We recently performed a fine-needle aspiration (FNA) on a patient aged 28 years, a father of 2 children, with a deep neck mass. Smears on site looked like melanoma, but there was no history. Immunohistochemistry and potentially molecular testing would be needed. For example, it could be useful to analyze the gene BRAF, coding for the serine/threonine protein kinase B-Raf, to identify mutations that confer survival benefit to melanoma patients treated with kinase inhibitors.1 Scenarios like this are playing out with increasing frequency in cytology.What exactly is the best way to triage material for the molecular laboratory?The short answer is that any cytology preparation (eg, smears, monolayer preparations, or cell blocks fixed by various means) can produce a valid molecular test result, and a key consideration is to use the preparation with sufficient numbers of tumor cells 2 (Table 1). [3][4][5][6][7][8][9][10][11][12][13] The devil is in the details, however, and the many publications that have established the suitability of cytology samples for molecular diagnosis continue to raise new questions. Beyond the scientific questions, molecular laboratories in the United States need to stay compliant with the Clinical Laboratory Improvement Act (CLIA). More challenging still could be the proposed new US Food and Drug Administration (FDA) regulations that would require laboratory-developed tests (LDTs) to be granted FDA approvals. 14 These regulations, if enacted as described, threaten the ability of cytology to help patients who need molecular testing. In this article, we have restricted attention only to BRAF mutation testing as an example of how technologies and the regulatory climate affect the ability of cytology laboratories to provide a valuable service for patients.
Three Factors That Affect the Suitability of Cytology Samples for Molecular TestingThe 3 key factors that affect the validity of a sequence-based molecular assay are the proportion of tumor cells in the sample, DNA quality, and DNA quantity. A low proportion of tumor cells could yield a dangerous falsenegative result, rendering a mutation undetectable, even with the most robust assays and controls. An advantage of cytology samples compared with surgical biopsies is that they frequently have a higher proportion of carcinoma cells. Newer microbiopsy devices may help to improve the procurement of heavily cross-linked sarcomas or desmoplastic carcinomas. The FDA-approved Roche Cobas BRAF test (Roche Molecular Systems, Inc, Pleasanton, Calif) detects mutant valine codon 600 (V600) alleles when tumor cells comprise 15% of total cells. It is not trivial to estimate the percentage of tumor cells, and data on the accuracy of pathologists' estimate of tumor cellularity have been limited to paraffin sections. By using an objective counting method as a gold standard, pathologists in