Abstract
BACKGROUND: Droplet digital (dd)PCR is a new-generation PCR technique with high precision and sensitivity; however, the positive and negative droplets are not always effectively separated because of the "rain" phenomenon. We aimed to develop a practical optimization and evaluation process for the ddPCR assay and to apply it to the detection of BRAF V600E in fine-needle aspiration (FNA) specimens of thyroid nodules, as an example. METHODS: We optimized seven ddPCR parameters that can affect "rain." Analytical and clinical performance were analyzed based on histological diagnosis after thyroidectomy using a consecutive prospective series of 242 FNA specimens. RESULTS: The annealing time and temperature, number of PCR cycles, and primer and probe concentrations were found to be more important considerations for assay optimization than the denaturation time and ramp rate. The limit of blank and 95% limit of detection were 0% and 0.027%, respectively. The sensitivity of ddPCR for histological papillary thyroid carcinoma (PTC) was 82.4% (95% confidence interval [CI], 73.6%-89.2%). The pooled sensitivity of BRAF V600E in FNA specimens for histological PTC was 78.6% (95% CI, 75.9%-81.2%, I(2)=60.6%). CONCLUSIONS: We present a practical approach for optimizing ddPCR parameters that affect the separation of positive and negative droplets to reduce rain. Our approach to optimizing ddPCR parameters can be expanded to general ddPCR assays for specific mutations in clinical laboratories. The highly sensitive ddPCR can compensate for uncertainty in cytological diagnosis by detecting low levels of BRAF V600E.