Mutational Status Can Indicate Drug Resistance

[CaroleHammett]

Source: http://www.oncologystat.com/journals/journal_scans/Detection_of_Mutations_in_EGFR_in_Circulating_Lung_Cancer_Cells.html

Detection of Mutations in EGFR in Circulating Lung Cancer Cells

CTCs Reveal EGFR Mutational Status During NSCLC Treatment

S Maheswaran, LV Sequist, S Nagrath, Ulkis, B Brannigan, CV Collura, E Inserra, Diederichs, AJ Iafrate, DW Bell, S Digumarthy, A Muzikansky, D Irimia, J Settleman, R Tompkins, TJ Lynch, M Toner, DA Harber, 2008 Jun 2 [EPub ahead of print: OncologySTAT™], New England Journal of Medicine

Mutational status can be indicative of drug resistance during treatment with epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs). Biopsies usually provide insufficient material for baseline molecular analysis, however, and genetic changes in epithelial tumor cells are not usually monitored during treatment. Evaluation of circulating tumor cells (CTCs) in the plasma of patients with metastatic disease is also limited by insensitive detection methods. Researchers at Massachusetts General Hospital (MGH) in Boston recently developed the CTC-chip, a microfluidic device that uses 78,000 EpCAM (epithelial cell adhesion molecule)-coated microposts to isolate, quantify, and analyze plasma CTCs. This device has the capability of isolating a pure sample of 132 CTCs/mL in patients with metastatic cancers. Thus, the technology could provide a sufficient quantity of CTCs to potentially be used to measure tumor response and the purity of the isolated cells enables repeat analysis of molecular markers.

Activating mutations in the EGFR gene are associated with response to TKIs in patients with non-small-cell lung cancer (NSCLC). Relapse, common after 1 year of treatment, is associated with an acquired secondary EGFR mutation, T790M, which blocks drug binding. The investigators evaluated the performance of CTC analysis of EGFR mutations in plasma samples from patients with NSCLC. The findings demonstrated sufficient reproducibility, CTC yield, and sample purity for accurate molecular analysis.

Patients recruited into the study fell into 1 of 2 separate groups. Group A patients (n = 31) were being treated at MGH and supplied blood samples. Group B patients from a multicenter study of gefitinib were included to increase the number of biopsy specimens (n = 15). Comparison of EGFR mutational analysis of blood, free plasma, and paraffin-embedded tissue was performed using the Scorpion Amplification Refractory Mutation System (SARMS), standard DNA sequencing, and both.

Spearman’s correlation was used to determine the relationship between CTCs and tumor burden. Fisher’s exact test was used to compare mutations across populations. Cox proportional hazards modeling was used for multivariate analysis of T790M status and Kaplan-Meier curves of progression-free survival (PFS).

For identification of CTCs, 23 blood samples were analyzed, 5 from treatment-naïve patients, 10 from gefitinib- or erlotinib-treated patients, and 8 from patients treated with chemotherapy (plus a TKI). Samples from 4 patients with wild-type EGFR were also included. CTCs were isolated from all patient samples (median, 74 cells/mL). The CTC levels did not correlate with tumor volume (Spearman’s coefficient = -0.028; P = .88), which suggested that additional factors are related to plasma CTC volume.

For detection of EGFR mutations in tumors, the SARMS assay was first validated using 26 paraffin-embedded NSCLC tumor samples with confirmed EGFR mutations and 8 wild-type specimens. Sensitivity and specificity were 96% and 100%, respectively. The allele-specific SARMS assay detected a rare EGFR mutation not detected by standard sequencing and also low-level T790M expression in 10 of 26 patients (38%). T790M was associated with significantly shorter PFS, a median of 7.7 months vs 16.5 months for patients not carrying the T790M mutation (hazard ratio = 11.5; 95% confidence interval; 2.94–45.1; P < .001).

For the detection of EGFR mutations in CTCs, EGFR mutations in CTCs were compared with those in tumor specimens. EGFR mutation-carrying CTCs were identified in 19 of 20 (95%) blood samples tested. T790M mutations were identified in 2 of 6 (33%) patients who responded to TKI treatment and 9 of 14 (64%) patients who had progressive disease (P = .34). Additional analysis was performed to compare the accuracy of EGFR detection in purified CTCs vs free plasma DNA (n = 18). EGFR mutations were detected in 17 of 18 (94%) CTC samples but in only 7 of 18 (39%) plasma DNA samples. In another analysis of 12 patients providing samples of primary tumor, plasma, and CTCs, EGFR mutations were detected in 11 of 12 (92%) CTC samples but in only 4 of 12 (33%) plasma DNA samples (P = .009).

Serial measurements were carried out in 4 patients who had confirmed EGFR mutations after initiation of gefitinib therapy. A substantial decline in CTC level was associated with gefitinib treatment in all cases. Progressive disease was associated with an increase in CTC level. A strong correlation existed between radiographic assessment of tumor volume and the relative number of CTCs during the course of gefitinib treatment. During this time, the EGFR genotype evolved, and activating EGFR mutations, as well as low-level expression of T790M, emerged.

In summary, this group of experiments demonstrated that the CTC-chip can be used to isolate CTCs of adequate purity and quantity for molecular analysis during TKI treatment of metastatic NSCLC. The findings showed correlations between the number of CTCs and tumor response, and also between the presence of T790M in pretreatment biopsy specimens and PFS. Sequence analysis of EGFR mutations in CTCs suggested that the isolated cells constitute a relatively pure tumor-derived cell population, which may have potential use in diagnostics and therapeutic intervention. Optimization and automation of CTC-chip technology are necessary to enable its large-scale use in clinical trials.

This summary was written by the OncologySTAT editorial team

Submitted by Carole