External Quality Assessment for Detection of Fetal Trisomy 21, 18, and 13 by Massively Parallel Sequencing in Clinical Laboratories

Abstract

An external quality assessment for detection of trisomy 21, 18, and 13 by massively parallel sequencing was implemented by the National Center for Clinical Laboratories of People's Republic of China in 2014. Simulated samples were prepared by mixing fragmented abnormal DNA with plasma from non-pregnant women. The external quality assessment panel, comprising 5 samples from pregnant healthy women, 2 samples with sex chromosome aneuploidies, and 13 samples with different concentrations of fetal fractions positive for trisomy 21, 18, and 13, was then distributed to participating laboratories. In total, 55.6% (47 of 84) of respondents correctly identified each of the samples in the panel. Seventeen false-negative and 87 gray zone results were reported, most [102 of 104 (98.1%)] of which were derived from for trisomy samples with effective fetal fractions <4%. No laboratories generated false-positive results. In addition, we observed varied diagnostic capabilities of different assays, with the assay on the basis of NextSeq CN500 performing better than others, whereas Z values generated by BGISEQ-100 fluctuated greatly. There were no significant correlations between the numbers of unique sequence reads and Z values from any trisomy sample generated by BGISEQ-100. Overall, most clinical laboratories detected samples containing effective fetal fractions >4%. Our study shows need for further laboratory training in the management of samples with low fetal fractions. For some assays, precision of Z values needs to be improved.