9548 Background: In metastatic non-small cell lung cancer (mNSCLC), the clinical trials NCT02492568 and NCT02444741 are the only known randomized comparisons of pembrolizumab alone versus pembrolizumab combined with radiation therapy (RT). When the trials were analyzed individually, some potential benefit was observed in the combination therapy group, but the relatively small sample size of each trial limited the detection of potential differences in response rates and outcomes. Hence, we perform a pooled analysis of these two randomized trials to validate and explore whether RT improves mNSCLC patient responses to immunotherapy. Methods: This was a pooled analysis of two randomized trials (NCT02492568 and NCT02444741) of pembrolizumab with or without RT for mNSCLC. Endpoints included the out-of-field overall response rate (ORR) and disease control rate (DCR), progression-free survival (PFS), overall survival (OS), and subgroup analysis of the different RT schemes. Results: In all, 131 patients were analyzed (n = 66 pembrolizumab; n = 65 pembrolizumab/RT (iRT)). ORR was 21% in the pembrolizumab arm vs. 38% in the iRT arm (p = 0.01); DCR was 53% in the pembrolizumab arm vs. 67% in the iRT arm (p = 0.0009); PFS was 4.4 m vs 8.3 m (p = 0.046); and OS was 9.2 m vs 19.2 m (HR 0.66; p = 0.040). Ablative RT (24Gy/3 fractions and 50Gy/4 fractions) had better ORRs of 48% and 54%, respectively, compared to 18% for non-ablative RT (45Gy/15 fractions) and 20% for pembrolizumab alone (p < 0.05, respectively). Conclusions: The addition of RT to immunotherapy significantly increased the ORR of unirradiated lesions and was additionally associated with significant improvements in PFS and OS. Ablative RT was associated with response rates significantly higher than those of non-ablative RT, possibly due to a detrimental effect of non-ablative RT on ALC. These hypothesis-generating findings require dedicated, large-volume, and randomized studies for corroboration. Clinical trial information: NCT02492568 and NCT02444741 .