<i>KEAP1</i> Mutation As a Predictive Biomarker in Checkpoint Blockade Immunotherapy

Background: Novel, powerful biomarkers are needed to determine which patients should or should not be offered immune checkpoint inhibitors (ICIs).<br><br>Methods: Using data from cBioPortal for Cancer Genomics database, we estimated the prevalence of <i>KEAP1</i> mutation, examined its predictive value in patients treated with ICIs, and investigated the characteristics of the immune system in <i>KEAP1</i>-mutant cancer.<br><br>Findings: Among 50907 patients with 35 cancers, <i>KEAP1</i> mutations were discovered in most tumor types (overall frequency, 1.34%; 95%, 1.03%-1.69%) with the highest level occurred in NSCLC. In immunotherapy, <i>KEAP1</i> mutation was associated with poor survival outcomes (hazard ratio, 1.45; 95% CI, 1.05-2.00; P = 0.006). It should be noted that an unexpected rapid death occurred early after the initiation of ICI therapy in patients with <i>KEAP1</i>-mutant cancer. Further analysis revealed that <i>KEAP1</i> mutation was associated with the downregulation of most examined immune-related features including 28 tumor-infiltrating lymphocytes, 24 immune-stimulators, 45 immune-inhibitors, 21 major histocompatibility complex molecules, 40 chemokines and 18 chemokine receptors.<br><br>Interpretation: <i>KEAP1</i>-mutant cancer is a generalized phenotype with a broader cancer spectrum than currently recognized. <i>KEAP1</i> mutation was a predictive biomarker for poor outcomes in patients treated with ICIs.<br><br>Funding Statement: This work was funded by National Natural Science Foundation of China (No. 31571417), China Postdoctoral Science Foundation (No. 2018M641862).<br><br>Declaration of Interests: The authors report no potential conflicts of interest.<br><br>Ethics Approval Statement: The study was deemed exempt from institutional board approval and patient informed consent because all data are de-identified and publicly available.