Recently Lefkowitz et al (Sequenom Laboratories, San Diego, CA) reported a validation study of fetal/placental copy number variants by sequencing the cell-free DNA of maternal plasma. The clinical sensitivity was 97.7% (42 of 43 copy number variants detected).

Unfortunately, the authors have confused the concepts of analytic validity (the correct response for the sample tested) with clinical validity (the correct response for the patient tested). Among the 43 cases, 9 were identified by cell-free DNA sequencing as being autosomal trisomies (3 trisomy 8, 1 trisomy 14, 3 trisomy 16, 1 trisomy 21, and 2 trisomy 22). All 9 had normal karyotypes (8 after amniocentesis and 1 after chorionic villus sampling). Ordinarily, these 9 would be classified as confined placental mosaicisms and the test results considered clinical false positives (test positive but fetus unaffected).

According to the study’s protocol, however, all 9 cases were considered correct clinical calls (true positives) because the abnormalities were confirmed by deeper sequencing of the cell-free DNA. Such deep sequencing can confirm only that the cell-free DNA results are analytically correct (ie, the placental cell-free DNA does indeed represent an abnormal karyotype). Finding a true placental copy number variant may, or may not, have clinical benefit, but such findings should not be confused with the fetal karyotype.

An additional important caveat is that the copy number variants used in this validation study are unlikely to represent an unselected series of copy number variants seen in pregnancy, many of which are smaller than 7 Mb and are not included in panels of select copy number variants. Given the overall reliability of cell-free DNA screening performance, it is especially important to avoid confusion over inappropriate claims of high clinical sensitivity because this has the potential to create distrust of laboratories offering cell-free DNA screening.