The Quantification Of EGFR Mutation Abundance And Its Association With The Efficacy Of EGFR-TKI Treatment For Advanced Non-small-cell Lung Cancer

ObjectiveEpidermal growth factor receptor (EGFR) mutation status is the most important biomarker of EGFR-tyrosine kinase inhibitor (EGFR-TKI) response in non-small-cell lung cancers (NSCLCs). However, the clinical efficacy of these drugs varies among individuals with EGFR mutation positive. So far, molecular predictor of EGFR-TKI efficacy in EGFR mutation-positive NSCLC has not been well defined. In our study, we aimed to verify the existence of EGFR mutation abundance, quantify such abundance in patients’ tumor tissues, and analysis its association with the effect of EGFR-TKI treatment.MethodsGenomic DNA was extracted from NSCLC patients’ tumor samples. We detected EGFR mutation status by amplification refractory mutation system (ARMS) and direct sequencing simultaneously. Then the following detection was performed:①Two cell lines NCI-H1650and NCI-H1975carrying exon19deletion and L858R mutation were introduced to our study. A standard curve was created by quantitative-PCR (q-PCR) and mutant EGFR copies as well as total EGFR-DNA amounts were quantified.②Fluorescence in situ hybridization (FISH) and cloning sequencing were used to analysis the EGFR gene amplification. The number of mutant EGFR clones and wild-type clones were evaluated, and the amplification ratio was defined as the value of mutant clones/wild-type clones.③The expression of mutant EGFR proteins was evaluated by immunohistochemistry (IHC) using antibodies specific for E746-A750deletion in exon19and L858R mutation in exon21. Patients enrolled in our study were routinely followed up, and the effect of TKI treatment was evaluated as well.ResultsPart one:The EGFR mutation abundance at molecular level was calculated by the amount of mutant copies as well as total EGFR copies from q-PCR analysis, and further corrected by the tumor cell percentage evaluated from HE staining. There was a good correlation between molecular abundance and the amplification ratio gained from cloning sequencing, which indicated the reliability of such quantification (Person value0.61, P<0.001). Molecular abundance of each sample was corrected by the corresponding amplification ratio to get the percentage of tumor cells carrying mutant EGFR, which was called the cellular abundance. We made a comparison between such cellular abundance and the proportion of stained tumor cells in mutation specific IHC analysis.18(85.7%) out of21specimens with higher cellular abundance also showed a relatively higher proportion of stained tumor cells. And in9cases harboring a low cellular abundance,6(66.7%) of them also gained a relatively low proportion of stained tumor cells. The statistical results revealed that the cellular abundance was moderately consistent with the proportion of stained tumor cells in each sample (Kappa value0.52, P=0.004).Part two:There were a total of129patients enrolled in this study.74samples were detected EGFR mutation positive by both ARMS and direct sequencing, and categorized into high abundance group. Another12samples were detected mutation positive by ARMS but negative by direct sequencing, and categorized into low abundance group. The median progression-free survival (PFS) of patients in high abundance group was significantly longer than that of patients with low abundance (17.0versus5.0months, P<0.001). In addition, based on the calculated molecular abundance, patients were categorized into two groups by a cut-off value of20%. Patients with higher molecular abundance gained a significantly longer PFS than those with relatively low abundance (median,20.0versus9.5months, P=0.003). With regard to the type of EGFR mutation, we found the difference in PFS between patients with high and low molecular abundance was apparent for the L858R mutation in exon21(median,24.0versus9.5months, P=0.002). But in the cases with the deletion in exon19, such difference was not significant (median,17.0versus18.0months, P=0.870).Based on the quantification result of cellular abundance, patients were categorized into two groups by a cut-off value of20%. The median PFS of patients in high abundance group was significantly longer than that of patients in low abundance group (20.0versus10.0months, P=0.013). The objective response rate (ORR) was69.9%in high abundance group, which was better than the ORR (40%) in the low one. With regard to the patients carrying exon19deletion, the difference in PFS between patients with high and low cellular abundance was statistically significant (median,19.0versus10.0months, P=0.035). As for cases with L858R mutation in exon21, patients with high cellular abundance gained a much longer PFS than those with low abundance (median,22.0versus10.0months, P=0.072), although no statistical significance was revealed for the reason of limited sample size. Results of mutation specific IHC revealed the percentage of stained tumor cells was significantly associated with the effect of EGFR-TKI treatment (median PFS,19.0versus8.0, P=0.002), while the evaluation of staining intensity didn’t show such statistical association (median PFS,18.0versus18.0, P=1.000). Thus, a further verification was made that the cellular abundance was correlated with the efficacy gained from EGFR-TKI therapy.ConclusionIn the current study, we quantify the abundance of EGFR mutation at both molecular and cellular level by ARMS combining direct sequencing, quantitative-PCR, FISH, cloning sequencing and mutation specific IHC. Each method has its own advantages and drawbacks, and their detecting results are verified by each other. The quantification of cellular abundance is less influenced by EGFR gene amplification and is more sensible for predicting the effect of EGFR-TKI treatment. In addition, EGFR mutation status is proved to be heterogeneous within single tumor tissue and the abundance of EGFR mutation varied among individuals. The results of our study also demonstrate that cellular abundance may predict the efficacy of EGFR-TKI therapy.