olled in the EURTAC trial, thus validating the use of the EGFR PCR test to select patients for treatment with anti-EGFR TKIs such as erlotinib. Median PFS survival was 9.7 versus 10.4 months for the erlotinib group and 5.2 versus 5.4 months 
for the LDTs and EGFR PCR test, respectively. The BORR was 58% versus 59.3% months for the erlotinib group and 15% versus 14.0% for the LDTs and EGFR PCR test, respectively. Among the 16 discordant specimens between the EGFR PCR test and LDTs, a third mutation testing method agreed with the EGFR PCR test result in 11 cases. Of seven cases that were mutation FD&C Green No. 3 chemical information detected by the EGFR PCR test and mutation not detected by the LDT, 5 were confirmed by MPP. These patients could have potentially benefited from anti-EGFR TKI therapy. The EGFR PCR test had a number of technical advantages over the LDT used in the EURTAC trial. The LDT required laser capture microdissection of multiple tissue sections and involved 3 separate assays with a median turnaround time of 4.5 days. By comparison the EGFR PCR test required macrodissection only if the tumor content was,10% and can be performed in one day using a single 5 mm section. Furthermore the EGFR PCR test is a commercially available kit-based assay that provides an automated result, rather than a manual process subject to interpretation and which can be performed by any qualified clinical laboratory. More than 80% of the specimens tested in this study were small biopsy specimens. The overall invalid rate for Sanger sequencing was 15.6% compared to the EGFR PCR assay at 9%. However, the invalid rate for the subset of specimens derived from resected specimens was 0% likely because of sufficient tissue availability. Thus the assay is extremely robust when performed on resected tumor specimens and has an approximately 90% success rate on biopsy specimens, which are often the only tumor sample available for testing in NSCLC. Sanger sequencing has been widely used to detect EGFR mutations. There were also 30 mutation not detected results for Sanger sequencing and 7 mutation not detected results for the EGFR PCR test. With 21 invalid results and 30 mutation not detected results, Sanger sequencing would have misclassified 38% of patients enrolled in the EURTAC trial. Similar invalid rates have been reported in three other studies, suggesting that this methodology has limitations when applied to DNA from FFPET samples. In addition, Sanger sequencing has shown poor sensitivity in samples containing less than 2025% mutant alleles. When we compared the PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/19632868 agreement between valid results for the EGFR PCR test with Sanger SLCG LDT N = 432 Mutation detected Mutation detected Mutation not detected Total 146 9 155 Mutation not detected 7 270 277 Total EGFR PCR test 153 279 432 N12 samples with inconclusive LDT results and 43 samples with invalid EGFR PCR test results were excluded. Positive percent agreement = 94.2%. Negative percent agreement = 97.5%. Overall percent agreement = 96.3%. PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/19630872 EGFR Mutation Testing in NSCLC in EURTAC Trial Sanger sequencing N = 406 Mutation detected Mutation detected Mutation not detected Total 112 4 116 Mutation not detected 34 256 290 Total EGFR PCR test 146 260 406 81 samples with invalid EGFR PCR test or Sanger sequencing results were excluded. Positive percent agreement = 96.6%. Negative percent agreement = 88.3%. Overall percent agreement = 90.6%. doi:10.1371/journal.pone.0089518.t004 sequencing, there were 38 discordant cases of which 30 were confirmed
Posted inUncategorized