypothesis was that consolidation SBRT would decrease secondary distant seeding. The primary endpoint was DF frequency at 12 months after SBRT. Secondary endpoints were patterns of failure, progression-free survival (PFS), local failure (LF), overall survival (OS), and toxicity per CTCAE v4. All pts signed informed consent. The trial was registered at clinicaltrials.gov. The trial closed early due to slow accrual. Results: Between 2015 – 2021, 27 out of 30 planned pts were enrolled. Median age was 59 years (range 42 – 78). The majority were female (n=17) and white (n=20). Most had EGFR driver mutations (n=22, 54.5% on osimertinib), followed by ALK (n=4, on alectinib or lorlatinib), and ROS1 fusions (n=1, on crizotinib). Median time from TKI start to SBRT was 6.4 months (range 5.0 – 11.2). Nearly all pts received consolidative SBRT to the residual lung primary only (n = 25, 92.6%). Median dose was 44 Gy (range 35 – 50) in 5 fractions (range 4 – 6). Median follow-up was 50.4 months. Half of first failures by 12 months were DF, for a rate of 22% (95% CI: 9-40%). Overall, 18/27 pts experienced recurrences, with 10 DF and 8 OF as sites of first failure. Brain metastases were 5/10 DF, with 3/5 occurring in pts on 1st-generation TKIs. Median PFS from SBRT was 14.7 months (95% CI: 8.3–46.4). 2-year LF of irradiated sites was 12% (95% CI: 3-27%). 2-year and median OS were 88% (95% CI 68-96%) and 70.8 months (95% CI 41.8 – NR), respectively. Treatment was well-tolerated, with no grade =3 adverse events related to SBRT. Conclusion: Consolidation SBRT to predominantly the primary lung site in pts with oncogene-driven metastatic NSCLC was not associated with decrease in frequency of DF which constituted 50% of first failures at 1 year. The lack of reduction in DF suggests that SBRT to the primary lung site only is insufficient. Consideration should be given to consolidation SBRT to not only the primary site but also additional original sites of metastasis.
