Evaluation of Initial Metastatic Tumor Location and Radiation Response to Determine Outcomes in Patients Who Received Combination Stereotactic Body Radiotherapy and Immunotherapy for NSCLC.

Abstract

PURPOSE/OBJECTIVE(S)\nCombination SBRT and immune checkpoint inhibition is safe and efficacious for metastatic NSCLC. It is unknown whether the site of metastatic disease at diagnosis or SBRT response can be used as predictors of PFS or OS. We report follow-up from a prospective phase I clinical trial examining predictors of OS and PFS based on initial metastatic tumor disease location, as well as SBRT response.\n\n\nMATERIALS/METHODS\nThirty-seven patients with treatment-naïve metastatic NSCLC were reviewed. Patients received SBRT to doses of 30-50 Gy in 3-5 fractions to 1-6 metastases, as well as concurrent (immunotherapy with SBRT) or sequential (immunotherapy after completion of SBRT) nivolumab/ipilimumab on a phase I clinical trial at a single institution. All metastatic lesions were evaluated and the sites of initial disease, including liver, bone, or thoracic only metastases, were evaluated to determine if those sites influenced OS or PFS. Treated metastases control (TMC) was defined as lack of progression for irradiated metastases. Landmark analysis was used to assess correlation of TMC and OS. The Kaplan-Meier method was used to analyze OS and PFS.\n\n\nRESULTS\nThirty-seven patients with 120 evaluable metastases were enrolled with a median follow-up of 17.0 months. Patients who had liver metastases at baseline (n\u202f=\u202f8) were found to have a worse PFS (log-rank P\u202f=\u202f0.048), but no difference in OS (log-rank P\u202f=\u202f0.243) when compared to patients who did not have liver metastases at baseline (n\u202f=\u202f29). There was a trend toward significance for better PFS in patients who had thoracic only metastases at baseline (n\u202f=\u202f6) compared to patients that had extra-thoracic metastases (n\u202f=\u202f31) at baseline (Gehan-Breslow-Wilcoxon P\u202f=\u202f0.061); however, there was no difference in OS. There was no statistically significant difference in PFS or OS for patients with bone metastases (n\u202f=\u202f9) at baseline compared to patients who did not have bone metastases at baseline (n\u202f=\u202f28). One- and two-year TMC were 93.8% and 91.7%, respectively. TMC corresponded to better OS (log-rank P\u202f=\u202f0.040).\n\n\nCONCLUSION\nLiver metastases present at diagnosis in metastatic NSCLC correlate with worse PFS. Treated metastases control rates also correlate with OS. These metrics may help predict outcomes and inform whether additional aggressive therapy is required for patients with metastatic NSCLC.