Non-classic EGFR mutations in a cohort of Dutch EGFR-mutated NSCLC patients and outcomes following EGFR-TKI treatment

BACKGROUND

Data on non-small-cell lung cancer (NSCLC) patients with non-classic epidermal growth factor receptor (EGFR) mutations are scarce, especially in non-Asian populations. The purpose of this study was to evaluate prevalence, clinical characteristics and outcome on EGFR-TKI treatment according to type of EGFR mutation in a Dutch cohort of NSCLC patients.

METHODS

We retrospectively evaluated a cohort of 240 EGFR-mutated NSCLC patients. Data on demographics, clinical and tumour-related features, EGFR-TKI treatment and clinical outcome were collected and compared between patients with classic EGFR mutations, EGFR exon 20 insertions and other uncommon EGFR mutations.

RESULTS

Classic EGFR mutations were detected in 186 patients (77.5%) and non-classic EGFR mutations in 54 patients (22.5%); 23 patients with an exon 20 insertion (9.6%) and 31 patients with an uncommon EGFR mutation (12.9%). Median progression-free survival (PFS) and overall survival (OS) on EGFR-TKI treatment were 2.9 and 9.7 months, respectively, for patients with an EGFR exon 20 insertion, and 6.4 and 20.2 months, respectively, for patients with an uncommon EGFR mutation. Patients with a double uncommon EGFR mutation that included G719X/L861Q/S768I had longer PFS and OS on EGFR-TKI treatment compared with patients with a single G719X/L861Q/S768I EGFR mutation (both P=0.02).

CONCLUSIONS

In our Dutch cohort, prevalence and genotype distribution of non-classic EGFR mutations were in accordance with previously reported data. The PFS and OS on EGFR-TKI treatment in patients with an uncommon EGFR mutation were shorter compared with patients with classic EGFR mutations, but varied among different uncommon EGFR mutations.

Abstract 4285: Structural variant breakpoint detection in advanced colorectal cancer

Introduction

Development of colorectal cancer (CRC) is accompanied by genomic alterations that drive tumor initiation and progression. Gains and losses of large chromosome segments result in DNA copy number alterations and subsequently quantitative changes in mRNA and protein expression levels. Interestingly, the accompanying chromosome breakpoints represent structural variants (SV) that may affect gene architecture and thereby normal gene function.

Aim

The aim of this study was to identify recurrent SV breakpoints in advanced CRC.

Methods

Previously a series of 352 advanced CRC samples from CAIRO and CAIRO2 clinical studies [Koopman et al. Lancet 2007; Tol et al. N Engl J Med 2009] was characterized for genome-wide DNA copy number alterations. DNA from formalin-fixed paraffin-embedded (FFPE) tumor and patient-matched normal tissue was subjected to high-resolution array-Comparative Genomic Hybridization (Agilent 180K arrays). Using these data, we now determined the prevalence of recurrent breakpoints in genes in CRC by computational analysis. In addition, multiplexed amplicon analysis involving 48 cancer-related genes (Illumina TruSeq Amplicon Cancer Panel) was applied to determine mutation frequencies. Multi-Dendrix was used to identify modules of (mutually exclusively mutated) cancer driver genes.

Results

We identified 748 recurrent SV breakpoints in genes (FDR<0.1). The highest frequency of recurrent breakpoints was detected in MACROD2, in up to ∼40% of CRC samples. Most recurrent breakpoints occurred in less than 5% of all tumors, and have not been reported before in CRC. Mutation frequencies in APC, TP53 and KRAS were conform expectations (60%, 58%, and 48%, respectively). Multi-Dendrix analysis revealed modules of cancer driver genes that included both the commonly mutated CRC cancer genes as well as genes with recurrent breakpoints, suggesting that several of the SV breakpoint genes are candidate drivers of the carcinogenic process.

Conclusions

We were able to pinpoint the prevalence of 748 recurrent SV breakpoint regions in genes using array-CGH data from 352 CRC samples. Moreover, our studies revealed several breakpoints in genes that are mutually exclusive with the commonly mutated APC, KRAS, and TP53 genes, and therefore represent novel candidate cancer driver genes. Further studies are required to investigate their functional and clinical significance.

Citation Format: E van den Broek, O Krijgsman, D Sie, JC Haan, M Komor, J Traets, DAM Heideman, MA van de Wiel, ID Nagtegaal, CJA Punt, B Carvalho, B Ylstra, GA Meijer, RJA Fijneman. Structural variant breakpoint detection in advanced colorectal cancer. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 4285. doi:10.1158/1538-7445.AM2014-4285