Liquid biopsy uncovers distinct patterns of DNA methylation and copy number changes in NSCLC patients with different EGFR-TKI resistant mutations

Abstract

Targeted therapy with tyrosine kinase inhibitors (TKI) provides survival benefits to a majority of patients with non-small cell lung cancer (NSCLC). However, resistance to TKI almost always develops after treatment. Although genetic and epigenetic alterations have each been shown to drive resistance to TKI in cell line models, clinical evidence for their contribution in the acquisition of resistance remains limited. Here, we employed liquid biopsy for simultaneous analysis of genetic and epigenetic changes in 122 Vietnamese NSCLC patients undergoing TKI therapy and displaying acquired resistance. We detected multiple profiles of resistance mutations in 51 patients (41.8%). Of those, genetic alterations in EGFR, particularly EGFR amplification (n = 6), showed pronounced genome instability and genome-wide hypomethylation. Interestingly, the level of hypomethylation was associated with the duration of response to TKI treatment. We also detected hypermethylation in regulatory regions of Homeobox genes which are known to be involved in tumor differentiation. In contrast, such changes were not observed in cases with MET (n = 4) and HER2 (n = 4) amplification. Thus, our study showed that liquid biopsy could provide important insights into the heterogeneity of TKI resistance mechanisms in NSCLC patients, providing essential information for prediction of resistance and selection of subsequent treatment.

Figure 1

The heterogeneity of mutation profiles of 122 Vietnamese NSCLC patients with acquired resistance to first and second generation EGFR TKI. (A) Distribution of TKI resistance associated mutations of 122 Vietnamese NSCLC. (B) Frequencies of EGFR sensitizing mutations (del19, L858R, or rare EGFR mutations) in different resistance mutation profiles. (C) Variant allele frequency (VAF, %) of EGFR sensitizing mutations in different resistance mutation profiles. Data are presented as median VAF, each data point represents one patient. **p < 0.01; ***p < 0.001 (one-way ANOVA, Kruskal–Wallis test).

Figure 2

EGFR dependent (on-target) and independent (off-target) resistance mechanisms showed distinct landscapes of genome-wide methylation and copy number changes. (A, C) Methylation density (MD) (A) and DNA copy number (C) alterations for each 1 Mb bin along the chromosomes (chr1-22). Cases with T790M (n = 5), EGFR amplification (n = 3) and T790M and EGFR amplification co-occurrence (n = 3) were grouped into the on-target group while HER2 amplification (n = 4) and MET amplification (n = 4) cases were grouped into the off-target group. Healthy individuals (n = 20) were used as a reference group to determine the baseline levels of MD and CNA. (B, D) Percent of hypo-methylated bins (B) and CNA bins (D) in cases with different resistance mutation profiles grouped into the on-target group or the off-target group. The component bar graphs showed the comparison between the two groups. Data are shown as mean ± SEM. **p < 0.01, ****p < 0.0001 (Mann–Whitney test). (E) Linear regression and Pearson’s correlation between proportion of hypo-methylated bins and CNA bins in 19 patients from both on-target and off-target groups.

Figure 3

Accumulation of genetic and epigenetic changes in cases with EGFR amplification correlated with the duration of response to TKI treatment. (A–C): Linear regression and Pearson’s correlation between proportion of CNA bins (A), hypo-methylated bins (B), EGFR amplification scores (C) and time to treatment resistance (TTTR) in 6 cases with EGFR amplification including 3 cases with co-occurring T790M.

Figure 4

EGFR dependent resistance mechanisms displayed markedly high levels of hyper-methylation in the transcriptional regions of genes involved in regulation of differentiation. (A) Heat-map showed differentially methylated regions (DMRs) among 450 target regions mapped to regulatory regions of cancer related genes across samples within the on-target group (T790M, T790M-EGFR amplification and EGFR amplification) and off-target group (HER2 and MET amplification). Healthy subjects were used as a reference group to determine the basal levels of methylation at target regions. Heatmap visualization was analyzed with ComplexHeatmap package version 2.8.0 (https://www.bioconductor.org/packages/release/bioc/html/ComplexHeatmap.html). Hierarchical clustering was done using Euclidean distance. Red in the scale bar represents a high level of methylation. (B) KEGG and Reactome pathway enrichment analysis using g: Profiler (https://biit.cs.ut.ee/gprofiler) for genes associated with 202 DMRs showing hyper-methylation in the on-target group compared to the off-target group.