Pembrolizumab has recently emerged as a PD-1 blockade immunotherapy treatment for lung cancer. It is critical that such treatment strategies for lung cancer should be chosen not only on the basis of histopathological features and the expression of targetable cell surface proteins (such as PD-1), but should rather be selected based on other determinants of treatment success or risk factors for poor prognosis. One method to forecast cancer trajectory is the identification of biomolecular signatures such as microRNAs (miRNAs), non-protein-coding RNA molecules that play a regulatory role in gene expression by modulating the translation or stability of messenger RNA. To find out which miRNAs have an important influence on anti-PD-1 treatment outcomes, we evaluated miRNA levels in sera from 38 lung cancer patients undergoing 3 months of pembrolizumab treatment. We selected a panel of miRNAs previously shown to be involved in lung cancer or PD-1 signaling and performed qPCR analysis. Overall, we observed a significant decrease in the levels of miR126-5p (4-fold), let-7a (5-fold), miR133a-3p (4-fold), miR3615 (2-fold), miR4516 (3-fold), miR16 (3-fold), miR34c-5p (2-fold), miR20b-5p (5-fold), miR106b-5p (5-fold), miR146a-5p (3-fold) and miR181b-5p (3-fold) in response to treatment indicating effectiveness of immunotherapy. Within our selected panel of miRNAs, we identified two markers relevant to cancer prognosis: miR-217, which is negatively associated with patient survival, and let-7a, which is positively associated with patient survival. Our findings suggest that circulating miRNAs can be used for future treatment evaluation and lung cancer prognosis, with potential as therapeutic targets.