Adequately defining tumor cell proportion in tissue samples for molecular testing improves interobserver reproducibility of its assessment

摘要: Gene mutation status assessment of tumors has become standard practice in diagnostic pathology. This is done using samples comprising tumor cells but also non-tumor cells, which may dramatically dilute the proportion DNA and induce false negative results. Increasing sensitivity molecular tests presently allows detection a targeted sample with small percentage cellularity remains essential to adequately interpret results tests. Comprehensive cell counting would provide most reliable approach time consuming, therefore rough global estimations are used, reliability been questioned view their potential clinical impact. The French association for quality assurance pathology (AFAQAP) conducted two external schemes, partly partnership group oncology cytogenomics (GFCO). purpose schemes was (1) evaluate how assessed on tissue samples, (2) identify reasons discrepancies, (3) recommendations standardization improvement. Tumor percentages lung colon cancer were estimated by 40–50 participants, 10 H&E virtual slides 20 conventional slides. average difference between lowest highest 66. largely due inadequate definition cellularity, reflecting confusion area occupied region. widest range interobserver variation observed dense or scattered lymphocytic infiltrates mucinous stroma. Estimations more accurate cases low cells. Macrodissection homogeneous reduced inter-laboratory variation. We developed rating system indicating impact discrepancy. Fewer discrepancies clinically relevant since study conducted. Although semi-quantitative remain somewhat subjective, improves when defined heterogeneous macrodissected analysis.