A-351 Experience and outcomes of cfDNA testing in the midst of ethnic and demographic variability at a large SafetyNet teaching hospital

Abstract

Background Biochemical assessment of risk of foetal aneuploidy, Trisomy’s T21, 18 and 13 are widely performed and have evolved over time from using a combination of maternal circulating biomarkers namely alpha-fetoprotein, chorionic gonadotropin, unconjugated esteriol, and dimeric inhibin A at about 21 weeks of gestation to recent development using cfDNA thought to exhibit much lower false positive rates and thus reduces unnecessary follow up invasive amniocentesis procedures. This study examined the impact of implementing a cfDNA testing program at a large SafetyNet hospital. Methods Retrospective review of all cfDNA results reported as well as electronic medical chart review for amniocentesis testing findings (if performed), ultrasound findings, and birth (pregnancy) outcomes were reviewed for final diagnosis. cfDNA testing was performed using Revvity chemagic cfDNA isolation kits, dedicated chemagic instruments using M-PVA Magnetic beads. Results A total of 19,328 patients had cfDNA screens during the study period where 156 (0.811%) patients had positive results. OF the positive screens, 41 were positive for T21, 56 were positive for T18, and 59 patients were positive for T13. Patients’ outcomes (true positives) were confirmed by reviewing medical charts for birth outcomes and results of secondary confirmatory testing (amniocentesis, ultrasound). Of the T21 positive patients, 6 were true positives (14.6%), and 19 were false positives (46.3%), the remainder outcomes were unknown, as for T18, 15 (2.8%) patients were true positives, 17 patients were false positives (30.4%), and the remainder outcomes were unknown, and for T13, 9 patients (15.3%) were true positives, 27 patients (45.8%) were false positives, and the rest were unknown or had other congenital anomalies. T 21 positive samples Z-scores ranged from 12.07 to 23.34 (median 19.69), T18 positive samples Z scores ranged from 4.74 to 45.31 (median 12.17), and T13 positive samples Z scores ranged from 8.65 to 33.33 (median 17.46). Z-scores exhibited marked imprecision and variability when patients BMI exceeded 40. Significant association between screen results and BMI was only observed for T18. Patients with true positive T18 had significantly lower BMI compared to patients with false positive results (p=0.02). Although T21 samples Z-scores were significantly higher among African American patients compared with White patients (p=0.007), no significant differences were observed for T18 or T13 (p= 0.3 and 0.2 respectively). Conclusion cfDNA in addition to offering low false positive results, it facilitated earlier detection at 10 weeks. Interestingly, samples from patients with high BMI (>40.0) exhibited variable Z-scores for T21, T18, and T13. The pathophysiology is not known but likely to be secondary to dilutional effects. The cause of false positive results and whether additional interventions are required remains under debate. Placental mosaicism and intra-uterine fetal demise were associated with false positive results.