Aim: A minority of patients gains advantage from immunotherapy (IO). Predictive variables of long-term benefit (LTB) are incompletely understood. Materials & methods: We retrospectively collected data about metastatic non-small-cell lung cancer patients treated with IO from April 2013 to July 2017. We defined LTB to IO as complete response (CR), partial response (PR) or disease stability as best response and maintaining it for ≥12 months. Results: Thirty-five of the 147 patients had LTB. More LTB patients than controls showed CR/PR as first and best response to IO. Only CR/PR as best response to IO retained association to LTB at multivariate analyses. Conclusion: Objective response appears as a central factor for LTB from IO.