A-324 Incorporating the Screening for Congenital Cytomegalovirus into a Newborn Genetic Testing Performed on Saliva Swab: An Intersection between Human Genetics and Infectious Diseases

Abstract

Abstract Background Newborn genetic testing is a powerful tool for improving the future healthcare of infants because of its potential to screening for a wide range of disorders with a single test. The assay can be performed in non-invasive sample such as saliva swabs. Hundreds of genes associated to disorders with significant morbidity and mortality (if undiagnosed) are analyzed. Despite the knowledge that infectious disorders such as congenital cytomegalovirus virus (cCMV) infection also has potential to cause serious illness to newborns (10–15% of apparently asymptomatic infants with cCMV may develop hearing loss) most of the newborn genetic testing does not test the saliva for cCMV. This validation aims to compare two laboratory-developed PCR-based CMV tests in saliva samples to add the cCMV screening into our newborn genetic testing. Methods Saliva swabs (OCR-100, DNA Genotek) from infants of our NGS-based newborn genetic testing were enrolled (n = 171). Parent signed informed consent agreeing with the used of the left-over specimens. Infants age ranged from 0 to 10 years. Saliva DNA was extracted by using QIAsymphony and DSP DNA mini kit (Qiagen). CMV was tested by two independent qPCR assays performed on LightCycler 480II using LightCycler Multiplex DNA Master (Roche). Each assays target two separated regions of the CMV genome (UL55 + UL123 and UL34 + UL80.5). A third assay targeting human CPOX gene was used as process control. The assays limits of detection were calculated by applying probit regression to the qualitative results returned from a serial dilution (1:10) of a positive pool (2 310 000 UI/mL) into a negative pool of samples. The accuracies for CMV detection were estimated by the total, positive and negative agreement between the compared sets of assays. The assay with lower limit of detection were chosen and its precision was evaluated by testing a subset of 26 saliva swabs (6 positives and 20 negatives) during 3 distinct days. For accuracy and precision experiments, Cq values >35 (∼750 UI/mL) was considered as not detected. Results The limits of detections were 40.7 UI/mL (95% CI:19.2–62.27 UI/mL) for UL55 + UL123 and 184 UI/mL (95% CI:64–304 UI/mL) for UL34 + UL80.5. CMV was detected in 8 and not detected in 163 saliva swabs by either UL55 + UL123 and UL34 + UL80.5 resulting in total, positive and negative agreements of 100% (95% CI:97–100%), 100% (95% CI:67–100%) and 100% (95% CI:97–100%), respectively. The median (min-max) Cq values for UL55 + UL123 were 30.75 (21.39–34.42) and for UL34 + UL80.5 were 30.81 (20.53–34.1). In the assay precision evaluation, UL55 + UL123 returned the expected result in 26/26, 26/26 and 25/26 samples on day 1, 2 and 3, respectively. The exception was a borderline positive sample that reach the Cq cut-off value (>35) on day 3 evaluation. Conclusion The UL55 + UL123 assay had lower limit of detection for CMV compared UL34 + UL80.5 and was chosen to be used in our cCMV screening. Both assays showed complete agreement when applied to saliva swabs samples. UL55 + UL123 presented acceptable precisions for positive and negative results. The cCMV screening by qPCR is now performed in parallel to NGS in our newborn genetic testing. cCMV test is only performed for newborns with less than 21 days of life, as recommend by specific guidelines.