(P09) Combined Use of Radiosurgery With Concurrent PD-1/PD-L1 Inhibition For Metastatic Brain Lesions of NSCLC

Abstract

Lung cancer is the leading cause of cancer death worldwide. The FDA first approved targeted PD-1/PD-L1 therapy for the treatment of advanced NSCLC in early 2015, and immunotherapies (IT) are becoming common in the setting of metastatic NSCLC. Some have hypothesized synergistic local and distant effects between stereotactic radiotherapy and IT, however data supporting safety and efficacy of their concurrent use has yet to be reported in the setting of SRS for brain metastasis from NSCLC. We evaluated overall survival (OS), time to intracranial progression (ICP), and acute adverse events (AEs) within our institution retrospective experience of PD-1/PD-L1 inhibitors and stereotactic radio surgery (SRS) for metastatic brain lesions of NSCLC. This study included all patients with brain metastases from NSCLC treated concurrently or sequentially with SRS and either PD-1 or PD-L1 inhibitors at our institution from October 2014 - June 2016. SRS was delivered using the Pefexion - GammaKnife. The time to ICP was measured from the time of SRS until radiological evidence of ICP by MRI. OS was calculated from the time of the first SRS treatment until death. Concurrent treatment was defined as receiving SRS within 30 days before or after immunotherapy. Comparison of these endpoints was made against patients who received radiation greater than 90 days prior to PD-1/PD-L1 inhibitors defined as non-concurrent or sequential therapy. Comparisons of time to ICP and OS were performed using the log-rank test and Kaplan-Meier estimated survival estimates. We identified twelve patients, who were treated with concurrent PD-1/PD-L1 inhibitors and SRS. Of these 12 patients, 2 received an additional SRS treatment for brain metastases while still receiving IT. Six of the 12 patients received SRS for 1 brain lesion while on IT; the other six patients had a range of 2-6 lesions treated in one session. Sixteen patients were identified that had received IT greater than 90 days after SRS intervention. Both time to ICP and OS were performed on a patient only basis calculating time from first SRS intervention. The median follow up was 10 months. OS was not statistically different between concurrent vs. non-concurrent groups with a median OS of 15 (1-20) months vs. 23.5 (7-54) months respectively (p=0.43). Time to ICP did not differ between our groups with time to ICP of 5.8 (1.1 - 20.5) months with concurrent therapy vs. 19.4 (1.4 - 54) months with non-concurrent therapy (p=0.70). Eight of 12 patients in the concurrent therapy group denied any acute AES, two patients reported a headache, one reported dizziness and another patient reported numbness. Our data suggest that concurrent use of SRS for brain metastases of NSCLC with PD-1/PD-L1 inhibitors may not be associated with an increase in acute AEs nor does it affect the time to ICP or OS. Time to ICP or OS after SRS were similar between patients who received concurrent vs. non-concurrent IT in our analysis and our reported median survival times and time to ICP is consistent with historically reported outcomes for patients treated with SRS alone.