Our systematic review and meta-analysis of ES in fetuses with SD and normal results on karyotyping or CMA revealed a 60% (CI, 48–72%) diagnostic yield for monogenetic disorders. Some studies revealed that the prenatal diagnostic yield of suspected SD is higher than that of the postnatal population33. Therefore, it is necessary to perform ES on fetuses suspected of skeletal dysplasia to improve diagnostic efficiency, which can also shorten the treatment process and improve diagnostic efficiency.
ES in fetuses with SD seems to have a greater diagnosis rate when compared to other systemic or ultrasonography subtypes of illnesses, and there are likely several factors that contribute to this finding. At the ultrasound diagnostic level, the fetal bones are relatively simple, and skeletal system diseases are more easily diagnosed by ultrasound10, whereas for the molecular diagnostic level, SD is shown to be more likely to be caused by a single genetic variant34–36. Additionally, the diagnostic yield may be significantly impacted by whether or not case selection is carried out prior to clinical implementation. According to our findings, there is a substantial difference in diagnostic yield depending on whether case selection is used (Selected: 65% vs. Unselected: 33%, p = 0.02). The PAGE study suggests that ES should ideally be performed on selected patients to avoid the uncertainty and dilemma in clinical management that would result from the detection of VUS and the low number of possible diagnoses9. Especially when faced with SD, a highly heterogeneous disease, the impact of case-selection measures on detection rates may be even greater.
Nowadays, there is also the "NGS gene panel" as an additional option for focusing on particular fetal abnormalities. The targeted skeletal gene panel has a greater sensitivity for SD and is less expensive and time-consuming than WES or whole genome sequencing (WGS)18. In our data, the diagnostic yield of the targeted skeletal gene panel for fetuses was 76% (CI, 62–89%). Even though the panels only comprise specific causative genes, employing skeletal gene panels in a clinical environment would have a significant diagnostic yield when accompanied by a high degree of case selection. However, as the phenotype-gene database of fetuses with skeletal anomalies continues to expand, specialized panels for SD based on the ISDS Nosology may limit the diagnosis of SD28. 36.4% (27/74) of the genes with pathogenic or possible pathogenic variants detected in our study were not included in the ISDS Nosology, while 27 genes are not on this list. It shows that SD can also be brought on by some specific genes that are occasionally not included in these pre-made NGS panels. Chandler et al, enlarged the analysis by 131 genes based on ultrasound findings after a case of negative fetal skeletal gene panel and they eventually identified a causative variant consistent with the fetal ultrasound findings17. It suggests that the NGS panel based on the ISDS Nosology still has its limitations in the prenatal setting. When the NGS panel tests are negative, genetic counselors need to understand that approximately one-third of genes with the pathogenic or likely pathogenic variants have not been analyzed, suggesting the residual risk is still high. Therefore, when using the NGS panel, it is required to talk about how to establish a more suitable analysis group or to take into account using the WES/WGS.
According to our data, the majority of pathogenic or potentially pathogenic variations are involved in FGFR3, COL1A1, and COL1A2, which together account for about half of all genes discovered. The most frequent diagnoses were achondroplasia, thanatophoric dwarfism type I (TD-1), and osteogenesis imperfecta (OI). Based on previous studies, the FGFR3 protein limits the ossification process to control bone growth37, and mutations primarily cause achondroplasia and TD-1. Moreover, the recent clinical guidelines also state that WES and skeletal gene panel sequencing are required for the diagnosis of achondroplasia, particularly in fetuses with clinical and imaging atypia38. Type I collagen-encoding genes (COL1A1 and COL1A2) mutations are primarily responsible for OI39,40. Currently, pathogenic variants in genes associated with SD are available at some scale in the human genetic variation database, however, the number of novel variants identified in our study accounted for 23.7% (82/253) of those. This finding once again reveals the significant genetic diversity of genes linked to SD and underscores the significance of the molecular diagnosis of fetal skeletal abnormalities. Additionally, With the use of WES/WGS, more and more pathogenic genes and novel variants will be discovered, which will contribute to clinical consultation, diagnosis, and treatment.
Equally important are issues about the ES turnaround time (TAT). Fast-turnaround ES can offer important information for prenatal care. In the clinical guidelines, it is advised to publish a final report and confirmation of results within 21 days41. Nowadays, researchers have used advanced tachycapture techniques, genetic databases, and effective bioinformatics systems to reduce the TAT of ES to about two weeks10,17,23. However, a novel clinical pathway that merits our attention is the simultaneous use of CNV-seq and ES, Chen et al. demonstrated this pathway can compress the TAT of the entire diagnostic process to less than two weeks while avoiding the lengthy waiting period associated with the use of a sequential CMA-ES strategy31. However, it cannot be overlooked that the concurrent use of the CMA and ES may place a financial burden on patients; as a result, adequate communication with patients is required when developing clinical pathways.
There are some limitations in this study. Despite the adoption of a random effects model, our findings showed significant heterogeneity. There are some reasons for this. The research we included has seven small sample studies (n < 20), several of which have been carefully selected to allow for a pathological debate. Additionally, due to different inclusion criteria and patient selection methods, this also increases the chance of increasing heterogeneity.
This systematic review and meta-analysis suggests that ES applied to fetuses with SD and normal karyotype or CMA can provide a considerable diagnostic yield. ES is more effective in the diagnosis of selected fetuses when compared to unselected fetuses. To fully understand the experience of how to choose fetuses with SD for ES and maximize the advantages of next-generation sequencing, more research is required.