Molecular diagnostic testing using DNA from saliva specimens markedly increased recently due to the ease of sample collection, compared to peripheral blood, during the COVID-19 pandemic. Published literature suggests that DNA from saliva is primarily composed of epithelial cells (70-90%), with the remainder being primarily leukocytes (10-30%). Here, we describe a case followed by our clinicians since 2007 characterized by developmental delay, autism, a somewhat coarse face with full cheeks, up-slanting palpebral fissures, thin corpus callosum, and a full-scale IQ of 60. This patient had an extensive work-up including high-resolution blood chromosome analysis, FISH for 22q microdeletion, three separate microarrays (various platforms), FMR1 molecular analysis, urine oligosaccharides analysis, an autism gene panel by NGS, and whole-exome sequencing, none of which identified a satisfactory diagnosis. These tests were performed on two peripheral blood samples collected at different times. Recently, a new microarray was ordered on a saliva sample from this patient, and an apparently non-mosaic gain of 18p was detected. The possibility of a sample swap was eliminated by comparing the SNP genotype of the saliva sample to the previously tested blood sample. The limit of detection for mosaicism in genomic microarrays is around 20%, so it is possible that the 18p duplication was present at a level undetectable by microarray in the peripheral blood samples. This case suggests that the differences between DNA obtained from saliva and peripheral blood may be, in some cases, more drastic than previously recognized. By testing primarily with DNA from peripheral blood, significant mosaic abnormalities may go undetected.

To read this article in full you will need to make a payment

DOI: https://doi.org/10.1016/j.cancergen.2022.05.021

© 2022 Published by Elsevier Inc.