213P - Results of the second pilot external quality assurance scheme for somatic EGFR mutation testing in non-small-cell lung cancer (NSCLC)

Date 30 September 2012
Event ESMO Congress 2012
Session Poster presentation II
Topics Biomarkers
Non-Small-Cell Lung Cancer, Metastatic
Presenter Nicola Normanno
Authors N. Normanno1, S. Patton2, F.H. Blackhall3, S. Murray4, K. Kerr5, M. Dietel6, M. Filipits7, S. Benlloch8, R.A. Stahel9, E. Thunnissen10
  • 1Dept. Biologia Cellulare E Bioterapie, Istituto Nazionale Tumori di Napoli, 80131 - Napoli/IT
  • 2Genetic Medicine, St Mary’s Hospital, Manchester/UK
  • 3Medical Oncology Department, Christie Hospital NHS Trust, M20 4BX - Manchester/UK
  • 4Genetics, GeneKOR, Athens/GR
  • 5Department Of Pathology, Aberdeen Royal Infirmary, Aberdeen/UK
  • 6Surgical Pathology, Charité, Humboldt-Universität zu Berlin, Berlin/DE
  • 7Medicine, Medical University of Vienna, Vienna/AT
  • 8Pangaea Biotech, USP Dexeus University Institute, Barcelona/ES
  • 9Oncology, University Hospital Zurich, Zurich/CH
  • 10Pathology, VU University Medical Center, Amsterdam/NL



The External Quality Assurance (EQA) process aims at establishing performance levels and may identify systematic errors in methodology that may not be revealed by internal QA processes. The European Molecular Genetics Quality Network (EMQN), the European Society for Pathology (ESP), the European Thoracic Oncology Platform (ETOP) and the European Society of Medical Oncology (ESMO) with other leading European groups collaborated in a pilot EQA scheme for somatic EGFR gene mutational analysis in NSCLC.

Material and methods

Samples generated from cell lines mimicking clinical fractions of neoplastic cell content were validated by 5 laboratories and then provided to participating laboratories. Each sample was supplied with a mock clinical case. Participating laboratories registered with the EMQN, performed the analysis using their usual method(s), and submitted their results within a 6 week timeframe. Anonymous results were assessed and made available to all participants.


Of a total of 117 labs from 30 countries registered, 91 labs returned results. Sequencing and the DxS Therascreen kit were the main methodologies used by the participants. The standard of genotyping was worrying, with a significant number of genotyping errors made. Only 42 (46.1%) participants reported the correct result for all 10 samples. The majority of errors were false-negative results (82.2%); however, 15.1% of errors were due to false-positive results and 2.7% to a combination of both errors. There were 42 (4.8%) analytical failures of which 88.1% were distributed across 6 laboratories. The performance of 18 (19.8%) labs fell below the criteria for acceptable performance (score <18 points). The standard of report formats was acceptable, a large proportion were clear, concise and easy to read. However, some participants reported the genotyping result in absence of any interpretation.


The technical performance of genotyping in EGFR mutation testing for NSCLC has room for improvement, evident from a high level of “true” diagnostic errors. Overall the standard of reporting was acceptable. Robust EQA can contribute to global optimization of EGFR testing.


All authors have declared no conflicts of interest.