Background: Immunologic checkpoint blockade with antibodies that target CTLA-4 or PD-1/PD-L1 have demonstrated promise in a variety of malignancies. However, the treatment response rate of these immunologic checkpoint blockades remains low. Identifying predictive biomarkers to assist patient selection for immunotherapy have become a priority in both clinical and research settings. Methods: Mutations in patients who responded to immunotherapy were identified by Next-Generation Sequencing (NGS). Relationship between mutation of PRKDC, mutation load and known immune biomarkers were analyzed using datasets from The Cancer Genome Atlas (TCGA). Following up, the PRKDC protein expression was evaluated in 439 gastric cancer patients by immunohistochemical staining and their MSI statuses were evaluated by PCR. Results: We first identified PRKDC mutations in two responders to immune checkpoint therapy (1 HCC, 1 gastric cancer). From published literature, we further discovered that 66.7% (2/3) of lung cancer patients and 63.6% (7/11) of melanoma patients whose tumor harbored PRKDC mutation and responded to immunotherapy. Most of these mutations detected in responders were either truncating or located in functional domains. Further analysis showed that PRKDC mutation is significantly associated with high mutation load in cervical cancer, colon adenocarcinoma, head and neck squamous cell carcinoma, lung adenocarcinoma, gastric adenocarcinoma and endometrial cancer (p= 0.008, p= 0.0108, p= 0.0166, p= 0.0183, p< 0.001 and p< 0.001, respectively). Interestingly, gastric cancer patients harboring PRKDC mutations or with MSI-H demonstrated significantly higher gene expression in PDL1, TIM3, LAG3, IFNG, CXCL9, CXCL10, GZMA and PRF1, compared to MSS patients (p= 0.0016, p= 0.0142, p= 0.0017, p= 0.0034, p= 0.0118, p< 0.0001, p= 0.0001, p< 0.0102, respectively). Finally, we discovered low expression of PRKDC was a poor prognostic factor and significantly correlated with MSI-H in gastric cancers. Conclusions: PRKDC may be a potential biomarker that can identify responders to immune checkpoint inhibition.