Cell-free DNA screening for fetal conditions other than the common aneuploidies

INTRODUCTION — Cell-free DNA (cfDNA) is produced by both the pregnant person and the placenta in fragments that are between 50 to 200 base pairs, which enter the maternal circulation [1]. Analysis of cfDNA for common autosomal aneuploidies (trisomies 21,18, and 13) using next-generation sequencing techniques has high sensitivity and specificity. It is commonly used prenatally to screen for these conditions, which account for 71 percent of all prenatally detected chromosome abnormalities [2]. Analyses for other genetic conditions are also possible, including autosomal trisomies other than trisomy 21, 18, and 13; sex chromosome aneuploidy; microdeletions and microduplications; and monogenic (single-gene) disorders, because the entire fetal genome is represented among the cfDNA fragments. Such analyses have become increasingly available; however, information about the sensitivity and specificity of these screening tests in the general obstetric population is limited. Offering screening for these other conditions increases the screen-positive rate, and therefore the rate of invasive procedures for diagnostic testing. Another issue is incidental information on the maternal genome, which is also represented among the cfDNA fragments.

Use of cfDNA for prenatal screening for genetic abnormalities other than common aneuploidies in the general obstetric population will be discussed here. Additional information about prenatal screening with cfDNA, including basic information about cfDNA (origins, fetal fraction, clearance, test failures, false-positive and -negative results) and using the test to screen for common aneuploidies, is available separately. (See "Prenatal screening for common aneuploidies using cell-free DNA".)

ANALYTIC METHODS

●Counting method – The counting method is the most common method laboratories use to analyze fragments of cfDNA in the maternal circulation. In this method, the fragment is evaluated and the chromosome of origin is determined. This is then compared with the expected percent contributed by that chromosome in a euploid pregnant patient carrying a euploid fetus.

●Massive parallel sequencing method – In this method, both maternal and fetal cfDNA fragments are sequenced and the chromosomes of origin are identified. This method can be used for all chromosomes ("shotgun" sequencing), or for a select chromosome of interest ("targeted or chromosome selective sequencing"), such as chromosome 21.

●Single nucleotide polymorphism method – A single nucleotide polymorphism (SNP) represents a common change in a single base. This method evaluates the sequences of the SNPs on cfDNA fragments from selected fetal and maternal chromosomes and compares them with a reference euploid pregnancy [3].

SCREENING FOR A LIMITED PANEL OF MICRODELETIONS/MICRODUPLICATIONS

Clinical utility — Microdeletion and duplication syndromes can be associated with neurodevelopmental disabilities and developmental delays. Clinically significant microdeletions and microduplications are identified by chromosomal microarray (CMA) in 6 percent of euploid fetuses with ultrasound abnormalities and 1.7 percent of euploid fetuses without ultrasound abnormalities in whom a karyotype was performed on a chorionic villus sample, amniocytes, or a percutaneous umbilical blood sample [4,5]. The microdeletions and duplications are spread throughout the genome and are not necessarily associated with maternal age.

A number of companies now offer cfDNA screening of maternal blood for specific sets of the more common microdeletion and microduplication syndromes. Specific microdeletions and duplications include: 22q11.2 microdeletion syndrome, Angelman syndrome, Cri-du-chat, 1p36 deletion syndrome, and Prader-Willi syndrome.

There are a number of limitations to screening for microdeletion and microduplication syndromes:

●Many laboratories are unable to identify chromosome imbalances that are <5 to 7 Mb in size because of limitations with depth of sequencing; however, 70 percent of pathogenic copy number variants determined through CMA are <7 Mb. The most common microdeletion syndrome is 22q11.2, which is the second most common genetic cause of developmental delay after trisomy 21. The majority (85 percent) of individuals with 22q11.2 microdeletion syndrome have a 2.54 Mb deletion, considerably smaller than the 5 to 7 Mb deletion range that could potentially be identified by cfDNA [6]. The clinical features, diagnosis, management, and prognosis of 22q11.2 deletion are reviewed separately. (See "DiGeorge (22q11.2 deletion) syndrome: Epidemiology and pathogenesis" and "DiGeorge (22q11.2 deletion) syndrome: Clinical features and diagnosis" and "DiGeorge (22q11.2 deletion) syndrome: Management and prognosis".)

●The clinical phenotype of some disorders may result from a true deletion or duplication, but the same disorder can also result from other causes, such as a single-gene pathogenic variant or uniparental disomy. For example, microdeletions of the Prader-Willi-critical region on paternal 15q11-q13a is the cause for Prader-Willi syndrome in 65 to 75 percent of cases. The remaining cases are a result of single-gene abnormalities or uniparental disomy. Thus, absence of a diagnostic microdeletion on cfDNA screening for Prader-Willi syndrome does not rule out this disorder and could lead to false reassurance in counseling [7]. The clinical features, diagnosis, management, and prognosis of Prader-Willi syndrome are reviewed separately. (See "Prader-Willi syndrome: Clinical features and diagnosis" and "Prader-Willi syndrome: Management".)

Predictive value — Analytic validation studies have demonstrated the technical ability of laboratories to perform this type of analysis; however, these studies have a number of limitations. Lack of confirmation of findings on either chorionic villus sampling, amniocentesis, or live birth testing makes the clinical validity of these tests difficult to determine, including the rates of false positives and false negatives.

A number of proof-of-concept studies have demonstrated the variable performance of cfDNA screening for detecting clinically significant microdeletions and microduplications. In a meta-analysis of cfDNA performance for detecting microscopic copy number variants, overall sensitivity was 66 to 86 percent and specificity was 98.0 to 99.8 percent, with overall positive predictive value of 31 to 45 percent, which the authors felt was overestimated because of baseline cohort risk [8]. The true sensitivity of these tests depends on a number of factors, including the size of the region being evaluated, the depth of the sequencing performed, and the fetal fraction of the cfDNA sample [9]. While 22q11.2 is relatively common (1 in 4000 to 1 in 5000 live births), other microdeletions are much less common and for many the prevalence is unknown. The positive predictive value is expected to be low for rare conditions and impossible to calculate for disorders where the prevalence is unknown, as is the case for many microdeletion and duplication syndromes. For very rare conditions, the positive predictive value is expected to be low, or unable to be calculated. Similar to rare autosomal trisomies, false positives will increase as more tests are added to each of the cfDNA screening panels [10].

SCREENING FOR MONOGENIC DISEASE

Clinical utility — Over 2300 monogenic (single-gene) disorders have a known identifiable molecular cause. For pregnancies at risk, definitive diagnosis of a monogenic disorder can only be made by testing fetal DNA obtained by chorionic villus sampling, amniocentesis, or umbilical vein sampling. Definitive testing is generally performed in the setting of known parental carrier status of a monogenic disorder, or in the setting of suspicious ultrasound findings, such as concern for a skeletal dysplasia.

A number of studies have demonstrated that the entire fetal genome can be evaluated from cfDNA in maternal blood and a number of case reports and proof of principle studies have shown that evaluation of cfDNA in maternal blood can predict monogenic disorders. Thus, cfDNA testing could be utilized in the setting of relevant family history of a monogenic disorder or abnormal ultrasound findings. However, the use of cfDNA for monogenic disease assessment is uniquely challenged. First, individual monogenic disorders are less common than individual autosomal aneuploidies and only a small numbers of individuals with any specific monogenic disorder are available to study. Second, detection of fetal variants has to occur in the setting of a large amount of maternal background cfDNA, as the fetal fraction of cfDNA obtained is only 5 to 20 percent of the total circulating cfDNA [11]. This is especially an issue for autosomal-recessive conditions in which the mother carries the same pathogenic variant as the fetus. For de novo or parental-dominant conditions, detecting a pathogenic variant amongst the maternal background is more straightforward. Despite these limitations and varied laboratory techniques, detection of sex-linked disorders (Duchenne/Becker muscular dystrophy), autosomal-dominant disorders (achondroplasia) and autosomal-recessive disorders (cystic fibrosis) have been described [12,13].

Autosomal-dominant disorders — Autosomal-dominant disorders in the fetus are caused by de novo gene changes or an inherited pathogenic variant. Both have been detected by cfDNA evaluation in small studies.

●De novo autosomal-dominant disorders – Multiple small investigational studies have identified some autosomal-dominant disorders using cfDNA screening and such testing is now clinically available. For example, targeted sequencing is utilized for "hot spot" areas in common genes associated with skeletal dysplasias [14,15]. One commercially available test includes genes associated with skeletal dysplasias and other autosomal-dominant disorders, such as craniosynostosis syndromes and Noonan spectrum disorders, among others. Utilization of this test could be considered in the setting of ultrasound findings concerning for skeletal dysplasia without an affected parent, although the positive and negative predictive values are unknown.

●Parentally inherited autosomal-dominant disorders – Because fetal cfDNA is evaluated in the presence of both maternal and fetal cfDNA, laboratory technique must take into consideration that the cfDNA obtained may demonstrate a maternal pathogenic variant with or without an inherited fetal pathogenic variant. Detection of a paternally inherited pathogenic variant may be more straightforward as this variant is present in the fetal cfDNA, but not in the maternal cfDNA [13]. Of note, diagnosing a fetus with an autosomal-dominant disorder may implicate a parent who also has the disorder but was unaware of their diagnosis. An example would be Noonan syndrome with a cardiac anomaly in the fetus but not the parent, although the affected parent may have other typical features on closer examination.

Predictive value — The test characteristics of cfDNA for autosomal-dominant disorders are limited to small studies. Sensitivity and specificity have ranged from 96 to 100 depending on the study and the disorder evaluated [16,17]. A number of issues have not been resolved. For example [11]:

●Counseling about positive results in the setting of incomplete penetrance and variable expressivity for these disorders is highly complex.

●False-negative results are a significant issue due to a lack of sequencing of the entire gene of interest.

●Little guidance is available on variants of uncertain significance

Autosomal-recessive disorders — Maternally inherited pathogenic variants, either because the fetus is a carrier of an autosomal-recessive disorder or is affected by an autosomal-recessive disorder with pathogenic variants inherited by both parents, may be detected using a strategy called "relative mutation dosage." In this strategy, the laboratory determines the "expected" amount of abnormal alleles present in the cfDNA obtained, knowing that the maternal cfDNA has an abnormal allele, and compares this with the observed mutant alleles found, which enables prediction of the fetal status as a carrier or as homozygous for the condition of interest. A similar strategy, "relative haplotype dosage" utilizes single nucleotide polymorphisms (SNPs) that can be evaluated for fetal inheritance of an abnormal paternal allele or maternal allele. Both strategies have significant limitations that have thus far limited its clinical use [13].

SCREENING FOR SEX-CHROMOSOME ANEUPLOIDY

Clinical utility — Sex chromosome aneuploidies are common, affecting up to one in 400 newborns [18]. Sex chromosome aneuploidy is often not suspected in the fetus or newborn because distinctive phenotypic features are not present. If cfDNA suggests a sex chromosome aneuploidy and it is confirmed by diagnostic testing on amniocytes or chorionic villi, the parent(s) can be given prognostic information, which may include infertility or neurobehavioral issues [19]. In some cases, intervention may mitigate some of these outcomes. Ideally, prenatal prognostic counseling should be based on prospective follow-up of children born following a prenatal diagnosis of the specific sex chromosome aneuploidy [20]. However, prognostic information is often limited prenatally because it largely relies on reports of postnatally ascertained cases, which often do not include cases with no or minimal phenotypic findings.

cfDNA screening also carries the potential risk for identifying a sex chromosome mosaic aneuploidy in the pregnant individual, such as 45,X mosaicism. (See "Clinical manifestations and diagnosis of Turner syndrome".)

Predictive value — Test performance is discussed separately. (See "Prenatal screening for common aneuploidies using cell-free DNA", section on 'Sex chromosome aneuploidies'.)

SCREENING FOR RARE AUTOSOMAL TRISOMIES

Clinical utility — The clinical utility of screening for autosomal trisomies other than trisomy 21, 18, and 13, particularly at early gestational ages, is unclear. Autosomal trisomies other than trisomy 21, 18, and 13 are rare, present in <0.1 percent of newborns, prior to the widespread use of prenatal screening and diagnosis [9]. The most common are trisomy 7, 15, 16, and 22. The clinical consequences of detecting these rare autosomal trisomies are less clear compared with trisomy 21, 18, and 13. In contrast to the common autosomal aneuploidies, the significance of the rarer trisomies is largely dependent on the chromosome involved and the proportion of cells in the embryo/fetus and the placenta that are affected. It is important to remember that cfDNA attributed to the embryo/fetus is of placental origin and although the genetic makeup of the embryo/fetus and placenta are usually identical since they originate from the same fertilization, this is not always the case.

Prenatal cfDNA screening can be performed as early as 9 to 10 weeks of gestation. If the result shows one of the rare autosomal trisomies this early in gestation, the prognosis can be difficult to estimate. Pregnancies with embryo/fetal trisomy 7, 15, 16, and 22 at this gestational age are at high risk for miscarriage. If miscarriage does not occur and no fetal anomalies are identified on ultrasound as the pregnancy progresses to a more advanced gestational age, then confined placental mosaicism (CPM) should be suspected. In one study, 97 percent of all rare autosomal trisomies detected by chorionic villus sampling appeared to be CPM [21]. The diagnosis of CPM is supported by amniocentesis for karyotype at ≥15 weeks showing absence of the trisomy. It is important because CPM may be associated with fetal growth restriction and/or other pregnancy outcomes [22] (see "Chorionic villus sampling", section on 'Confined placental mosaicism').

If a euploid fetus is confirmed on amniocentesis and the autosomal trisomy detected by cfDNA involved chromosome 6,7,11,14,15, or 20, then the possibility of uniparental disomy (both homologous chromosomes inherited from one parent) needs to be considered and can be diagnosed by evaluation of DNA-based polymorphic markers [23]. The overall risk of uniparental disomy with CPM for one of these chromosomes is about 2 percent, but depends on the specific chromosome [24].

Another diagnostic outcome is that the genetic amniocentesis result supports the diagnosis of fetal mosaicism for the rare autosomal trisomy. The lack of information about the natural history of these mosaicisms both in utero and postnatally makes counseling on the predicted outcome extremely difficult [10]. For such pregnancies that continue, outcomes may include fetal anomalies, fetal growth restriction, fetal death, and neonatal death.

Predictive value — There are few studies that have evaluated the predictive value of cfDNA for rare autosomal trisomies. These disorders have a low prevalence and the studies are limited by small sample sizes. In one meta-analysis, the pooled positive predictive value for cfDNA for detecting rare autosomal trisomies was very low (11.46 percent, 95% CI 2.49-18.76) [25]. Even so, the low positive predictive value was felt to be an overestimation as many studies included patients who had ultrasound abnormalities. Data were insufficient to provide accurate ascertainment of sensitivity and specificity because most studies only offered confirmatory tests to patients with high-risk results.

SPECIAL CLINICAL SCENARIOS

Fetal RhD status — RhD is an autosomal-dominant trait, rather than a disorder. Use of cfDNA after 10 weeks of gestation to determine fetal RhD status has significant clinical potential. For example:

●In an alloimmunized pregnancy, if the fetus is RhD positive, monitoring maternal anti-D titers is appropriate because of the risk for hemolytic disease of the fetus and newborn. If the fetus is RhD negative, the fetus is not at risk, so such monitoring can be avoided. This strategy essentially eliminates the need for amniocentesis to determine fetal RhD status. A number of laboratories in Europe and at least one laboratory in the United States perform this test commercially. This topic is discussed in more detail separately. (See "RhD alloimmunization in pregnancy: Management", section on 'Cell-free DNA testing'.)

●In a non-alloimmunized pregnancy in which the pregnant person is RhD negative, if the fetus is RhD negative, maternal administration of anti-D immune globulin can be avoided. This topic is discussed in more detail separately. (See "RhD alloimmunization: Prevention in pregnant and postpartum patients", section on 'Fetal D-negative status known based on evaluation of cell-free DNA'.)

Suspected sex-linked disorders — Fetal sex determination is an area where the clinical utility of cfDNA screening may be highly valuable. For example, fetal sex virilization in a XX fetus affected with congenital adrenal hyperplasia (CAH) begins at 9 weeks of gestation, several weeks before the fetal sex can be evaluated accurately by ultrasound. Early maternal dexamethasone treatment may reduce this virilization process; however, if the fetus is XY and early dexamethasone is administered, then the medication is given without benefit and possible harm. In families where both parents are known carriers for CAH, early sex determination can allow for appropriate administration of dexamethasone to only predicted XX fetuses [13].

Fetal sex determination can also be useful for families with known X-linked disorders (such as hemophilia and Duchenne muscular dystrophy) to inform decisions about pursing invasive testing for predicted male fetuses who are at 50 percent risk of being affected by an X-linked disorder.

Performance data are reviewed separately. (See "Prenatal screening for common aneuploidies using cell-free DNA", section on 'Sex chromosome aneuploidies'.)

POSITIONS OF SELECTED PROFESSIONAL SOCIETIES — The use of cell-free fetal DNA for evaluation of rare autosomal trisomies and microdeletions and duplication, and monogenic disorder assessment is a frequent subject for investigation but has not been recommended by professional societies. Selected professional society positions on cfDNA for abnormalities other than common autosomal aneuploidies include:

●American College of Obstetricians and Gynecologists (ACOG) (validated 2020) and the Society for Maternal-Fetal Medicine (SMFM) – ACOG recommends not performing cfDNA screening for rare autosomal aneuploidies (eg, trisomy 16 and trisomy 22), microdeletion testing, and genome-wide screening of large copy number variants because screening accuracy and the false-positive rate are not established [26]. Likewise, although cfDNA screening for a limited number of microdeletions is available, this testing has not been validated clinically and not recommended.

●The American College of Medical Genetics (ACMG) strongly recommends offering cfDNA screening for fetal sex chromosome aneuploidy, and suggests its use for 22q11.2 deletion syndrome [19]. ACMG recommended not using it to screen for rare autosomal trisomies or other copy number variants.

●The International Society for Prenatal Diagnosis (ISPD) recommends against cfDNA screening for rare autosomal trisomies, subchromosomal imbalances, and microdeletion and microduplication syndromes in unselected populations, but states screening for sex chromosome aneuploidy is sufficiently accurate to be offered alongside autosomal aneuploidy screening with specific pretest counseling and consent [27]. ISPD strongly recommends genetic counseling for all patients with a high-chance cfDNA result and offering diagnostic testing to these patients.

They also recommend offering at least one first-trimester ultrasound scan for dating, diagnosis of multiple pregnancy, and confirmation of fetal viability before performing cfDNA screening, and offering diagnostic testing (chromosomal microarray) for fetuses with ultrasound abnormalities, including nuchal translucency measurement ≥3.5 mm, regardless of the prior cfDNA result.

Additional society guideline links can be found separately. (See 'Society guideline links' below.)

COUNSELING — Despite professional recommendations, a number of these tests are available commercially, including cfDNA for rare autosomal trisomies, cfDNA for microdeletions and duplications, and cfDNA for select autosomal-dominant disorders. Patient pre- and posttest counseling prior to receiving these tests is becoming increasingly complex.

Pretest counseling involves [10]:

●Scope and nature of the conditions being tested.

●Test characteristics (detection rate, false-positive rate, no call rate). A description of the meaning of false-positive and false-negative tests, and that false negatives may be more common for rare conditions.

●The recommendation to confirm positive results through invasive testing.

●Detection of disorders of maternal or paternal health relevance.

●Uncertain findings associated with mosaicism and unexpected findings.

Posttest counseling involves:

●Offering invasive testing to any patient who has cfDNA screen-positive result.

●Referring all patients with positive results to a genetic counselor to discuss the pros and cons of different modalities for invasive genetic testing. Parental testing should be considered (maternal for aneuploidies, both parents for single gene) to best inform the cfDNA interpretation.

●Informing all patients with a screen-negative that a negative test does not rule out the possibility of an affected fetus.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Prenatal genetic screening and diagnosis".)

SUMMARY AND RECOMMENDATIONS

●General obstetric population – Prenatal cell free DNA (cfDNA) screening for fetal genetic variants other than common autosomal aneuploidies (21, 18, 13) is technically possible and commercially available (see 'Analytic methods' above). Professional medical societies generally recommend not using cfDNA tests to screen for microdeletions and duplications, monogenic disorders, or rare autosomal trisomies in the general obstetric population because of lack of high-quality evidence of detection rates, false-positive rates, and the clinical significance of a positive test. However, the American College of Medical Genetics (ACMG) does recommend offering cfDNA screening for fetal sex chromosome aneuploidy, and suggests its use for 22q11.2 deletion syndrome. (See 'Screening for a limited panel of microdeletions/microduplications' above and 'Screening for monogenic disease' above and 'Screening for rare autosomal trisomies' above and 'Screening for sex-chromosome aneuploidy' above and 'Positions of selected professional societies' above.)

●Special clinical scenarios

•Fetal RhD status – Prenatal cfDNA testing is an accepted approach for determining fetal RhD status in pregnancies complicated by maternal alloimmunization as it informs pregnancy management: titers are monitored if the fetus is RhD positive and can be avoided if RhD negative. Test results can also be used to optimize use of anti-D immune globulin by avoiding administration to RhD-negative pregnant people carrying a RhD-negative fetus. (See 'Fetal RhD status' above.)

•Fetal sex – Prenatal cfDNA testing can be useful for early identification of fetal sex when this information is medically important for decision-making, such as in pregnancies at risk for fetal congenital adrenal hyperplasia (CAH), hemophilia, or Duchenne muscular dystrophy. For these X-linked conditions, parents may pursue an invasive procedure for diagnostic testing for predicted male fetuses and avoid such procedures and testing for predicted female fetuses. (See 'Suspected sex-linked disorders' above.)