Expanded Carrier

S cree ning
Anthony R. Gregg, MD, MBA

KEYWORDS
 Prenatal carrier screening  Expanded carrier screening  Genetic counseling
 Perceptions of Uncertainties in Genome Sequencing (PUGS) scale

KEY POINTS
 The goal of prenatal carrier screening is to provide information to couples considering or
with an ongoing pregnancy.
 Couples who select this elective screening have an opportunity to learn whether there is
mendelian risk rather than population risk of an affected fetus.
 Couples that select screening have the option of screening for a limited number of condi-
tions, or an expanded number of conditions.
 The Perception of Uncertainties in Genome Sequencing scale offers a framework around
which pretest and posttest counseling can be considered.

SCREENING

The term screening is important to understand. Classically, this refers to applying an

imperfect test (eg, Papanicolaou smear for cervical cancer) or process (eg, travel
questionnaire used during the Ebola epidemic) to identify asymptomatic patients
within a population. The imperfection in a screening test refers to the acceptance of
false-positive and false-negative results to improve case finding. Carrier screening
in genetics attempts to determine, among asymptomatic patients, those capable of
passing genes or genetic risk to future generations. This context refers to the popula-
tion addressed by screening. Prenatal carrier screening is intended to identify asymp-
tomatic, reproductive-aged individuals, so that risk of carrying an affected child can be
specified. The target population is couples who are pregnant or planning to become
pregnant.1 There are 2 other dimensions to which the word screening is attached.
These dimensions relate to the number of genes (ie, conditions) and the number of ge-
netic variants. The ability to analyze multiple genes and variants simultaneously (with
the same sample) set the stage for screening panels. These panels first reflected

Conflict of Interest: The author reports no conflict of interest.

Department of Obstetrics and Gynecology, University of Florida College of Medicine, PO
Box 100294, Gainesville, FL 32610-0294, USA
E-mail address: [email protected]

Obstet Gynecol Clin N Am 45 (2018) 103–112

https://doi.org/10.1016/j.ogc.2017.10.005 obgyn.theclinics.com
0889-8545/18/ª 2017 Elsevier Inc. All rights reserved.
104 Gregg

genotyping (slow throughput) efforts to return results for more than one gene and
variant simultaneously. With current sequencing technology, reported genes and var-
iants are now arbitrary and reflect inconsistent efforts by companies to return results
for conditions of interest to society, providers, and their patients. Advances in
sequencing technology led to a new term, expanded carrier screening (ECS). In the
prenatal setting this expansion allows consideration of panethnic screening, which
means the population is not restricted. Expansion also means the number of genes
and their variants reported is a function of the laboratories and any restrictions placed
on them (eg, time, reimbursement, or regulation). In the clinical setting, ECS has addi-
tional implications. Prenatal carrier screening has extended beyond asymptomatic
carriers of recessive conditions to include presymptomatic carriers of X-linked and
semidominant conditions. The application of screening now has a spectrum and
this key concept can be depicted using a Rubik cube (Fig. 1).
The Rubik cube matrix has 3 axes. The X axis depicts the populations screened, the
Y axis the variants considered in the screening strategy, and the Z axis represents the
genes or conditions being evaluated. Red, blue, and green dots denote the magnitude
along each axis from highest (red) to lowest (green). The most intensive screening pro-
gram would be to interrogate an entire population (panethnic; X axis) using sequencing
technology (Y axis) and then reporting out information for a large number of genes (Z
axis). The least intensive (all green dots on the cube) would be the application of po-
lymerase chain reaction for delta F508 in a family with a proband known to have cystic
fibrosis (homozygous delta F508) based on clinical symptoms and prior application of
a diagnostic panel. The least intensive application (all green on the cube) is one in
which genotyping methods are used for diagnostic purposes. In this case, the risk
of affected family members is mendelian in magnitude. When the risk of an asymptom-
atic carrier moves from population-based (eg, 1/25 for cystic fibrosis) risk to

Fig. 1. The Rubik cube model of prenatal diagnosis and carrier screening has 3 axes: popu-
lation, screened variants, and genes. The red, blue, and green dots indicate stratification
across the spectrum. CF, cystic fibrosis; PCR, polymerase chain reaction; S, syndrome.
 Expanded Carrier Screening 105

mendelian (eg, 1/3 for an adult sibling of an affected family member), this moves along
the axes of the cube from a screening to a diagnostic paradigm.

HISTORICAL PERSPECTIVE

Genotyping technology available in the early 1990s allowed the molecular diagnosis of
conditions characterized by a severe phenotype (eg, cystic fibrosis) within families. At
that time the goal was to interrogate a known gene and only a few known pathogenic
alleles. High throughput was not a characteristic of these approaches. This process
was molecular inquiry for diagnostic purposes. As stated, this utility is represented
by the green dots on the Rubik cube axes (see Fig. 1). Professional organizations
considered the application of genotyping technologies for individual conditions2,3
and for specific ethnic groups.4 Prenatal carrier screening used genotyping with the
intent of identifying recessive alleles in asymptomatic people. The conditions/genes
chosen for screening were enriched in subgroups of the population and were chosen
by virtue of the known pathogenic variants, known genes, and severe phenotypes.
Two major scientific accomplishments changed the speed at which screening for
mendelian conditions was adopted. Publication of the first complete draft of the hu-
man genome sequence5,6 led the way. The term draft is used to recognize that regions
of the genome were/are not fully characterized, which means that a negative
screening result always leaves a residual risk (RR; a chance that the patient remains
a carrier). RR is calculated as: carrier frequency in a population screened  (1 detec-
tion rate for that population). Over time, sequencing data uncover new pathogenic var-
iants and some pathogenic variants are determined to be nonpathogenic as more
clinical information becomes available.7,8 These changes resulting from new informa-
tion increase and decrease the detection rate respectively. Furthermore, population
carrier frequency varies by ethnicity. In populations characterized as closed with
respect to reproduction, fewer pathogenic variants account for a large proportion of
disease.9 For example, in the Ashkenazi Jewish population, 2 pathogenic variants in
the HEXA (hexosaminidase A) gene cause more than 90% of the severe disease
phenotype. In non–Ashkenazi Jewish, the same 2 variants account for 35% of the se-
vere disease.10
Hardy-Weinberg equilibrium can be used to determine the frequency of a pheno-
type within a population.11 However, this can only be done under specific conditions
(Box 1, Table 1). The binomial equation is used when there are 2 possible alleles, p
and q. Allele frequency can be determined by counting chromosomes with p or q

Box 1
Hardy-Weinberg equilibrium: criteria

Large population
Random mating
Reproductive fitness is not altered by alleles
No migration to affect allele frequency
No genetic drift to affect allele frequency
Alleles do not undergo mutations

Data from Hardy GH. Mendelian proportions in a mixed population. Science 1908;28(706):49–
50.
106 Gregg

Table 1
Hardy-Weinberg equilibrium: equations

Hardy-Weinberg (p 1 q)2 5 q2 1 2pq 1 p2

Sum of all alleles p1q51
Alleles at a locus (n) (p1 1 p2 1 .pn)2

q may be the frequency of the variant allele or all variant alleles, depending on use of equation.
p is the frequency of the wild-type (normal) allele in the population.
n is the number of alleles possible at a specific locus.
Data from Hardy GH. Mendelian proportions in a mixed population. Science 1908;28(706):49–50.

alleles in a population or identifying disease frequency (q2). Taking the square root pro-
vides q. The equation 1 q gives us p. Once allele frequencies are known, the bino-
mial provides frequencies for homozygous unaffected (p2), heterozygous carriers
(2pq) and those affected (q2).
The second achievement was (and continues to be) advancing sequencing technol-
ogy so that the high price tag to sequence the first human genome (0.5–1 billion dol-
lars) decreased precipitously to $1000 or less. With significant funding through the
National Human Genome Research Institute (NHGRI), sequencing technology has
advanced in such a way that high-throughput sequencing at low cost for an entire
exome is now possible.12,13 All exons (regions of the genome that code amino acids)
of a person’s genome are collectively called an exome. Sequence data for an entire
exome can now be obtained in less than 3 days. The analysis and link to clinical impor-
tance (bioinformatics) are now rate-limiting steps that determine clinical utility. Cost,
speed, and reproducibility of nucleotide sequence are no longer limiting. Referring
to the Rubik cube, in less than 3 decades progress has moved from the green on
the axes to the red. These scientific accomplishments moved screening into the multi-
dimensional application referred to as ECS.

PURPOSE

The first iteration of panethnic prenatal carrier screening focused on the recessive
conditions, cystic fibrosis and spinal muscular atrophy.14,15 The purpose was to apply
genotyping technology to all asymptomatic reproductive-aged women and their part-
ners planning pregnancy or with an early ongoing pregnancy. Screening results can be
used to calculate the risk of having a fetus affected by either cystic fibrosis or spinal
muscular atrophy. With this information, couples can make reproductive decisions
and educate themselves around the needs of children with these disabilities. These
two conditions allow risk estimation because they are single-gene recessive disorders
with known genetic variants whose allele frequencies are known (Table 2). The pop-
ulation carrier frequency (line 1 in Table 2) and the detection rate (line 3 in Table 2)
are the important variables that the determine risk for any recessive condition before
screening or after screening.
The latest iteration of panethnic carrier screening incorporates next-generation
sequencing (NGS) technology or the ability to sequence across regions of the
genome that include genes of interest. The access to more data has changed
the focus. Screening now includes not only common recessive conditions but
rare recessive conditions, semidominant (eg, factor V Leiden) conditions, and
X-linked (eg, fragile X mental retardation) conditions. Importantly, women and their
partners may learn they are not only asymptomatic carriers but presymptomatic
 Expanded Carrier Screening 107

Table 2
Example of individual residual and pregnancy risk assessment for cystic fibrosis in northern
Europeans

Description Formula Risk

1 Population carrier frequencya — 1/25
2 Pregnancy risk before screening 1/25  1/25  1/4 1/2,500
3 Screening DRb — 9/10
4 RRc 1 DR 1/10
5 Individual risk after negative screen 1/25  1/10 1/250
6 Pregnancy risk after negative screening 1/250  1/250  1/4 1/250,000

Abbreviation: DR, detection rate.

a
Ethnic specific (eg, northern European, African American, Ashkenazi Jewish).
b
Function of number of mutations and population screened, in this case American College of
Medical Genetics panel for cystic fibrosis.
c
RR is the risk of being a carrier when screening test is negative.

carriers too. For example, young women (<40 years old) identified as carriers of
fragile X with 40 to 199 copies of the fragile X mental retardation repeat (CGG)
are at risk for premature ovarian failure. This condition was observed in 2.2% to
18.6% of carriers compared with a general population frequency of w1%. Men
with repeats between 55 and 199 are at risk for fragile X–associated ataxia tremor
syndrome.16,17
Some patients want maximum information (the most genes/conditions and the most
variants), whereas others do not. It is critically important that, through counseling, the
goals of women and their partners align with the results that will be delivered to them.
Some people do not want to know that they are at increased risk for later-onset con-
ditions; they may only want information about fetal risk. Although it is important to
know which Rubik cube block most closely meets each person’s goals, further refine-
ment may be necessary.

CARRIER FREQUENCY AND RESIDUAL RISK

One benefit patients receive from genetic counseling after screening for single-gene
disorders is the calculation of the RR of being a carrier even after receiving negative
test results. RR stems from an inability to screen and detect all possible genomic var-
iations in all genes that cause the conditions being screened. Patients who screen
negative can have a child affected with the condition being screened because of a fail-
ure to recognize or understand variants as pathogenic or decisions not to report all
variants. False-negative and false-positive results are a feature of screening tests.
Those providing genetic counseling know that all disease-causing variants are not
elucidated and convey to patients that results after genetic testing are incomplete.
This is true whether low-throughput genotyping or NGS platforms are used.
With the implementation of ECS, RR is difficult to measure and harder for patients
and counselors to calculate. The number of conditions screened is large and summa-
tion
Pn of RR for each condition (i) is required. The formula for this summation is:
i 5 1 RRi (where n is the number of conditions being screened). Because the fre-
quency of rare population variants is not usually known, the contribution of each to
the detection rate (DR) in a general population is not known. RR is a function of the
DR (ie, RR51-DR), the inability to calculate DR and the large number of conditions
screened make calculating a meaningful RR elusive (see Table 2). In addition, it would
108 Gregg

be impractical to provide RR estimates for each condition separately during pretest or

posttest counseling sessions.
Many companies offer ECS. However, the number of conditions screened and the
specific conditions screened for vary. Furthermore, the number of pathogenic variants
reported for each condition is not uniform across companies that offer screening;
therefore, DR varies across companies. Although laboratories are aware of the criteria
used to define variant categories, there are no regulations ensuring that, after NGS, all
variants identified and reported are valid. Validating rare variations in the genome can
be a difficult task, so the conditions included for reporting and the variants reported
are prone to becoming marketing tools and may not always contribute to the goals
of population-based carrier screening.

VARIANTS AND CONDITIONS

Variant classification was defined through a joint workshop of the Association of Mo-
lecular Pathology and the American College of Medical Genetics and Genomics in
2013.18 Importantly, the terms mutation (ie, permanent change in the genome) and
polymorphism (ie, change in the genome that is present in at least 1% of the popula-
tion) were eliminated from the current classification system. The term variant replaced
both terms and was further divided into a 5-tier classification scheme (Table 3). The
likely pathogenic and likely benign tiers carry at least 90% certainty.
The driver of conditions that should be reported on carrier screening panels is
controversial and ranges from a view that anything that can be reported should be re-
ported to a more conservative view that only conditions that are frequent, severe, and
actionable should be included. The problem relates to who defines each of these vari-
ables. Without citing data and by consensus, the American Congress of Obstetricians
and Gynecologists (ACOG) determined that only conditions with a frequency of 1 in
100 or greater should be included.19 When health care providers were surveyed,
they rated a shortened life span from birth to adolescence and conditions that affect
intellectual functioning as the most severe.20 Actionable conditions are difficult to
define. The decisions made by prospective parents once they become aware of an
affected pregnancy are personal and range from doing nothing to learning as much
as possible before the anticipated birth to pregnancy termination. These actions are
likely heterogeneous across regions of the United States. Therefore, until more is
learned about the factors that determine how individuals and couples weigh severity
and action ability, clinicians are left with a frequency of 1 in 100 (based on no data) as a
deterrent to industry marketing strategies.

IMPLEMENTING EXPANDED CARRIER SCREENING

Professional organizations move more slowly compared with the pace of technology.
Without advocating for or against ECS, professional organizations coalesced to

Table 3
Nucleotide sequence variation classification

Pathogenic Certain
Likely pathogenic >90% certainty
Uncertain significance —
Likely benign >90% certainty
Benign Certain
 Expanded Carrier Screening 109

provide a framework that can be useful when implementing ECS into clinical prac-
tice.21 There are two basic approaches for prenatal carrier screening:
1. Gene-by-gene screening of a limited number of conditions based on ethnicity
2. ECS, which screens many conditions and variants simultaneously and is panethnic

Advantages and Disadvantages

As the number of conditions and variants reported increases in a screening program,
the proportion of people who screen positive also increases. People more often screen
positive for something. The advantages of ECS compared with gene-by-gene
screening include the opportunity to learn more. Learning more means more
information can be shared with family members. Furthermore, identifying presymptom-
atic risk allows the development of strategies to modify risk and stay informed. For
some patients, these are reasons enough to choose ECS. For others, learning more
may bring anxiety, especially when presymptomatic conditions are revealed. Anxiety
can also emanate from the identification of variants with uncertain significance.
Selection of ECS carries disadvantages beyond potential anxiety. Carrier screening
can be performed sequentially (one partner followed by the second based on results)
or simultaneously. At baseline, simultaneous screening using NGS costs more than
sequential screening. As stated earlier, the probability of at least one member of a
couple screening positive is very high when sequential screening is employed, costs in-
crease when a carrier is identified and the partner choses NGS in an effort to learn about
all possible variants. This sequence can result in misinformation due to low stringency
reporting of variants. Further counseling is often necessary and counseling costs follow
laboratory costs. A review of the advantages and disadvantages of gene-by-gene
screening compared with ECS is best worked through during pretest counseling.

Counseling
A basic tenet in genetic counseling is to provide nondirective counseling.22 For this to
take place, patients must be educated in a medically literate fashion. All patients
should be familiar with the concepts presented in the Rubik cube (see Fig. 1), and
they should be introduced to the technology platform and that it is imperfect. The con-
cepts of DR and the relationship to RR should be discussed. Effective counseling re-
quires that providers understand the principles of mendelian inheritance. Importantly,
the patient’s goals when participating in screening should be understood.
ACOG recommends offering panethnic carrier screening for 2 conditions: cystic
fibrosis and spinal muscular atrophy. A complete blood count is a general screen
for anemia and is a first-line screen for some hemoglobinopathies. Hemoglobin
electrophoresis is recommended for patients with African, Mediterranean, Middle
Eastern, southeast Asian, or West Indian ancestry. ACOG also recommends offering
Ashkenazi Jews screening for Tay-Sachs and Canavan diseases as well as familial
dysautonomia.1
When patients select screening for more conditions than those recommended by
ACOG independent of ethnicity, they are selecting ECS and they may receive results
for some conditions that could affect their own health. These conditions include domi-
nant, semidominant, and X-linked conditions. It is not necessary for those who provide
pretest counseling to be familiar with the treatment, prognosis, and clinical course of
all conditions included on ECS panels.
The hemoglobinopathies represent a group of conditions characterized by abnor-
malities of alpha or beta chain production or oxygen affinity. The use of molecular
testing is associated with a greater RR compared with hemoglobin electrophoresis
110 Gregg

in screening for hemoglobinopathies, because the pathogenic variants that result in

b-thalassemia are often private (specific to families). Furthermore, the genomic
changes associated with a-thalassemia are deletions. The most severe a-thalas-
semia phenotype requires deletions in cis (on the same chromosome). Although
NGS can determine dosage (the number of copies of a gene), a limitation of NGS
is its inability to determine cis or trans (on opposite chromosomes).
Tay-Sachs disease is often cited as a condition for which NGS is not well suited.
Some laboratories use NGS and report the results for 3 common pathogenic variants
observed in Ashkenazi Jews. For non–Ashkenazi Jews and in cases in which 1 mem-
ber of a couple is a known carrier, enzyme analysis is preferred. Because of concerns
around ethnic intermixing, enzyme analysis is generally preferred to molecular testing
and results in the lowest RR after screening.21
There is agreement that the best time to offer carrier screening is before conception.
Carrier screening should be offered as early as possible during an ongoing pregnancy.
Although sequential screening may be well suited before an ongoing pregnancy,
simultaneous screening seems to be the best approach when screening is performed
during an established pregnancy.
Importantly, when a family history of a mendelian disorder is identified or already
known, referral to a trained genetic professional is indicated. Confirmation of the dis-
order, genetic variant in the family, and the determination that the familial variant is
included on a carrier panel or as part of NGS results is crucial. When the familial
variant is not included, special efforts are required to properly inform the concerned
couple.

Pretest and Posttest Counseling

NGS used for carrier detection brings uncertainties. A useful approach when providing
pretest counseling is to consider the Perceptions of Uncertainties in Genome
Sequencing (PUGS) scale framework. Although this scale was introduced as a tool
to evaluate patient perceptions of uncertainties in the use of genome sequencing,23
it highlights the need to address specific areas in pretest and posttest counseling.
This scale considers 3 general domains: clinical (ie, health care implications), affective
(ie, impact of results on psychosocial health and behavior), and evaluative (whether re-
sults are reliable and actionable). These domains translate into important pretest and
posttest counseling concerns by patients and reflect areas that should accordingly be
discussed (Tables 4 and 5).

Table 4
Pretest counseling using Perceptions of Uncertainties in Genome Sequencing domains

Domain # Consideration by Those Screened

Clinical 1 Conditions reported have varied severity
Affective 1 Screening is voluntary
2 Results are confidential and protected by GINA
3 Plan for returning results
Evaluative 1 Requires accurate paternity information
2 RR remains after a negative screen
3 May identify a recessive condition in an asymptomatic person
4 Asymptomatic carriers of semidominant, dominant, and X-linked disorders
may be identified

Abbreviation: GINA, Genetic Information Nondiscrimination Act (signed into US law by President
George W. Bush).
 Expanded Carrier Screening 111

Table 5
Posttest counseling using Perceptions of Uncertainties in Genome Sequencing domains

Domain # Consideration by Those Screened

Clinical 1 Access to reliable information regarding conditions for which both members
of couple screen positive should be provided
Affective 1 A copy of carrier screening results should be provided
2 Provide written information that can be used for sharing positive screening
results with other family members if desired
Evaluative 1 Pregnancies are considered low risk when couples have discordant results
(one screens negative and the other positive)
2 When one partner screens positive and the other negative for Tay-Sachs, the
negative partner should be screened with enzyme analysis

SUMMARY

Prenatal carrier screening has expanded to include a larger number of genes and
variants offered to all couples with ongoing pregnancy or considering a pregnancy.
Panethnic screening for cystic fibrosis and spinal muscular atrophy and screening
for a limited number of conditions based on ethnicity are recommended by ACOG.
Screening is elective and counseling should be nondirective. As providers offer
and patients consider ECS, their goals should align with those of the laboratories
used for screening. This process includes consideration of ethnicity, as well genes
and variants being reported. Although RR calculations have become an obsolete
part of posttest counseling when ECS is selected, it is still important for patients
to understand that there is a risk that they may be a carrier for a condition in the
setting of a negative screening test result. The PUGS scale offers a useful under-
standing of the pretest and posttest counseling concerns that should be considered
as part of ECS implementation.

REFERENCES

1. ACOG Committee Opinion Summary 691: carrier screening for genetic condi-
tions. Obstet Gynecol 2017;129:597–9.
2. Workshop on Population Screening for the Cystic Fibrosis Gene. Statement from
the National Institutes of Health workshop on population screening for the cystic
fibrosis gene. N Engl J Med 1990;323:70–1.
3. ACOG Committee Opinion 101: current status of cystic fibrosis carrier screening.
Int J Gynaecol Obstet 1992;39:143–5.
4. ACOG Committee Opinion 298: prenatal and preconceptional carrier screening
for genetic diseases in individuals of Eastern European Jewish descent. Obstet
Gynecol 2004;104:425–8.
5. Lander ES, Linton LM, Birren B, et al. Initial sequencing and analysis of the hu-
man genome. Nature 2001;409:860–921.
6. Venter JC, Adams MD, Myers EW, et al. The sequence of the human genome.
Science 2001;291:1304–51.
7. Grody WW, Cutting GR, Klinger KW, et al. Laboratory standards and guidelines
for population-based cystic fibrosis carrier screening. Genet Med 2001;3:149–54.
8. Watson MS, Cutting GR, Desnick RJ, et al. Cystic fibrosis population carrier
screening: 2004 revision of American College of Medical Genetics mutation
panel. Genet Med 2004;6:387–91.
112 Gregg

9. Abelovich D, Lavon IP, Lerer I, et al. Screening for five mutations detects 97% of
cystic fibrosis (CF) chromosomes and predicts a carrier frequency of 1:29 in the
Jewish Ashkenazi population. Am J Hum Genet 1992;51:951–6.
10. Kaback MM, Desnick RJ. Hexosaminidase A deficiency. Gene Reviews. 2001. Avail-
able at: https://www.ncbi.nlm.nih.gov/books/NBK1218/. Accessed September 30,
2017.
11. Hardy GH. Mendelian proportions in a mixed population. Science 1908;XXVIII:
49–50.
12. Reuter JA, Spacek DV, Snyder MP. High-throughput sequencing technologies.
Mol Cell 2015;58:586–97.
13. Heather JM, Chain B. The sequence of sequencers: the history of sequencing
DNA. Genomics 2016;107:1–8.
14. American College of Obstetricians and Gynecologists, American College of Med-
ical Genetics. Preconception and prenatal carrier screening for cystic fibrosis;
clinical and laboratory guidelines. Washington, DC; Bethesda (MD): ACOG;
ACMG; 2001.
15. Prior TW. Carrier screening for spinal muscular atrophy. Genet Med 2008;10:
840–2.
16. Sherman S, Pletcher BA, Driscoll DA. Fragile X syndrome: diagnostic and carrier
testing. Genet Med 2005;7(8):584–7.
17. Loesch D, Hagerman R. Unstable mutations in the FMR1 gene and the pheno-
types. Adv Exp Med Biol 2012;769:78–114.
18. Richards S, Aziz N, Bale S, et al. Standards and guidelines for the interpretation
of sequence variants: a joint consensus recommendation of the American Col-
lege of Medical Genetics and Genomics and the Association for Molecular Pa-
thology. Genet Med 2015;17(5):405–8.
19. ACOG Committee Opinion Summary 690: carrier screening in the age of genomic
medicine. Obstet Gynecol 2017;129:e35–40.
20. Lazarin GA, Hawthorne F, Collins NS, et al. Systematic classification of disease
severity for evaluation of expanded carrier screening panels. PLoS One 2014;
9:e114391.
21. Edwards JG, Feldman G, Goldberg J, et al. Expanded carrier screening in repro-
ductive medicine–points to consider: a joint statement of the American College of
Medical Genetics and Genomics, American College of Obstetricians and Gyne-
cologists, National Society of Genetic Counselors, Perinatal Quality Foundation,
and Society for Maternal-Fetal Medicine. Obstet Gynecol 2015;125:653–62.
22. Rogers CR. Counseling and psychotherapy. Cambridge (MA): Houghton Mifflin;
1942.
23. Biesecker BB, Woolford SW, Klein WMP, et al. PUGS: a novel scale to assess per-
ceptions of uncertainties in genome sequencing. Clin Genet 2017;92:172–9.