OR WAIT 15 SECS
Expanded screening answers questions that were unanswerable in the past, but brings with it new uncertainties.
Dr Gregg is the BL Stalnaker Professor, Chief of Maternal-Fetal Medicine and Director of Obstetrics at UF Health Shands Hospital, University of Florida, Gainesville. He has no conflicts of interest to report in respect to the content of this article.
Next generation sequencing (NGS) technology represents a challenge to traditional prenatal genetic screening for recessive conditions. Specifically, what has been dubbed “expanded carrier screening” (ECS) makes gene-by-gene or a la carte genetic screening menus obsolete. Before the advent of NGS, sequencing was labor-intensive, costly, and slow, and it was difficult to screen large numbers of people for multiple genetic conditions and to evaluate many pathogenic variants (aka mutations) within the same gene. With NGS, genetic information (nucleotides) can be evaluated (sequenced) rapidly and at low cost.1
The hierarchical approach to carrier screening for single-gene disorders (positive family history followed by racial/ethnic identification) has evolved to include panethnic, multiple-condition screening, using more pathogenic variants. Expanded carrier screening utilizes genetic technology, panels with specific genes and variants as targets or sequencing strategies with results for specific genes whose pathogenic variants are selectively reported. These technologies can be applied using the hierarchical approach or they can be used in a panethnic fashion to identify carriers in those with unsuspecting ancestry or unknown family history. Importantly, when family history for a single gene disorder is positive, ECS may not be the best choice because family-specific and already-identified pathogenic variants may not be included on commercially available screening panels. Genetic counseling by a trained professional should be considered before implementing ECS in cases of a family history of a single-gene disorder. At the heart of ECS is a requirement that patients and providers understand 2 basic principles.
Principle 1. A population-based approach is used to identify carrier parents who are at risk of having a child with a serious, often intellectually disabling, heritable condition. Traditionally screening focused on recessive conditions, in which mother and father are carriers but do not manifest a phenotype. Today’s ECS “panels,” however, frequently include semi-dominant (eg, factor V Leiden, prothrombin gene, alpha 1- antitrypsin) and sex linked (eg, fragile X mental retardation) conditions. This means that screened parents can learn that they have a previously unrecognized health risk. These conditions might also pose health implications for the unborn child, children already in the home, and relatives. The contrary is also true. Some genes and variants included for screening by ECS laboratories are dubious with respect to pathogenicity. A prime example is MTHFR, and there are others. Furthermore, some conditions on panels demonstrate variable penetrance (age of onset or absence of expression) and variable expressivity (phenotypic variation). When ECS approaches fail to consider these concerns, patient anxiety and cost can escalate.
Adopting ECS either selectively or panethnically clearly can add complexity to posttest counseling compared to gene-by-gene screening for cystic fibrosis (CF) and spinal muscular atrophy (SMA). This aspect of ECS has troubled many and has led to the slow adoption of ECS into clinical practice. One response by industry has been to allow patients and providers to have some autonomy in selecting conditions to be screened. Another response has been to remove potentially important genetic information from screening panels (eg, Factor V Leiden, prothrombin gene). An ethical argument can be made that if health risks of a parent or other family member can be mitigated by results obtained after ECS, patients should be allowed access to this information. If the ethical argument holds, then obstetricians should not be troubled if patients are informed about their own health risks. Recall that when performing fetal imaging patients are not counseled that incidental findings are possible. They are not told that these incidental findings could inform patients of increased health risks. Often ovarian cysts and placenta implantation concerns are identified, but they do not lead to a call to abandon fetal imaging, nor is concealment from the patient an option. Relying on a knowledge base, the response is to convey accurate and useful information. Pretest counseling today requires a brief discussion so that patients understand results of screening may inform them of their own health risks and risk to other family members.
A “serious” condition is easy to recognize by the ravages it imposes. Likewise, the unaffected or unimpaired are easy to spot. Moving from each end of the spectrum creates a problem in defining “serious,” because the term has many shades of gray. The most appropriate conditions for prenatal ECS are difficult to name when they are not the most or least severe. This is due in part to variable expressivity (a spectrum of signs and symptoms that may be expressed differently among persons with the same genetic condition) and/or variable penetrance (proportion of people that will manifest signs or symptoms despite having a pathogenic variant in a disease-causing gene). Just as patients with placenta previa, preterm labor, and seizure disorder can have a varied clinical course, so too can patients with cystic fibrosis or sickle cell disease.
The problem also lies in the phenotype of the conditions screened. Debate over the conditions to include within the gray zone lacks scientific inquiry and recognition that this may be best determined by a patient or the family undergoing screening. Educated healthcare professionals were asked to rank phenotypes by category6 (x-axis in Figure 1). Conditions that impact lifespan (especially in infancy and childhood), affect intellectual function, and those that can be treated were considered most appropriate for screening (Tier 1). Conditions that impair fertility rank least important (Tier 4). In between are a host of conditions that were not easy to separate (Score of 6-8), including conditions characterized by reduced adult life expectancy, birth defects (especially those that impact mobility), vision loss, cancer predisposition, deafness, and mental health.6 Expanded carrier screening can assess conditions across the spectrum. Pretest counseling should include a general discussion of the types of conditions being screened, emphasizing “spectrum” and patient-specific importance.
Principle 2. A negative screening test does not eliminate the possibility that a patient is a carrier.
Recall that residual risk is risk that remains after a negative test result. Ob/gyns are familiar with this concept. After a negative mammogram it is still possible that the test missed a lesion. The same can be said of colposcopy, colonoscopy, and other forms of screening. Before ECS, calculating residual risk was easy (carrier frequency x (1- detection rate)). The factors required to calculate residual risk, populations specific carrier frequency, and detection rate were known when gene-by-gene screening was performed on targeted racial ethnic groups. This calculation was possible because screening occurred for relatively few conditions that had high visibility compared to conditions on ECS platforms. Today, some conditions screened for are so rare that detection rate and populations carrier frequency are unknown. Adding more pathogenic variants to the screen will improve the detection rate, but the zeal to do this at low cost has been complicated by the observation that not all variants screened are pathogenic. False positives occur when clinical validity of variants is not considered. False-negative and false-positive results are inherent in any screening program. The solution is not to abandon the technology, but to rely on resources that can facilitate interpretation of test results.
Negative results are straightforward. Patients can be told they were screened for a large number of conditions and that the results for those specific conditions suggest a very low likelihood that they are carriers. Emphasize that it is still possible to be a carrier, because there is always a residual risk. Therefore, the chance of having an affected child for the conditions screened will depend on the carrier frequency in the population and the variants tested. It may be difficult to calculate the residual risk for each condition screened, because population carrier frequency (2pq) may be unknown, or ethnic-specific and the percent contribution of specific variants may be unknown.
When results are positive, check that the variant reported is pathogenic before providing information to patients. Laboratory-employed genetic counselors can provide information and are an important resource. Web-based resources may also be helpful (eg, genetics home reference, gene reviews). Consider the utility of the Clinical and Functional Translation of CFTR site (www.CFTR2.org) to clarify pathogenic variants in cases in which a rare CFTR variant is identified. In obstetrics and gynecology there is often a need to consult radiologists and pathologists after receiving non-genetic test results. Providers readily search Internet resources, textbooks, and journals in order to convey accurate information to patients. The principle with ECS and genetic testing in general is no different; enlist help when needed.
It is virtually impossible to accurately calculate residual risk for each condition on an ECS panel (although some laboratories report values). Genetic counselors and others have expressed concern over lack of precise residual risk data. When discussing ECS it may be helpful to consider residual risk as diagrammed (Figure 2). On the x axis are the conditions being screened. Risk is on the y axis. Each condition is depicted as a filled circle before screening (eg, pretest risk = population carrier frequency) and an open circle after screening (eg, posttest risk = residual risk). The downward arrow depicts the detection rate. This arrow is elongated by a employing more pathogenic variants in the screen and/or when the pathogenic variants screened make up a substantial portion of disease burden. CF, a condition with relatively high carrier frequency in certain populations (A in the figure), is screened using mutations that move the arrow towards the x axis (eg, zero risk), but the residual risk never reaches this position. This is because laboratories do not screen/report all of the nearly 400 variants that might be present. Condition B is frequent, but less common than CF, similar to that of SMA with a low residual risk after screening (~98% detection rate). Hurler syndrome is depicted by C, low carrier frequency and modest detection rates leaving a residual risk greater than that for SMA. Notice that some conditions have a low residual risk by virtue of their low carrier frequency even when the number of variants screened is minimal (C in the figure). Likewise, some conditions may have a high residual risk because of a high carrier frequency and few pathogenic variants are screened.
In the era of NGS, gene-by-gene screening is replaced by efforts to screen the aggregate of known recessive disorders. Therefore, we must consider how residual risk across a spectrum of conditions (common to rare) changes the residual risk for the entire pregnancy. The addition of conditions, from common to rare, to an ECS platform causes the residual risk of the entire pregnancy to fall, but not at a constant rate. Screening first for common conditions with known common pathogenic variants has the greatest impact. As conditions are added, the number of rare conditions screened increases. Eventually, conditions with modest carrier frequency and low detection rates do little to further lower the residual risk of the pregnancy (Figure 3).
Why then do laboratories that offer ECS continually add to their platforms conditions with which even trained geneticists are unfamiliar and that can befuddle a clinician when a carrier patient is encountered? First, the additional cost is trivial. Second, the motive may be altruistic. Patients may benefit from the addition of very rare conditions. Third, as the ECS space becomes more competitive, companies will need to do more marketing to survive.
Two serious and hidden pitfalls, which are recognized by most geneticists, lurk in the condition menu of ECS platforms.
Hemoglobinopathies are best screened using red cell indices on the complete blood count (CBC) and hemoglobin electrophoresis. The ACOG guideline on this topic remains relevant today.7 Every clinically significant point mutation in the β chain is not on a molecular based “panel” and α-thalassemia deletions are not usually on carrier screening platforms.
Carrier testing for Tay-Sachs disease is complicated by a wide range of carrier frequencies across racial ethnic groups. Furthermore, there is a lower detection rate when the carrier frequency is low (short arrow, Figure 2) as is the case in certain populations. The pathogenic variants used on ECS platforms for low carrier frequency groups are not robust. Therefore carrier testing using NGS as it is applied is inefficient. The filled circle (Figure 2) is constrained and doesn’t move to a lower position when starting in a low carrier frequency range. Screening for Tay–Sachs disease is performed most efficiently using hexosaminadase A enzyme analysis for low carrier frequency groups (non-Ashkenazi). Because of racial/ethnic mixing, it may be best to include this as part of any screening approach.
Carrier screening can be performed before conception or antenatally, and sequentially (maternal then paternal) or simultaneously Most patients are not screened in the preconception period. In many cases, this is because of issues with insurance and public healthcare coverage for genetic testing rather than lack of awareness on the part of patients or their physicians. Patients need education around these problems, and as a result, the window of time for consent and pretest counseling often is so short that maternal and paternal carrier screening have to be done simultaneously.
Documenting the information provided during pretest counseling is important. Key concepts include:8
• All genetic screening and testing is voluntary and confidential.
• Accurate fetal risk assessment requires screening of the biologic father.
• Testing positive for at least one condition is common.
• Patients may learn about health risks important to them and to family members.
• Conditions screened for range in spectrum of severity-some are mild and others are quite severe.
• A negative test result does not eliminate the possibility of being a carrier (residual risk)
• Repeat screening in subsequent pregnancies is not necessary.
• A plan to provide accurate information is needed in cases of a positive test result.
The follow-up required after implementing carrier screening depends on several factors: The nature of the condition screened (recessive, semi-dominant or X-linked), test result (positive or negative), spectrum of the condition, and literacy level of the patient. Follow-up respects the patients’ need to react, respond, and adjust to their results. Having a plan consistent with the provider’s office infrastructure is important (eg, is a geneticist on site, who calls results to patients). These tips may be helpful.
Designate 2 people in the office to assist in tracking genetic screening reports. This allows uninterrupted review of results when employees are on vacation.
• Be sure to train staff on the voluntary nature and confidentiality of genetic testing.
• To ensure privacy, request that ECS reports for each partner are distinctly separate.
Establish a tracking system for reports with different names.
When one member of a couple tests positive, sequencing the partner’s gene is not recommended. Instead the partner should also be screened.8 Establish a relationship with a professional trained in genetics and genetic counseling. Other family members who might benefit from test results should be considered. Be prepared to discuss specific conditions, spectrum of conditions, treatment options, and available support groups. Rely on legitimate resources (not blogs).
• Fetal testing can be performed using amniocentesis or chorionic villus sampling.
• Laboratories that perform ECS generally do not handle fetal specimens.
• Identify a resource lab for testing fetal specimens.
• Recognize that families will individualize their use of genetic screening and testing information.
Professional organizations have not stated whether all patients should be made aware of ECS. However, this method of screening is increasingly being covered by private insurance as well as state and federal healthcare programs. ECS can be offered to patients without fear as long as basic principles used in everyday ob/gyn practice are adopted to allay anxiety. As the entirety of medicine adopts molecular-based diagnostics and therapeutics, clinicians have a duty to stay current on the implementation of these into clinical practice.
Genetics is not an arcane discipline practiced only by certified geneticists. Genetics should be viewed in the same way as ultrasound, which is now a part of most medical disciplines. Proactive learning can fend off efforts to mandate certification as these technologies become a routine part of state-of-the-art obstetric care.
1. Stratton MR, Campbell PJ, Futreal PA. The cancer genome. Nature. 2009;458:719-724.
2. ACOG Committee Opinion. Number 101: Current Status of cystic fibrosis carrier screening. The American College of Obstetricians and Gynecologists 1991;101.
3. ACOG Committee Opinion. Number 325: Update on carrier screening for cystic fibrosis. Obstet Gynecol. 2005;106:1465-1468.
4. ACOG Committee Opinion. Number 486: Update on Carrier Screening for cystic fibrosis. The American College of Obstetricians and Gynecologists 2011;486:1-4.
5. Prior TW. Carrier screening for spinal muscular atrophy. Genet Med. 2008;10:840-842.
6. Lazarin GA, Hawthorne F, Collins NS, Platt EA, Evans EA, Haque IS. Systematic Classification of Disease Severity for Evaluation of Expanded Carrier Screening Panels. PLoS One 2014;9:e114391.
7. ACOG Practice Bulletin No. 78: Hemoglobinopathies in pregnancy. Obstet Gynecol 2007;109:229-237.
8. Edwards JG, Feldman G, Goldberg J et al. Expanded carrier screening in reproductive medicine-points to consider: a joint statement of the American College of Medical Genetics and Genomics, American College of Obstetricians and Gynecologists, National Society of Genetic Counselors, Perinatal Quality Foundation, and Society for Maternal-Fetal Medicine. Obstet Gynecol 2015;125:653-662.