Epidermal growth factor receptor mutation analysis in previously unanalyzed histology samples and cytology samples...

Epidermal growth factor receptor (EGFR) mutation testing is standard practice after lung adenocarcinoma diagnosis, and provision of high-quality tumor tissue is ideal. However, there are knowledge gaps regarding the utility of cytology or low tumor content histology samples to establish EGFR mutation status, particularly with regard to the proportion of testing performed using these sample types, and the lack of an established link with efficacy of treatment.

The randomized phase III Iressa Pan-ASia Study (IPASS; ClinicalTrials.gov identifier NCT00322452) of first-line gefitinib versus chemotherapy analyzed samples meeting preplanned specifications (n=437 evaluable for EGFR mutation; n=261 mutation-positive). This supplementary analysis assessed tumor content and mutation status of histology (n=99) and cytology samples (n=116) which were previously unanalyzed due to sample quality, type, and tumor content (<100 cells). Objective response rate (ORR) and change in tumor size with gefitinib treatment were assessed.

EGFR mutation testing was successful in 80% and 19% of previously unanalyzed histology and cytology samples, respectively. Mutations were detected in 54 tumors previously described as mutation-unknown (histology, n=45; cytology, n=9). ORRs in mutation-positive cytology (83%) and histology (74%) subgroups were consistent with previous analyses (71%). Tumor size decrease was consistent across previously analyzed and unanalyzed samples (all mutation subgroups), with less consistency across ORRs in mutation-negative cytology (16%) and histology (25%) subgroups versus the previous analysis (1%).

Histology samples with low tumor content and cytology samples can be used for EGFR mutation testing; patients whose mutation status was confirmed using these sample types achieved a response to treatment consistent with those confirmed using high-quality histology samples. Better sample quantity/quality can potentially reduce false-negative results.