All patients underwent concomitant tissue and blood-based
DNA genotyping using the cobas blood test, followed by longitu-
dinal blood-based monitoring using the same test for EGFR exon
19 deletions, L858R and T790M mutations, and/or all KRAS
codon 12 (G12X) alterations. The positive predictive value of
plasma ctDNA analysis was 100% for EGFR19 deletions, EGFR
L858R, and KRAS G12X and 79% for T790M. Interestingly, sen-
sitivity for both EGFR and KRAS was higher for patients with
multiple metastatic sites and those with hepatic or bone metastases
than those with a single metastatic site on these organs.
Indeed, liquid biopsy technologies are considered by regu-
latory agencies, such as the FDA and the European Medicines
Agency (EMA), as in vitro diagnostic devices. Therefore, it is
subject to approval based on the criteria of safety and effectiveness.
To accelerate the development, adoption, and reimbursement of
blood (and other body fluids) liquid biopsy, these generic rules
must be translated into standards for clinical concordance, valida-
tion, and utility studies. The FDA has published a draft guidance
document to facilitate this process [54]. Liquid biopsy is crucial in
view of the fact that Illumina, a prominent manufacturer of NGS
apparatus, and other large companies, such as Johnson & Johnson,
Qiagen, and Roche, as well as at least 30 other new companies, are
steadily developing diagnostic blood tests for multiple cancer types.
To date, however, liquid biopsy has merely entered the clinical
practice of NSCLC management. In January 2015, the EMA
granted the approval of In Vitro Diagnostic Medical Device
(IVD) marketing in Europe to the therascreen EGFR RGQ PCR
Kit. And in 2016, the FDA approved the cobas EGFR Mutation
Test v2 (an updated version of the prior cobas EGFR Mutation
Test) for application in the USA. Both of these IVDs can detect
EGFR mutations in plasma ctDNA with comparable accuracy to
those in DNA from tumor tissue specimens by Sanger sequencing,
as revealed by large clinical trials of small-molecule EGFR TKIs in
patients with EGFR-mutated NSCLC. The therascreen assay
enables the detection of 19 different exon 19 deletions and 3 dis-
tinct exon 20 insertions (although it cannot be used to distinguish
which specific deletion or insertion is present), as well as L858R,
L861Q, G719X, S768I, and T790M mutations in EGFR.
In the Lung-LUX3 trial, the overall percentage agreement
(OPA) between the therascreen test and the tissue-based test results
was 92.2% (95% CI, 89.0–94.8%) [55]. For 27 of 28 discordant
sample pairs (96%), the presence of EGFR mutation was detected in
plasma ctDNA, but not in DNA from tumor tissue samples
[56]. On the one hand, the clinical utility of the therascreen test
was confirmed in a phase IV trial of gefitinib [57], in which 12 of
201 patients with no tumor tissue available for genotyping were
found to harbor EGFR mutations by this liquid biopsy assay.The Introduction and Clinical Application of Cell-Free Tumor DNA 55