KEAP1/NFE2L2 Mutations Predict Lung Cancer Radiation Resistance That Can Be Targeted by Glutaminase Inhibition

Abstract

Tumor genotyping is not routinely performed in localized non-small cell lung cancer (NSCLC) due to lack of associations of mutations with outcome. Here, we analyze 232 consecutive patients with localized NSCLC and demonstrate that KEAP1 and NFE2L2 mutations are predictive of high rates of local recurrence (LR) after radiotherapy but not surgery. Half of LRs occurred in tumors with KEAP1/NFE2L2 mutations, indicating that they are major molecular drivers of clinical radioresistance. Next, we functionally evaluate KEAP1/NFE2L2 mutations in our radiotherapy cohort and demonstrate that only pathogenic mutations are associated with radioresistance. Furthermore, expression of NFE2L2 target genes does not predict LR, underscoring the utility of tumor genotyping. Finally, we show that glutaminase inhibition preferentially radiosensitizes KEAP1-mutant cells via depletion of glutathione and increased radiation-induced DNA damage. Our findings suggest that genotyping for KEAP1/NFE2L2 mutations could facilitate treatment personalization and provide a potential strategy for overcoming radioresistance conferred by these mutations. SIGNIFICANCE: This study shows that mutations in KEAP1 and NFE2L2 predict for LR after radiotherapy but not surgery in patients with NSCLC. Approximately half of all LRs are associated with these mutations and glutaminase inhibition may allow personalized radiosensitization of KEAP1/NFE2L2-mutant tumors.This article is highlighted in the In This Issue feature, p. 1775.

Figure 1:. KEAP1/NFE2L2 mutations are predictive biomarkers of local recurrence after radiotherapy.

(A) Study design (images were produced and modified from Servier Medical Art, see Acknowledgements). (B) Recurrent mutations in patients with and without local recurrence (LR) after chemoradiotherapy (CRT) or stereotactic ablative radiotherapy (SABR). (C) Association of recurrent mutations with LR. Competing risk comparison performed using Gray’s test with multiple hypothesis testing correction. (D) Pie chart fraction of LR events occurring in tumors with KEAP1 or NFE2L2 mutations. (E) Location of KEAP1 and NFE2L2 mutations from CRT and SABR cohorts. (F) Incidence of local recurrence stratified by KEAP1/NFE2L2 mutation status for patients with stage IIB-IIIC NSCLC receiving CRT. (G) Incidence of LR stratified by KEAP1/NFE2L2 mutation status for patients with stage IA1-IIB NSCLC treated with SABR. (H) Incidence of LR stratified by KEAP1/NFE2L2 mutation status for patients with stage IA1-IIB NSCLC treated with surgical resection.

Figure 2:. Functional analysis of KEAP1/NFE2L2 mutations found in the radiotherapy cohorts identifies pathogenic and passenger mutations.

(A) Strategy for assessing KEAP1/NFE2L2 mutation functional classification using isogenic H1299 knock-out cell lines generated by CRISPR-Cas9 (images were produced and modified from Servier Medical Art, see Acknowledgements). (B) Western blot analysis for KEAP1, NFE2L2, and NFE2L2 target proteins in parental and KEAP1^NULL H1299 cells. (C) Clonogenic survival of parental and KEAP1^NULL H1299 cells after 10 Gy of ionizing radiation (n=5, **P<0.001). (D) DNA damage assessment by γH2AX foci immunofluoresence analysis 5 min after exposure to 2 Gy of ionizing radiation in parental and KEAP1^NULL H1299 cells (n=4, ***P<0.0001). (E) Clonogenic survival of parental and KEAP1^NULL H1299 cells in the presence or absence of BSO (100 nM) and/or NAC (100 mM) treated with or without 4 Gy of ionizing radiation (n=4; Student’s t-test, *P<0.01). (F) Expression of the NFE2L2 target gene HMOX1 in KEAP1^NULL cells by qRT-PCR after transfection of empty plasmid, plasmid containing wild-type KEAP1, or plasmids containing KEAP1 constructs with mutations observed in the CRT and SABR cohorts (n=4, *P<0.01) (G) Clonogenic survival of KEAP1^NULL cells transfected with plasmids containing wild-type or mutant KEAP1 constructs 24 hours before exposure to 5 Gy of ionizing radiation. Experiments were performed in groups, with empty vector and wild-type controls in each group (n=3-4; *P<0.01, ** P<0.001 compared to ‘empty’ by Student’s t-test). (H) Expression of NFE2L2 target genes in NFE2L2^NULL or KEAP1^NULL cells by qRT-PCR after transfection of plasmids containing wild-type or mutant NFE2L2 (n=4, *P<0.01). (I) Clonogenic survival of NFE2L2^NULL cells transfected with plasmids containing wild-type or mutant NFE2L2 constructs 24 hours before exposure to 3 Gy of ionizing radiation (n=3; ** P<0.001 by Student’s t-test). (J) Summary of functional assay results and mutation classification for each KEAP1 and NFE2L2 mutation found in the CRT and SABR cohorts.

Figure 3:. Pathogenic KEAP1/NFE2L2 mutations but not passenger mutations or expression analysis predict local recurrence after radiotherapy.

(A) Incidence of local recurrence after CRT or SABR in patients with KEAP1/NFE2L2 mutations stratified by functional classification. (B) Tumor volumes for patients with pathogenic KEAP1/NFE2L2 mutations (K/N^MUT). For patients who did not develop local recurrence (LR) the volume of the largest lesion is shown (*P=0.03). (C) Incidence of local recurrence in patients with pathogenic KEAP1/NFE2L2 mutations stratified by radiotherapy type. (D) Overall survival of patients in the SABR cohort stratified by presence or absence of pathogenic KEAP1/NFE2L2 mutations (WT, wild-type). (E) Overall survival of stage I-II patients from the TCGA lung adenocarcinoma and squamous cell cohorts who were treated with RT and not surgery, stratified by presence or absence of KEAP1/NFE2L2 mutations. (F) RNA-seq analysis of tumor cells from FFPE tumor biopsies of patients in the CRT and SABR cohorts (n=41). CIBERSORTx was used to deconvolve tumor cell expression (26). The heatmap depicts single sample gene set enrichment analysis (ssGSEA) scores of a previously defined NFE2L2 target gene expression signature (NFE2L2 sig.) and expression of the individual signature genes (27). (G) NFE2L2 target gene ssGSEA scores in tumor biopsies from patients in the CRT and SABR cohorts stratified by the presence or absence of pathogenic KEAP1/NFE2L2 mutations. (H) Incidence of local recurrence after CRT or SABR stratified by ssGSEA scores for NFE2L2 signature. Legend is same as in F. Stratification threshold was obtained by choosing the highest significance value by log-rank for LR based on 1,000 re-sampling iterations (I) Overall survival of patients from the TCGA lung adenocarcinoma and squamous cell cohorts who were treated with RT and not surgery, stratified by ssGSEA scores for NFE2L2 signature. The optimal cutpoint identified in H was used.

Figure 4:. Glutaminase inhibition preferentially radiosensitizes KEAP1 mutant lung cancer cells.

(A) Schematic depicting potential interaction between glutaminase inhibition and ionizing radiation in KEAP1 mutant cells (images were produced and modified from Servier Medical Art, see Acknowledgements). (B) Western blot analysis for ASCT2, NFE2L2, and HMOX1. Left: parental, KEAP1^NULL, and NFE2L2^NULL H1299 cells. Right: transient siRNA knock-down of NFE2L2 in H1975 NSCLC cells (wild-type for KEAP1 and NFE2L2). (C) Gene set enrichment analysis (GSEA) of a previously defined glutamine metabolism signature using RNA-seq data from parental and KEAP1^NULL H1299 cells. (D) As in C but using RNA-seq data from the tumor biopsies of patients in the CRT and SABR cohorts, comparing samples with and without pathogenic KEAP1 mutations. (E) Oncoprint of genes recurrently mutated in NSCLC (24) in the three cell line used for the experiments with CB-839. (F) Clonogenic survival of parental and KEAP1^NULL H1299 cells in the presence or absence of CB-839 (100 nM; 24 hour pre-treatment) and 4 Gy of ionizing radiation (n=4; *P<0.01, **P<0.001). Results were normalized against untreated cells. (G) Clonogenic survival of KEAP1^NULL and parental H1299 cells (KEAP1 wildtype) in the presence or absence of CB-839 (100 nM; 24 hour pre-treatment) and 4 Gy of ionizing radiation (n=4; *P<0.01). (H) Clonogenic survival of H1437 cells (KEAP1 wildtype) with and without siRNA knock-down of KEAP1 in the presence or absence of 2 Gy and 500 nM CB-839 (n=4; *P<0.01). (I) Clonogenic survival of parental and KEAP1 transfected A549 cells (KEAP1 mutant) in the presence or absence of CB-839 (0.1 nM; 24 hour pre-treatment) and 2 Gy of ionizing radiation (n=4; *P<0.01, **P<0.001). Results for panel F through I were normalized to untreated cells. (J) Intracellular reactive oxygen species (ROS) levels measured by DCFDA intensity via FACS in parental and KEAP1^NULL H1299 cells in the presence or absence of CB-839 (100 nM; 24 hour pre-treatment) and 2 Gy of ionizing radiation (n=4; *P<0.01). (K) Glutathione (GSH) to glutathione disulfide (GSSG) ratio in parental and KEAP1^NULL H1299 cells in the presence or absence of CB-839 (1 μM; 24 hour pre-treatment; n=4; *P< 0.05, **P< 0.01). (L) Clonogenic survival of KEAP1^NULL H1299 cells treated with or without 2 Gy of ionizing radiation and/or 100 nM CB-839 in the presence or absence of NAC (1 mM) (n=4; *P<0.01). (M) Western blot analysis for γH2AX in parental and KEAP1^NULL H1299 cells in the presence or absence of CB-839 (100 nM) treated with or without 2 Gy IR.