Fetal exome sequencing: yield and limitations in a tertiary referral center

Objective: To explore the indications for and diagnostic outcomes of fetal exome sequencing in a tertiary referral center. Methods: Between 2012 and 2017, 77 unrelated fetal samples from pregnancies referred to our center underwent exome sequencing. The cohort included 37 fetuses, 36 products of conception (from cases of pregnancy termination or intrauterine fetal death), one case with DNA from both the fetus and a previous termination of pregnancy, and three cases with DNA of unknown origin. Exome sequencing was performed on DNA extracted from amniocytes or fetal tissue and, in some cases, from parental peripheral blood. Indications, turnaround time, diagnostic rates and pregnancy outcomes were investigated. Diagnostic yield was analyzed according to consanguinity (yes or no), sample type (proband only, or trio or other) and referral indication (malformation or isolated nuchal translucency (NT)). Results: The most common indication for fetal exome sequencing was multiple malformations (21/77, 27%), followed by isolated brain malformation (15/77, 19%). Twelve (16%) fetuses were referred for isolated increased NT. Exome analysis was diagnostic for 16 fetuses (21%); when subclassified into fetal malformations vs isolated increased NT it became clear that exome analysis did not reveal any known or probable pathogenic variants in cases referred for isolated increased NT, whereas, among the remaining fetuses, a molecular diagnosis was reached in 16/65 (25%). Proband-only cases received a diagnosis more often than did cases that had trio exome sequencing. Conclusions: Exome sequencing has the potential to provide molecular diagnoses in cases in which conventional prenatal cytogenetic testing is negative. Referral bias of consanguineous cases could account for the high diagnostic rate of proband-only sequencing. Syndrome-specific prognostic information enables parents to make informed decisions, whereas challenges include time limitations and variant interpretation in the setting of non-specific fetal findings. As we report only established disease–gene associations, further segregation and functional studies in a research setting are expected to increase significantly the diagnostic yield.