Abstract
Aim: To evaluate the utility of plasma circulating tumor DNA (ctDNA) using the peptide nucleic acid-locked nucleic acid (PNA-LNA) clamp method to detect epidermal growth factor receptor (EGFR) mutations in non-small cell lung cancer (NSCLC) patients who had progressed under treatment with EGFR-tyrosine kinase inhibitors (TKIs). Patients and Methods: Blood samples were collected from patients with EGFR mutation-positive NSCLC who had progressed on EGFR-TKIs between March 2016 and August 2016 at the Kyoto University Hospital. Extracted ctDNA was analyzed using the PNA-LNA clamp method. In eligible patients, tissue re-biopsy was also performed and EGFR mutation status was compared between tissue and plasma samples. Results: Thirty-one patients were enrolled in this study. Known activating EGFR mutations and the T790M mutation were detected in 18 (58%) and 5 patients (16%), respectively. Twenty-five patients underwent tissue re-biopsy. Adequate samples for mutation analysis were obtained from 21 patients and 10 patients were found to be tissue T790M-positive. Among these 10 patients, 4 patients were positive for T790M in plasma ctDNA (sensitivity 40% and specificity 100%). All patients with T790M-positive plasma ctDNA responded to osimertinib. Conclusion: Sensitivity of the PNA-LNA clamp method in detecting the plasma EGFR T790M mutation was moderate with elevated, however, specificity. Plasma EGFR T790M testing may be adequate for the initial step; however, tissue re-biopsy should be considered for plasma T790M-negative patients because of its high false-negative rate.
Despite initial responses to epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKIs), most patients eventually develop drug resistance and relapse around 12 months after treatment initiation (1-3). The EGFR T790M mutation has been reported to be a major resistance mechanism accounting for approximately 50-60% of acquired resistance to EGFR-TKIs (4). A previous study demonstrated that osimertinib, a third-generation EGFR-TKI, has clinical activity for patients with EGFR T790M-positive non-small cell lung cancer (NSCLC) (5, 6). Consequently, T790M mutation has become a new biomarker for guidance of treatment after the failure of first- or second-generation EGFR-TKIs. Although tissue re-biopsy is recommended to confirm this mutation, obtaining sufficient tissue for mutation analysis is not always easy. Thus, non-invasive genotyping of plasma circulating tumor DNA (ctDNA) has been developing as an alternative method (7-9).
To date, several studies have reported that the EGFR mutation status detected in plasma ctDNA was highly concordant with that detected in tumor tissue (10-12); however, data are still insufficient and tissue-based testing is considered to be a gold standard. This study aimed to evaluate the utility of plasma ctDNA for EGFR mutation detection using the peptide nucleic acid-locked nucleic acid (PNA-LNA) clamp method in NSCLC patients who had progressed on EGFR-TKIs.
Patients and Methods
Patients were recruited between March 2016 and August 2016 at Kyoto University Hospital, Kyoto, Japan. All patients had advanced EGFR mutation-positive NSCLC and were under previous EGFR-TKIs. Patients were excluded if they had de novo T790M mutation before EGFR-TKI treatment. Whole blood samples (14 ml) were collected in ethylenediaminetetraacetic acid (EDTA) tubes. Samples were mixed thoroughly and plasma was isolated by centrifugation at 1,700 × g for 10 min at room temperature was stored at −80°C until DNA extraction. EGFR mutations, G719X, exon 19 deletion, T790M, L858R and L861Q were analyzed in the central laboratory, LSI Medience Corporation (Kyoto, Japan) using the PNA-LNA clamp method. The laboratory procedures for mutation detection were the same as previously described (13, 14). In addition, tissue re-biopsy was also performed on eligible patients and EGFR mutation in tissue samples was analyzed by the PNA-LNA clamp method as well. Concordance of EGFR mutation status between matched samples from tissue re-biopsy and plasma ctDNA was assessed by sensitivity and specificity. Objective tumor response was evaluated in accordance with the Response Evaluation Criteria in Solid Tumors (RECIST) 1.1. Progression-free survival (PFS) was defined as the period from the date of initiation of osimertinib treatment to the date of disease progression or death due to any cause. Written informed consent was obtained from all participants. This study was approved by the ethics committee of Kyoto University (Kyoto, Japan).
Results
Patients' characteristics. Blood samples were collected from 31 patients between March 2016 and August 2016. Clinical characteristics of the 31 patients are listed in Table I. Median age was 66 (range=39-82) years. Twenty patients had never smoked. Regarding the EGFR mutation subtype, exon 19 deletion was detected in 15 patients, L858R mutation in 15 patients, while an uncommon mutation was found in one patient.
Plasma ctDNA analysis. Thirty-one plasma samples were analyzed for EGFR mutation status. Of the 31 samples, the known activating EGFR mutations and T790M mutation were detected in 18 (58%) and 5 samples (16%), respectively, whereas 13 of the 31 plasma samples were negative for both the known activating EGFR mutations and T790M mutation. Sensitivity of plasma genotyping for the known EGFR mutations was 60% for both exon 19 deletion and L858R mutation.
EGFR mutation status between tissue and plasma samples. Tissue re-biopsy samples were obtained from 25 patients, including primary lesions from 16 patients, pleural effusion from 5 patients, lung metastases from two patients, bone metastases from one patient and liver metastases from one patient. Among them, 21 samples were adequate for mutation analysis; three tissue samples did not include malignant cells and one sample was transforming to small cell lung carcinoma.
The results of known activating EGFR mutation and T790M mutation analysis of tissue samples and plasma samples from 21 patients are listed in Table II. Ten were positive for exon 19 deletion and 11 were positive for L858R mutation. Thus, the sensitivity of plasma genotyping was 60% for exon 19 deletion and 64% for L858R mutation. T790M was detected in 10 patients (48%). Of those 10 patients, four patients were T790M-positive in plasma samples (sensitivity=40%). In contrast, T790M was not detected in 11 patients (52%) in tissue samples, while all of them were T790M-negative (specificity=100%).
Comparison of plasma T790M-positive and -negative patient characteristics among the tissue T790M-positive patients. Of the 10 tissue T790M-positive patients, four patients were plasma T790M-positive and six patients plasma T790M-negative (Table III). The number of prior EGFR-TKI regimens was higher for plasma T790M-positive patients. All plasma T790M-positive patients were M1b and the serum carcinoembryonic antigen (CEA) level was higher for plasma T790M-positive patients.
Clinical response to osimertinib. Eleven patients, including one patient who was plasma T790M-positive but tissue T790M status-unknown, were administered osimertinib. Response to osimertinib is shown in Figure 1. Eight patients achieved partial response (PR) and 3 patients had stable disease (SD). No patient achieved complete response (CR) or had progressive disease (PD). The objective response rate (ORR) in T790M-positive patients for all patients, tissue-positive patients, as well as plasma-positive patients, was 73%, 70% and 100%, respectively. Although sensitivity was moderate, the response rate for plasma T790M-positive patients was 100%. The median PFS was 8.1 months (95% confidence interval (CI)=6.3-not reached (NR) (Figure 2), similar with previous reports.
Discussion
Although tissue biopsy is a standard procedure to characterize tumor genetics, tissue samples are not always available. Thus, non-invasive plasma ctDNA for EGFR mutation testing has been developing recently as an alternative method to tissue-based testing. To date, several methods of detecting EGFR mutations from plasma ctDNA have been reported (15). The PNA-LNA clamp method is a highly sensitive polymerase chain reaction (PCR) method and one of the major methods applied to assess EGFR mutations in tissue or cytology samples in Japan (12, 16); however, data for the PNA-LNA clamp method with plasma ctDNA are limited for validating the T790M mutation status in patients who progressed on previous EGFR-TKIs treatment.
In this study, the sensitivity and specificity of the PNA-LNA clamp method for T790M mutation in plasma ctDNA were 40% and 100%, respectively, and the overall concordance between tissue and plasma samples was 71%. In addition, the response rate to osimertinib in plasma T790M-positive patients was equivalent to that in tissue T790M-positive patients (100% vs. 70%). Furthermore, one of six patients who were unable to undergo tissue re-biopsy was plasma T790M-positive and responded to osimertinib. Considering these results, plasma-based testing using the PNA-LNA clamp method may be feasible as an initial step to detect T790M in patients who had relapsed after EGFR-TKI treatment and it may be acceptable to use osimertinib in plasma T790M-positive patients detected by the PNA-LNA clamp method without tissue re-biopsy. On the other hand, our study also indicated that tissue biopsy is necessary when plasma T790M is negative because of the high false-negative rate, in line with previous studies using different assays (17-19).
In the current study, all plasma T790M-positive patients had extrathoracic metastases (M1b). In a previous work, plasma T790M positivity was more frequent in M1b disease compared to M0/M1a disease (20). In addition, the serum CEA level was numerically higher for plasma T790M-positive patients than plasma T790M-negative patients in this study. These results suggest that high tumor burden or diffuse tumor spread may be reflected in the high amount of plasma ctDNA. The possibility of false-negatives should be considered, especially in patients with M0/M1a disease and normal serum CEA levels.
In conclusion, our study demonstrated the modest sensitivity and high specificity of the PNA-LNA clamp method for detecting plasma T790M in NSCLC patients previously treated with EGFR-TKIs. It also showed that plasma T790M positivity could be a predictive factor for osimertinib. Thus, plasma EGFR T790M testing by the PNA-LNA clamp method may be adequate for the initial step; however, as a result of its high false-negative rate, tissue biopsy is inevitable in plasma T790M-negative patients. As far as understanding these limitations, liquid biopsy using the PNA-LNA clamp method can be routinely used to detect T790M in NSCLC patients treated previously with EGFR-TKIs in clinical practice.
Acknowledgements
This study was supported, in part, by a Grant-in-Aid for Cancer Research from the Japanese Ministry of Education.
Footnotes
Conflicts of Interest
The Authors declare no conflicts of interest.
- Received March 15, 2017.
- Revision received March 27, 2017.
- Accepted March 31, 2017.
- Copyright© 2017, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved