Background: Immunotherapy (IO) provides a significant benefit for a subgroup of non-small cell lung cancer (NSCLC) patients. Radiotherapy (XRT) might enhance the efficacy of IO. We evaluated the impact of the specifics of XRT treatments on the OS of IO-treated NSCLC patients. Methods: Metastatic NSCLC patients treated with IO were retrospectively identified. Parameters included demographics, tumor characteristics, IO and XRT details. Correlation between the parameters and OS was tested with Cox regression. Results: 453 patients were included. No XRT was given to 167 (36.9%) patients, whereas XRT prior and after IO had 182 (40.2%) and 104 (22.9%) patients, respectively. XRT total doses between 30 and 40 Gy had better overall survival (OS) compared to non-irradiated patients (hazard ratio (HR) 0.5, 95% CI 0.25–1.0, p = 0.049). Worse outcome was seen with total doses ≤ 10 Gy (HR 1.67, 95% 1.13–2.5, p = 0.01), XRT fractions of 4.1–8 Gy (HR 1.48, 95% CI 1.05–2.1, p = 0.027) and XRT to the bone (HR 1.36, 95% CI 1.01–1.8, p = 0.04). Several clinical parameters correlated with OS in the univariate analysis of the IO-treated patients. While, in the multivariate analysis, only ECOG-PS, treatment line, type of IO, albumin and NLR remained statistically significant. Conclusion: Specific doses, fractions and sites of XRT correlated with the OS of IO-treated NSCLC patients in the univariate analysis, although not in the multivariate analysis.