RESUMO
Molecular testing on metastatic lung adenocarcinoma or on non-small cell non-squamous lung carcinoma often relies on small specimen. In this group of patient with poor specimen adequacy, we analyzed the rate of EGFR, KRAS, BRAF and HER2 mutations compared to their rate in optimal specimen. We analyzed discrepancies in molecular testing results in patients with iterative analysis on several samples. We performed a retrospective study of 1538 samples consecutively analyzed. 263/665 (39,5%) biopsies and 37/708 (5,2%) surgical specimens were considered as samples with poor specimen adequacy (p<0,0001). A lower tumor cell content was associated with a lower rate of KRAS mutation: 15,8% in samples with <10% of tumor cells or <100 tumor cells versus 29,8% in samples with >10% tumor cell and >100 tumor cells (p=0,001). KRAS mutational rate was at 11,1% in cytology specimens, significantly lower than in biopsy or surgical specimens respectively at 28,2% and 28,5% (p=0,0002). Tumor cell content was not associated with mutational rate for EGFR, BRAF and HER2 mutations. DNA quantity was not associated with mutational rate for EGFR, KRAS, BRAF and HER2. A discrepancy in molecular testing was found in 16 patients. For 5 patients there was also a discrepancy for TTF-1 expression. On the 11 without TTF-1 discrepancy, specimen adequacy was not fulfilled in 10 cases at least for tumor content. Discrepancies were found in the case of low cellularity, poor cell content or testing on cytological specimens. Tumor cell content is a crucial parameter for molecular analysis rather than the type of specimen or the DNA quantity. Discrepancies in molecular testing results are rare but might suggest the presence of another tumor type, the emergence of another clone or a molecular testing in a sample with low cell content.