Tumour content ratio matters for detecting epidermal growth factor receptor mutation by cobas test in small biopsies; a retrospective study.

BACKGROUND: Recent studies indicate the benefit of treatment with osimertinib over that with conventional epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKI) for untreated EGFR-mutated non-small cell lung cancer (NSCLC). Cobas ver2 is the only companion diagnostic method for detecting EGFR mutations with osimertinib treatment. We clinically experience false negative cases with this test, but its actual sensitivity is unknown. Moreover, no study has suggested the importance of tumour dissection, and most facilities do not routinely perform them on small biopsies. The purpose of this study was to evaluate the sensitivity of cobas in clinical practice and clarify the role of dissection as a component of the cobas testing. METHODS: We examined 132 patients with EGFR-mutated NSCLC diagnosed by bronchoscopy and confirmed with PCR clamp. Patients were tested with cobas and the EGFR-positive rate was calculated. Samples with undetected EGFR mutations were retested after tumour dissection and the rate of samples whose EGFR mutation was corrected to positive was assessed. To evaluate tumour cellularity, the tumour content ratio was assessed by calculating tumour cell count over the total cell count on the slide. RESULTS: The positive rate of EGFR mutation identification was 76% with cobas, although EGFR mutation-negative patients retained responses to TKI therapy equivalent to positive patients did; however, the tumour content ratio of negative samples was significantly lower than that of positive samples. Twenty-nine negative samples underwent dissection and 24% were corrected to positive. Moreover, 53% of the samples with a tumour content ratio below 10% was negative for cobas, but 33% of these turned positive after dissection. CONCLUSIONS: Cobas had a high false negative rate in clinical practice, and tumour content ratio might be associated with this rate. Dissection could improve the sensitivity of cobas, especially in samples with low tumour cellularity.