International Journal of Gynecology and Obstetrics 124 (2014) 181–184
Contents lists available at ScienceDirect
International Journal of Gynecology and Obstetrics journal homepage: www.elsevier.com/locate/ijgo
ETHICAL AND LEGAL ISSUES IN REPRODUCTIVE HEALTH
Ethical and legal aspects of noninvasive prenatal genetic diagnosis Bernard M. Dickens ⁎ Faculty of Law, Faculty of Medicine, Joint Centre for Bioethics, University of Toronto, Toronto, Canada
a r t i c l e
i n f o
Keywords: Cell-free fetal DNA testing Ethics of fetal testing Fetal genetic diagnosis Genetic counseling Noninvasive fetal testing Termination of pregnancy Whole-genome sequencing
a b s t r a c t The new technology that will allow genetic testing of a fetus within the first trimester of pregnancy by isolating cell-free fetal DNA (cffDNA) in the mother’s blood raises a range of ethical and legal issues. Considered noninvasive, this test is safe and reliable, and may avoid alternative genetic testing by amniocentesis or chorionic villus sampling, which risks causing spontaneous abortion. Ethical and legal issues of cffDNA testing will become more acute if testing expands to fetal whole-genome sequencing. Critical issues include the state of the science or diagnostic art; the appropriateness of offering the test; the implications of denying the test when it is available and appropriate; disclosure and counseling following test results; and management of patients’ choices on acquiring test results. A challenge will be providing patients with appropriate counseling based on up-to-date genetic knowledge, and accommodating informed patients’ legal choices. © 2013 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction A means to detect a range of fetal genetic anomalies by testing maternal blood has been developed, described as cell-free fetal DNA (cffDNA) testing [1]. This is considered a minimum risk, noninvasive test, in contrast to fetal genetic and related testing by amniocentesis and chorionic villus sampling (CVS). However, the test raises a spectrum of ethical and legal issues. These include the state of the science or diagnostic art; the appropriateness of offering the test; the implications of denying the test when it is available and appropriate; disclosure and counseling following test results; and management of patients’ choices on acquiring test results. Resort to maternal serum to screen for fetal characteristics is not new. Alpha-fetoprotein testing of maternal blood has been used as a screening technique for over 30 years [2] to determine whether the more invasive amniocentesis is indicated. Amniocentesis may detect fetal neural tube defects such as spina bifida and anencephaly, in addition to some single-gene chromosomal abnormalities. The new test offers and promises to detect a wide range of genetic fetal anomalies and is considered to achieve high levels of sensitivity for trisomies 21, 18, and 13 in high-risk populations, with a 98% probability of correct diagnosis and a specificity above 99.5%, meaning that the proportion of non-affected instances—the true negative rate—would be very reliably determined [1]. The new noninvasive means of fetal genetic diagnosis may appeal to practitioners and expectant parents. Unlike amniocentesis and CVS, it presents no risk of causing spontaneous abortion and can be undertaken earlier in pregnancy: at 9 weeks of gestation, compared with 16 weeks ⁎ Faculty of Law, University of Toronto, 84 Queen’s Park, Toronto, Ontario M5S 2C5, Canada. Tel.: +1 416 978 4849; fax: +1 416 978 7899. E-mail address: [email protected].
plus a further 2 weeks for culture results in amniocentesis and 10–13 weeks for CVS, which is more complicated and carries twice the risk of spontaneous abortion [3] (p. 65). In its December 2012 Committee Opinion on noninvasive prenatal testing for fetal aneuploidy [4], the American College of Obstetricians and Gynecologists (ACOG) cautiously approved cffDNA testing for women at high risk of bearing fetuses affected by trisomies 13, 18, or 21, but as a primary screening test rather than as a replacement for the precision obtained with CVS or amniocentesis. The test is also approved as a follow-up test for women with a positive first- or secondtrimester screening test result. The Committee Opinion notes the limitation of the lack of outcome data for testing with low-risk populations, for which it cannot therefore be recommended. With high-risk populations, pretest counseling regarding current limitations of the test is recommended, with referral for genetic counseling for pregnant women with positive test results. The ACOG Committee Opinion has been welcomed by companies that offer the technology for the test. They anticipate its improvement and sufficient cost reduction for widespread application, not limited to high-risk populations. Similarly, popular news media have indulged their appetite for scientific innovation with an immediate personal impact by publicizing not only the primary medical purposes but also some secondary social effects, with such article headings as “Prenatal testing: earlier and more accurate than ever” followed by the subheading “Parents-to-be can now safely determine their baby’s gender, father, and certain chromosomal abnormalities during the first trimester” [5]. Popular reactions went beyond reassuring women in high-risk populations that they could enjoy the prospect of delivering healthy children. They speculated on a reduction close to elimination of births of children with Down syndrome and related genetic conditions, rendering parents who favor the births of affected fetuses and the born children themselves anomalous and stigmatized in their communities,
0020-7292/$ – see front matter © 2013 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijgo.2013.11.001
182
B.M. Dickens / International Journal of Gynecology and Obstetrics 124 (2014) 181–184
and an increase in sex-selective abortion. Whether the new test offers a preponderance of social benefit over risk or vice versa is placed in the eye of the beholder. 2. State of the diagnostic art The ACOG Genetics Committee Opinion [4] presents a balanced assessment of the current utility of cffDNA testing; its values and limitations; how it may be appropriately offered and undertaken; and how it fits within existing resources for fetal genetic diagnosis. A surge of enthusiasm to offer and provide improved diagnosis by cffDNA testing is understandable, driven by clinicians’ intentions to serve their patients better and by patients’ hopes that early testing will provide the reassurance of a reliable negative result or time to consider difficult options in the face of diagnosis of fetal anomaly. Promotion of cffDNA testing is also driven by aggressive advertising by test technology suppliers keen to develop a market for their product and to maximize their market share. A cautious note has been sounded, however, against premature or excessive resort to the new technology. It has been observed, for instance, that “the diffusion of [cffDNA] testing into routine prenatal care may be occurring too quickly. Professional societies do not recommend these tests for normal-risk pregnancies because their clinical utility in the general population is not well established” [1] (p. 499). Concern has been raised regarding tests with high-risk populations because sensitivity and specificity tests were conducted on “collections of archived samples with known karyotypes that intentionally included a large proportion of specimens from women with known aneuploid fetuses” [1] (p. 499). This biasing of samples may be justified on scientific grounds related to the effectiveness of the test but it leaves questions about the generalizability of the outcome conclusions, even among high-risk populations for whom the test is considered most appropriate. The limited evidence about the performance of testing in the general population and in twin pregnancies [6], and about the positive predictive value of the tests [7] underscores concerns about overselling and overuse of cffDNA testing. Its cost may be a restraint on present use because costs of the 4 versions of the test currently available in the USA range from $795 to more than $2000 [1] (p. 501) but, should costs fall significantly, routine resort to cffDNA testing will become a concern that reproductive health professionals and funders of reproductive healthcare services, both governmental and private, will have to address. Several biotechnological procedures such as in vitro fertilization and stem cell treatments have been criticized for having moved prematurely from research into therapy without the necessary intermediate stage of disinterested clinical evaluation—a move propelled by the hope that they will satisfy patients’ needs and demands, and by commercial agents’ incentives to create markets and reap returns on investments. 3. Appropriateness of offering testing Ethical concerns raised by prospects of cffDNA testing include, but transcend, clinical diagnosis of trisomies in fetuses at high risk of abnormality. They arise more gravely because easy, safe access to fetal DNA might make fetuses amenable to techniques that test for a much broader range of genetic abnormalities such as susceptibility to breast cancer and late-onset disorders, and even extend to whole-genome or whole-exome sequencing. What was once a cavernous divide between the outer reaches of imaginative science fiction and the reality of the limited capacity of prevailing biotechnology is becoming progressively narrowed, making it foreseeable to achieve complete gene sequencing of an early fetus in utero by resort to cffDNA testing. Initial concerns are more mundane and immediate. If cffDNA testing were to become relatively inexpensive and routine, women offered the test as part of their prenatal care might give consent as to any other request for blood sampling. Anecdotal hazards of routine blood drawing extend to thrombophlebitis, pulmonary embolism, and death but
complications such as infection are rare, and associated mild bruising is usually transient. The law of informed consent to medical procedures requires disclosure of significant hazards, including low risks of severe consequences, but blood sampling is often taken as the legal paradigm of a minimum-risk procedure. Accordingly, little beyond brief discomfort and mild bruising is usually disclosed, and patients give consent to the procedure without much or any counseling. However, as the ACOG Committee on Genetics observes, “[t]o offer a cell free fetal DNA test, pretest counseling regarding these [described] limitations is recommended. The use of a cell free fetal DNA test should be an active, informed choice and not part of routine prenatal laboratory testing” [4]. Concern has been expressed that offering cffDNA testing on a wide scale would undermine informed consent [8] and risk trivializing a procedure that might compel exceptionally difficult decisions to be made, with lifelong consequences. A related concern is that, because the “chances of an affected pregnancy ending in miscarriage decrease with gestational age, early testing will more often burden women with ‘unnecessary’ decision making concerning pregnancies that may spontaneously miscarry” [8] (p. 273). Noninvasive prenatal testing of fetal characteristics has been advanced to identify significant chromosomal abnormalities, but as genetic diagnosis becomes refined the potential emerges to identify minor abnormalities, genetic features of unknown significance, and normal features of sex and inheritance—particularly paternity—that patients may disfavor for the children they might deliver. Disclosure of fetal inheritance of genes predisposing, for instance, to breast cancer may provide opportunities for parents to seek to control their future children’s lives to an extraordinary degree and to terminate pregnancies on irrational and alarmist grounds. In pediatric genetics, it is often considered inappropriate to test minors for their liability to experience late-onset disorders, notably Huntington’s disease [9] (p.232–3), lest disclosure may cause parents to limit their children’s opportunities for enjoyment of their lives. However, disclosure may allow parents the advantage of time to make suitable plans. This raises concerns about whether there should be guidelines or limits for what tests and disclosures are appropriate regarding cffDNA testing. A key ethical concern with prenatal genetic testing of more minor inheritance is that disclosures may precondition parents’ expectations of their children’s capacities and personalities, and trigger “genetic determinism:” that is, the belief that individuals’ genes exclusively or primarily determine their capacities and characteristics. This belief revives the historical, unresolved debate about the interaction of nature (meaning genes) and nurture (meaning upbringing and social environment) in shaping an individual’s character and personality [9] (pp. 281–299). A related concern is that, as children mature, they may be unduly influenced by what they perceive, and/or what their parents indicate, to be their genetic destiny. They may accordingly attempt to pursue, or to resist, their genetic predestination, rather than any independent choices or chance opportunities. They may be denied the freedom to flourish as their instincts and circumstances allow, and to express an independent personality. They may also be compelled to forfeit an alleged right that has been recognized particularly in the context of genetic diagnosis: the right not to know [8] (p. 275). Ethics and law applicable to healthcare have come to emphasize informed consent, requiring health service providers’ disclosures of information material to patients’ choices. The obligation, however, is not to impose information but to offer it. The goal is to serve individuals’ choices not only of treatment options but also of receipt of information. Patients may accept the offer and make an informed choice, or forgo the information and either accept the recommended treatment on trust or decline the treatment. Individuals may choose not to take an opportunity to learn their medical prognoses based on genetic or other tests. Parents who obtain cffDNA testing, or other genetic testing, of their fetuses may deny the children they rear the right of choice to be free of this knowledge.
B.M. Dickens / International Journal of Gynecology and Obstetrics 124 (2014) 181–184
4. Denial of accessible testing Costs may keep cffDNA tests unavailable for some time, even in high-resource settings, but it is anticipated that costs will fall and testing will become more widely accessible. A body of experience already exists, in both ethical analysis and particularly legal analysis and jurisprudence, of implications of health service providers failing to inform pregnant women of their choice of resort to accessible prenatal testing. Patients’ informed consent to options serves patients’ autonomy but differs from their autonomy in that consent refers to patients’ rights to accept or reject offered or recommended treatment, whereas autonomy refers to patients’ rights to demand treatment that is not offered. Omission to offer a treatment may occur, for instance, because it is unavailable or considered futile or contraindicated. Patients’ autonomy in this sense is rarely available, although a court has, for example, required a Roman Catholic hospital to provide emergency contraception to rape victims over its officers’ conscientious objection [10]. Informed consent requires disclosure, however, of available treatment options, and patients with high-risk pregnancies should usually be informed of how the risk may be identified, quantified if possible, and managed. Ethical concerns arise if providers object to prenatal testing because of their conscientious objection to an outcome such as abortion or patients’ voluntary sterilization—to which test results may lead. They are fully entitled to refuse to participate in procedures to which they object but disclosure is required of lawful procedures available to their patients to which providers may object. Disclosure of such indicated options is part of providers’ legal duty of care and does not constitute their participation in any eventual procedures. The FIGO Ethical Guidelines on Conscientious Objection require, in guideline 2, that [11]: Practitioners provide [their] patients with timely access to medical services, including giving information about the medically indicated options of procedures for their care and of any such procedures in which their practitioners object to participate on grounds of conscience. Guideline 6 provides that [11]: Patients are entitled to be referred in good faith, for procedures… that their practitioners object to undertaking, to practitioners who do not object. Referral for services does not constitute participation in any procedures agreed upon between patients and the practitioners to whom they are referred. Practitioners’ failures, usually due to oversight rather than conscientious refusal, to inform their pregnant patients—particularly those at high risk of delivering severely impaired children—of medically available options of which they would have taken advantage to terminate their pregnancies have resulted in litigation in several countries. Patients’ legal proceedings are unlikely to succeed where the abortions the patients complain they were denied would have been unlawful, of course, but wrongful denial of information about lawful procedures has given rise to so-called “wrongful birth” actions. In the early years, courts unfamiliar with these claims tended to reject them on the ground that every child’s birth is a blessing, but more recently they have become willing to “bust the blessing balloon” [12] and find legal liability. This is usually for negligence causing parents injury where the parents establish that, but for the failure of disclosure, they would have aborted the pregnancies. Courts have disagreed with each other about the basis on which compensation may be awarded [3] (pp. 79–99). Birth of an apparently healthy or normal child may be considered to justify only token or nominal compensation, but where caring for an unhealthy or severely disabled child is likely to be financially and practically burdensome to parents compensation may cover the expenses of accommodating the child’s particular needs and a sum to represent the parents’ inconvenience, including lost employment opportunities.
183
Where it is foreseeable that a physically and/or mentally disabled child would be taken into care by a governmental agency, parents’ compensation for costs of care may be reduced. Awards often have to be speculative about children’s futures and parents’ ability to care for them where courts award only lump-sum payments, but if payments may be awarded or agreed, such as with medical insurance companies, on an annuity or periodic basis they may be adjusted to the children’s evolving circumstances. In contrast to wrongful birth claims are claims, sometimes called “wrongful life” claims, that are occasionally brought by or on behalf of children born with severe impairments. These claims allege that the children should have been aborted in utero. Claims that children would have been born in normal health but for providers’ medical negligence may succeed because courts can quantify the difference between enjoyment of a normal life and enjoyment of life diminished by impairment. However, when the alternative to children’s impaired lives would not have been unimpaired lives but their abortion, once described as “the utter void of nonexistence” [13], courts may find compensation for such children impossible to calculate, although some courts have been able to overcome this difficulty [14]. 5. Disclosure and counseling on test results Medical information obtained from patients in clinical settings, however contained or recorded, should in principle be available to the patients and under their control. Patients’ rights to their medical information are not necessarily to the raw data, which they are liable to find incomprehensible, but to an interpretation of the data relevant to their circumstances and questions. If cffDNA testing develops to allow comprehensive fetal genetic diagnosis, the quality of genetic counseling available to patients will be significant, and this will be based on the information they receive. Providers’ professional judgment will be required to cater disclosure of genetic data from fetuses to patients’ needs to know, based on their objectively perceived circumstances and their subjectively expressed wishes, in order to spare patients the confusion and dysfunction of information overload. More may be learned of fetal genetic conditions, traits, and susceptibilities than patients can keep in rational perspective. Information available from prenatal genetic testing may have to be filtered, for instance by primary reference to trisomies and conditions that comparably impair health as understood by WHO (i.e. affecting physical, mental, or social wellbeing) and only secondarily to patients’ interests in information less related to fetal health. Nevertheless, patients’ requests for information must be answered honestly. The ethical duty of disclosure applies even when providers believe that patients will apply the information for purposes to which the providers conscientiously object, as has been seen above [11]. Similarly, counseling on options that arise following results of prenatal testing should include all lawful accessible medical options. Counselors who feel unable to disclose an abortion, sterilization, or comparable option are ethically required to refer patients to other counselors who are willing to present these options. Access to competent genetic counseling will be difficult in many cases, not just in resource-poor settings. It may be supposed that centers capable of conducting cffDNA testing will be equipped with counselors for pre-test and post-test services. If and where testing becomes routine, however, counseling by specially trained genetic counselors may become the exception rather than the rule, and responsibility for counseling on test potentials and results may be shared by general providers of reproductive healthcare. This raises the question of the standard of care or skill to which they will be legally held. The law does not require perfection, but competence usually according to standards of the practitioner’s profession. That is, the law does not require that correct decisions be made but that decisions be made correctly. Decisions must be made and judgment exercised by reference to what would be considered relevant and done
184
B.M. Dickens / International Journal of Gynecology and Obstetrics 124 (2014) 181–184
by a reasonably competent provider in a particular provider’s professional field of practice, acting with ordinary care [15]. Professional standards are established by evidence of professional education and training, and of general practice within a particular provider’s specialty or profession. The standard of competence and skill the profession sets for its practitioners will be strongly influential. The legally required standard of care or skill remains a matter of law, however, and is not just what a profession finds satisfactory. This is not peculiar to medicine, but applies to all professions, including engineering, law, and pharmacy. Very exceptionally, courts may find that a standard of performance accepted in a particular profession falls below the level of protection to which the public is entitled. Professional standards evolve over time. If cffDNA testing becomes routine, practitioners in reproductive health may be required to know what genetic information relevant tests may disclose and the implications of test results for the outcome of pregnancy; for children’s health; for family strategies and means to care for disabled children; and for what such children’s futures may offer. Practitioners may be required to counsel patients on the basis of such knowledge or to refer their patients to alternative sources of relevant information, including literature or pamphlets in language at a level of patients’ comprehension. Accordingly, patients’ routine access to cffDNA testing may add genetic understanding and counseling to required teaching and training curricula within the reproductive healthcare field and be a criterion of approval for accreditation of training programs and for professional licensure. It has been observed in the USA that “[t]he introduction of noninvasive prenatal diagnosis would require the ill-prepared medical system to change how patients are counseled and how cases are managed” [16]. 6. Patients’ choices following test results When test results cannot afford pregnant patients reassurance that their fetuses can be born as healthy children, patients may face difficult and even agonizing choices. An initial difficulty is to determine parameters of a child’s “normal” genetic diagnosis and a normally healthy future. In popular culture, normality may be assessed by superficial criteria such as appearance and ability to function in society. For children, this may mean ability, for instance, to attend a regular school, play safely with other children, and participate in sports. Medical assessment may be more statistical or normative, and consider risk levels of dysfunction, recognizing that—in the course of nature—all individuals bear genetic traits that may condition them to functional limitations or that make them susceptible to impairments. These may or may not materialize, depending on factors such as diet or environment that they may or may not be able to regulate. Newer forms of genetic diagnosis and health prognosis may show that every individual departs to some extent from a statistical or biological genetic norm. A challenge is to have patients understand the contrast between the possibility that a child will suffer effects of its genetic inheritance or predisposition and the probability of such effects, the assessment of risk, the possibility—although apparently low in cffDNA testing—of a falsepositive test result, and uncertainties due to genetic variants of unknown clinical significance. An immediate key decision for patients, where they have the choice of legal abortion, may be whether or not to continue their pregnancies. If they have the option of legal abortion but—as a matter of principle—would not accept this procedure, the test may be of value to them to prepare for any diagnosed conditions
in the children they deliver. Where testing is available, for instance in high-risk pregnancies, it should not be denied to eligible patients on the ground that they will refuse to terminate their pregnancies. Denial would appear as discrimination on grounds of conscience or religion, and characterize the test as abortion opponents have claimed it is: as a “search and destroy” exercise. Practitioners who would not participate in services or counseling that includes an abortion option should not suffer sanction or discrimination but should refer their eligible patients to accessible practitioners who do not object, in accordance with FIGO ethical guidelines [11]. All appropriately trained practitioners should counsel patients who receive cffDNA test results showing a trisomy and/or a comparable genetic condition about its implications and the management of affected children. Counseling should be non-directive and designed to present available options for patients to consider. Where abortion would not be lawful, counselors have to consider whether patients have feasible options of traveling to another, more accommodating, country or jurisdiction. For instance, the European Court of Human Rights has upheld the restrictive abortion law in Ireland on the ground that Irish law allows its nationals to receive lawful abortion services in other countries [17]. In 2012, for instance, 3982 women providing home addresses in the Republic of Ireland had abortions in the UK [18]. Conflict of interest The author has no conflicts of interest. References [1] Morain S, Greene MF, Mello MM. A new era in noninvasive prenatal testing. N Engl J Med 2013;369(6):499–501. [2] Brock DJH, Scrimgeour JB, Steven J, Barron L, Watt M. Maternal plasma alpha feto protein screening for fetal neural tube defects. Br J Obstet Gynaecol 1978;85:575–81. [3] Mason JK. The Troubled Pregnancy: Legal Wrongs and Rights in Reproduction. Cambridge: Cambridge University Press; 2007. [4] American College of Obstetricians and Gynecologists Committee on Genetics, The Society for Maternal-Fetal Medicine. Committee Opinion Number 545, December 2012. Noninvasive Prenatal Testing for Fetal Aneuploidy; 2012. [5] Abrams L. Prenatal testing: earlier and more accurate than ever. The Atlantic; December 11, 2012. [6] Nicolaides KH, Syngelaki A, Ashoor G, Birdir C, Touzet G. Noninvasive prenatal testing for fetal trisomies in a routinely screened first trimester population. Am J Obstet Gynecol 2012;207(5):374 eI-374, e.6. [7] Norton ME, Rose NC, Benn P. Noninvasive prenatal testing for fetal aneuploidy: clinical assessment and a plea for restraint. Obstet Gynecol 2013;121:847–50. [8] de Jong A, Dondorp WJ, de Die-Smulders CEM, Frints SGM, de Wert GMWR. Noninvasive prenatal testing: ethical issues explored. Eur J Hum Genet 2010;18(3):272–7. [9] Kevles DJ, Hood L. The Code of Codes: Scientific and Social Issues in the Human Genome Project. Cambridge, MA: Harvard University Press; 1992. [10] Brownfield v Daniel Freeman Marina Hospital. 256 California Reporter 240 (Court of Appeal, Second District, Division 4, California, 1989); 1989. [11] FIGO Committee for the Study of Ethical Aspects of Human Reproduction and Women’s Health. Ethical Issues in Obstetrics and Gynecology. London: FIGO; 2012 25–7. [12] Ranous JS, Sherrin JJ. Busting the blessing balloon; liability for the birth of an unplanned child. Albany Law Rev 1975;39:221–51. [13] Gleitman v Cosgrove. 227 Atlantic Reporter 2d 689 (New Jersey Supreme Court, 1967); 1967. [14] Procanik v. Cillo. 478 Atlantic Reporter 2d 755 (New Jersey Supreme Court, 1984); 1984. [15] Dickens BM. Medical errors: legal and ethical responses. Int J Gynecol Obstet 2003;81(1):109–14. [16] Benn PA, Chapman AR. Practical and ethical considerations of noninvasive prenatal diagnosis. JAMA 2009;301(20):2154–6. [17] Westeson J. Reproductive health information and abortion services: standards developed by the European Court of Human Rights. Int J Gynecol Obstet 2013;122(2):173–6. [18] UK Government. Abortion Statistics for England and Wales, 2012. Department of Health; 2013 52 . Table 12a.
Contents lists available at ScienceDirect
International Journal of Gynecology and Obstetrics journal homepage: www.elsevier.com/locate/ijgo
ETHICAL AND LEGAL ISSUES IN REPRODUCTIVE HEALTH
Ethical and legal aspects of noninvasive prenatal genetic diagnosis Bernard M. Dickens ⁎ Faculty of Law, Faculty of Medicine, Joint Centre for Bioethics, University of Toronto, Toronto, Canada
a r t i c l e
i n f o
Keywords: Cell-free fetal DNA testing Ethics of fetal testing Fetal genetic diagnosis Genetic counseling Noninvasive fetal testing Termination of pregnancy Whole-genome sequencing
a b s t r a c t The new technology that will allow genetic testing of a fetus within the first trimester of pregnancy by isolating cell-free fetal DNA (cffDNA) in the mother’s blood raises a range of ethical and legal issues. Considered noninvasive, this test is safe and reliable, and may avoid alternative genetic testing by amniocentesis or chorionic villus sampling, which risks causing spontaneous abortion. Ethical and legal issues of cffDNA testing will become more acute if testing expands to fetal whole-genome sequencing. Critical issues include the state of the science or diagnostic art; the appropriateness of offering the test; the implications of denying the test when it is available and appropriate; disclosure and counseling following test results; and management of patients’ choices on acquiring test results. A challenge will be providing patients with appropriate counseling based on up-to-date genetic knowledge, and accommodating informed patients’ legal choices. © 2013 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction A means to detect a range of fetal genetic anomalies by testing maternal blood has been developed, described as cell-free fetal DNA (cffDNA) testing [1]. This is considered a minimum risk, noninvasive test, in contrast to fetal genetic and related testing by amniocentesis and chorionic villus sampling (CVS). However, the test raises a spectrum of ethical and legal issues. These include the state of the science or diagnostic art; the appropriateness of offering the test; the implications of denying the test when it is available and appropriate; disclosure and counseling following test results; and management of patients’ choices on acquiring test results. Resort to maternal serum to screen for fetal characteristics is not new. Alpha-fetoprotein testing of maternal blood has been used as a screening technique for over 30 years [2] to determine whether the more invasive amniocentesis is indicated. Amniocentesis may detect fetal neural tube defects such as spina bifida and anencephaly, in addition to some single-gene chromosomal abnormalities. The new test offers and promises to detect a wide range of genetic fetal anomalies and is considered to achieve high levels of sensitivity for trisomies 21, 18, and 13 in high-risk populations, with a 98% probability of correct diagnosis and a specificity above 99.5%, meaning that the proportion of non-affected instances—the true negative rate—would be very reliably determined [1]. The new noninvasive means of fetal genetic diagnosis may appeal to practitioners and expectant parents. Unlike amniocentesis and CVS, it presents no risk of causing spontaneous abortion and can be undertaken earlier in pregnancy: at 9 weeks of gestation, compared with 16 weeks ⁎ Faculty of Law, University of Toronto, 84 Queen’s Park, Toronto, Ontario M5S 2C5, Canada. Tel.: +1 416 978 4849; fax: +1 416 978 7899. E-mail address: [email protected].
plus a further 2 weeks for culture results in amniocentesis and 10–13 weeks for CVS, which is more complicated and carries twice the risk of spontaneous abortion [3] (p. 65). In its December 2012 Committee Opinion on noninvasive prenatal testing for fetal aneuploidy [4], the American College of Obstetricians and Gynecologists (ACOG) cautiously approved cffDNA testing for women at high risk of bearing fetuses affected by trisomies 13, 18, or 21, but as a primary screening test rather than as a replacement for the precision obtained with CVS or amniocentesis. The test is also approved as a follow-up test for women with a positive first- or secondtrimester screening test result. The Committee Opinion notes the limitation of the lack of outcome data for testing with low-risk populations, for which it cannot therefore be recommended. With high-risk populations, pretest counseling regarding current limitations of the test is recommended, with referral for genetic counseling for pregnant women with positive test results. The ACOG Committee Opinion has been welcomed by companies that offer the technology for the test. They anticipate its improvement and sufficient cost reduction for widespread application, not limited to high-risk populations. Similarly, popular news media have indulged their appetite for scientific innovation with an immediate personal impact by publicizing not only the primary medical purposes but also some secondary social effects, with such article headings as “Prenatal testing: earlier and more accurate than ever” followed by the subheading “Parents-to-be can now safely determine their baby’s gender, father, and certain chromosomal abnormalities during the first trimester” [5]. Popular reactions went beyond reassuring women in high-risk populations that they could enjoy the prospect of delivering healthy children. They speculated on a reduction close to elimination of births of children with Down syndrome and related genetic conditions, rendering parents who favor the births of affected fetuses and the born children themselves anomalous and stigmatized in their communities,
0020-7292/$ – see front matter © 2013 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijgo.2013.11.001
182
B.M. Dickens / International Journal of Gynecology and Obstetrics 124 (2014) 181–184
and an increase in sex-selective abortion. Whether the new test offers a preponderance of social benefit over risk or vice versa is placed in the eye of the beholder. 2. State of the diagnostic art The ACOG Genetics Committee Opinion [4] presents a balanced assessment of the current utility of cffDNA testing; its values and limitations; how it may be appropriately offered and undertaken; and how it fits within existing resources for fetal genetic diagnosis. A surge of enthusiasm to offer and provide improved diagnosis by cffDNA testing is understandable, driven by clinicians’ intentions to serve their patients better and by patients’ hopes that early testing will provide the reassurance of a reliable negative result or time to consider difficult options in the face of diagnosis of fetal anomaly. Promotion of cffDNA testing is also driven by aggressive advertising by test technology suppliers keen to develop a market for their product and to maximize their market share. A cautious note has been sounded, however, against premature or excessive resort to the new technology. It has been observed, for instance, that “the diffusion of [cffDNA] testing into routine prenatal care may be occurring too quickly. Professional societies do not recommend these tests for normal-risk pregnancies because their clinical utility in the general population is not well established” [1] (p. 499). Concern has been raised regarding tests with high-risk populations because sensitivity and specificity tests were conducted on “collections of archived samples with known karyotypes that intentionally included a large proportion of specimens from women with known aneuploid fetuses” [1] (p. 499). This biasing of samples may be justified on scientific grounds related to the effectiveness of the test but it leaves questions about the generalizability of the outcome conclusions, even among high-risk populations for whom the test is considered most appropriate. The limited evidence about the performance of testing in the general population and in twin pregnancies [6], and about the positive predictive value of the tests [7] underscores concerns about overselling and overuse of cffDNA testing. Its cost may be a restraint on present use because costs of the 4 versions of the test currently available in the USA range from $795 to more than $2000 [1] (p. 501) but, should costs fall significantly, routine resort to cffDNA testing will become a concern that reproductive health professionals and funders of reproductive healthcare services, both governmental and private, will have to address. Several biotechnological procedures such as in vitro fertilization and stem cell treatments have been criticized for having moved prematurely from research into therapy without the necessary intermediate stage of disinterested clinical evaluation—a move propelled by the hope that they will satisfy patients’ needs and demands, and by commercial agents’ incentives to create markets and reap returns on investments. 3. Appropriateness of offering testing Ethical concerns raised by prospects of cffDNA testing include, but transcend, clinical diagnosis of trisomies in fetuses at high risk of abnormality. They arise more gravely because easy, safe access to fetal DNA might make fetuses amenable to techniques that test for a much broader range of genetic abnormalities such as susceptibility to breast cancer and late-onset disorders, and even extend to whole-genome or whole-exome sequencing. What was once a cavernous divide between the outer reaches of imaginative science fiction and the reality of the limited capacity of prevailing biotechnology is becoming progressively narrowed, making it foreseeable to achieve complete gene sequencing of an early fetus in utero by resort to cffDNA testing. Initial concerns are more mundane and immediate. If cffDNA testing were to become relatively inexpensive and routine, women offered the test as part of their prenatal care might give consent as to any other request for blood sampling. Anecdotal hazards of routine blood drawing extend to thrombophlebitis, pulmonary embolism, and death but
complications such as infection are rare, and associated mild bruising is usually transient. The law of informed consent to medical procedures requires disclosure of significant hazards, including low risks of severe consequences, but blood sampling is often taken as the legal paradigm of a minimum-risk procedure. Accordingly, little beyond brief discomfort and mild bruising is usually disclosed, and patients give consent to the procedure without much or any counseling. However, as the ACOG Committee on Genetics observes, “[t]o offer a cell free fetal DNA test, pretest counseling regarding these [described] limitations is recommended. The use of a cell free fetal DNA test should be an active, informed choice and not part of routine prenatal laboratory testing” [4]. Concern has been expressed that offering cffDNA testing on a wide scale would undermine informed consent [8] and risk trivializing a procedure that might compel exceptionally difficult decisions to be made, with lifelong consequences. A related concern is that, because the “chances of an affected pregnancy ending in miscarriage decrease with gestational age, early testing will more often burden women with ‘unnecessary’ decision making concerning pregnancies that may spontaneously miscarry” [8] (p. 273). Noninvasive prenatal testing of fetal characteristics has been advanced to identify significant chromosomal abnormalities, but as genetic diagnosis becomes refined the potential emerges to identify minor abnormalities, genetic features of unknown significance, and normal features of sex and inheritance—particularly paternity—that patients may disfavor for the children they might deliver. Disclosure of fetal inheritance of genes predisposing, for instance, to breast cancer may provide opportunities for parents to seek to control their future children’s lives to an extraordinary degree and to terminate pregnancies on irrational and alarmist grounds. In pediatric genetics, it is often considered inappropriate to test minors for their liability to experience late-onset disorders, notably Huntington’s disease [9] (p.232–3), lest disclosure may cause parents to limit their children’s opportunities for enjoyment of their lives. However, disclosure may allow parents the advantage of time to make suitable plans. This raises concerns about whether there should be guidelines or limits for what tests and disclosures are appropriate regarding cffDNA testing. A key ethical concern with prenatal genetic testing of more minor inheritance is that disclosures may precondition parents’ expectations of their children’s capacities and personalities, and trigger “genetic determinism:” that is, the belief that individuals’ genes exclusively or primarily determine their capacities and characteristics. This belief revives the historical, unresolved debate about the interaction of nature (meaning genes) and nurture (meaning upbringing and social environment) in shaping an individual’s character and personality [9] (pp. 281–299). A related concern is that, as children mature, they may be unduly influenced by what they perceive, and/or what their parents indicate, to be their genetic destiny. They may accordingly attempt to pursue, or to resist, their genetic predestination, rather than any independent choices or chance opportunities. They may be denied the freedom to flourish as their instincts and circumstances allow, and to express an independent personality. They may also be compelled to forfeit an alleged right that has been recognized particularly in the context of genetic diagnosis: the right not to know [8] (p. 275). Ethics and law applicable to healthcare have come to emphasize informed consent, requiring health service providers’ disclosures of information material to patients’ choices. The obligation, however, is not to impose information but to offer it. The goal is to serve individuals’ choices not only of treatment options but also of receipt of information. Patients may accept the offer and make an informed choice, or forgo the information and either accept the recommended treatment on trust or decline the treatment. Individuals may choose not to take an opportunity to learn their medical prognoses based on genetic or other tests. Parents who obtain cffDNA testing, or other genetic testing, of their fetuses may deny the children they rear the right of choice to be free of this knowledge.
B.M. Dickens / International Journal of Gynecology and Obstetrics 124 (2014) 181–184
4. Denial of accessible testing Costs may keep cffDNA tests unavailable for some time, even in high-resource settings, but it is anticipated that costs will fall and testing will become more widely accessible. A body of experience already exists, in both ethical analysis and particularly legal analysis and jurisprudence, of implications of health service providers failing to inform pregnant women of their choice of resort to accessible prenatal testing. Patients’ informed consent to options serves patients’ autonomy but differs from their autonomy in that consent refers to patients’ rights to accept or reject offered or recommended treatment, whereas autonomy refers to patients’ rights to demand treatment that is not offered. Omission to offer a treatment may occur, for instance, because it is unavailable or considered futile or contraindicated. Patients’ autonomy in this sense is rarely available, although a court has, for example, required a Roman Catholic hospital to provide emergency contraception to rape victims over its officers’ conscientious objection [10]. Informed consent requires disclosure, however, of available treatment options, and patients with high-risk pregnancies should usually be informed of how the risk may be identified, quantified if possible, and managed. Ethical concerns arise if providers object to prenatal testing because of their conscientious objection to an outcome such as abortion or patients’ voluntary sterilization—to which test results may lead. They are fully entitled to refuse to participate in procedures to which they object but disclosure is required of lawful procedures available to their patients to which providers may object. Disclosure of such indicated options is part of providers’ legal duty of care and does not constitute their participation in any eventual procedures. The FIGO Ethical Guidelines on Conscientious Objection require, in guideline 2, that [11]: Practitioners provide [their] patients with timely access to medical services, including giving information about the medically indicated options of procedures for their care and of any such procedures in which their practitioners object to participate on grounds of conscience. Guideline 6 provides that [11]: Patients are entitled to be referred in good faith, for procedures… that their practitioners object to undertaking, to practitioners who do not object. Referral for services does not constitute participation in any procedures agreed upon between patients and the practitioners to whom they are referred. Practitioners’ failures, usually due to oversight rather than conscientious refusal, to inform their pregnant patients—particularly those at high risk of delivering severely impaired children—of medically available options of which they would have taken advantage to terminate their pregnancies have resulted in litigation in several countries. Patients’ legal proceedings are unlikely to succeed where the abortions the patients complain they were denied would have been unlawful, of course, but wrongful denial of information about lawful procedures has given rise to so-called “wrongful birth” actions. In the early years, courts unfamiliar with these claims tended to reject them on the ground that every child’s birth is a blessing, but more recently they have become willing to “bust the blessing balloon” [12] and find legal liability. This is usually for negligence causing parents injury where the parents establish that, but for the failure of disclosure, they would have aborted the pregnancies. Courts have disagreed with each other about the basis on which compensation may be awarded [3] (pp. 79–99). Birth of an apparently healthy or normal child may be considered to justify only token or nominal compensation, but where caring for an unhealthy or severely disabled child is likely to be financially and practically burdensome to parents compensation may cover the expenses of accommodating the child’s particular needs and a sum to represent the parents’ inconvenience, including lost employment opportunities.
183
Where it is foreseeable that a physically and/or mentally disabled child would be taken into care by a governmental agency, parents’ compensation for costs of care may be reduced. Awards often have to be speculative about children’s futures and parents’ ability to care for them where courts award only lump-sum payments, but if payments may be awarded or agreed, such as with medical insurance companies, on an annuity or periodic basis they may be adjusted to the children’s evolving circumstances. In contrast to wrongful birth claims are claims, sometimes called “wrongful life” claims, that are occasionally brought by or on behalf of children born with severe impairments. These claims allege that the children should have been aborted in utero. Claims that children would have been born in normal health but for providers’ medical negligence may succeed because courts can quantify the difference between enjoyment of a normal life and enjoyment of life diminished by impairment. However, when the alternative to children’s impaired lives would not have been unimpaired lives but their abortion, once described as “the utter void of nonexistence” [13], courts may find compensation for such children impossible to calculate, although some courts have been able to overcome this difficulty [14]. 5. Disclosure and counseling on test results Medical information obtained from patients in clinical settings, however contained or recorded, should in principle be available to the patients and under their control. Patients’ rights to their medical information are not necessarily to the raw data, which they are liable to find incomprehensible, but to an interpretation of the data relevant to their circumstances and questions. If cffDNA testing develops to allow comprehensive fetal genetic diagnosis, the quality of genetic counseling available to patients will be significant, and this will be based on the information they receive. Providers’ professional judgment will be required to cater disclosure of genetic data from fetuses to patients’ needs to know, based on their objectively perceived circumstances and their subjectively expressed wishes, in order to spare patients the confusion and dysfunction of information overload. More may be learned of fetal genetic conditions, traits, and susceptibilities than patients can keep in rational perspective. Information available from prenatal genetic testing may have to be filtered, for instance by primary reference to trisomies and conditions that comparably impair health as understood by WHO (i.e. affecting physical, mental, or social wellbeing) and only secondarily to patients’ interests in information less related to fetal health. Nevertheless, patients’ requests for information must be answered honestly. The ethical duty of disclosure applies even when providers believe that patients will apply the information for purposes to which the providers conscientiously object, as has been seen above [11]. Similarly, counseling on options that arise following results of prenatal testing should include all lawful accessible medical options. Counselors who feel unable to disclose an abortion, sterilization, or comparable option are ethically required to refer patients to other counselors who are willing to present these options. Access to competent genetic counseling will be difficult in many cases, not just in resource-poor settings. It may be supposed that centers capable of conducting cffDNA testing will be equipped with counselors for pre-test and post-test services. If and where testing becomes routine, however, counseling by specially trained genetic counselors may become the exception rather than the rule, and responsibility for counseling on test potentials and results may be shared by general providers of reproductive healthcare. This raises the question of the standard of care or skill to which they will be legally held. The law does not require perfection, but competence usually according to standards of the practitioner’s profession. That is, the law does not require that correct decisions be made but that decisions be made correctly. Decisions must be made and judgment exercised by reference to what would be considered relevant and done
184
B.M. Dickens / International Journal of Gynecology and Obstetrics 124 (2014) 181–184
by a reasonably competent provider in a particular provider’s professional field of practice, acting with ordinary care [15]. Professional standards are established by evidence of professional education and training, and of general practice within a particular provider’s specialty or profession. The standard of competence and skill the profession sets for its practitioners will be strongly influential. The legally required standard of care or skill remains a matter of law, however, and is not just what a profession finds satisfactory. This is not peculiar to medicine, but applies to all professions, including engineering, law, and pharmacy. Very exceptionally, courts may find that a standard of performance accepted in a particular profession falls below the level of protection to which the public is entitled. Professional standards evolve over time. If cffDNA testing becomes routine, practitioners in reproductive health may be required to know what genetic information relevant tests may disclose and the implications of test results for the outcome of pregnancy; for children’s health; for family strategies and means to care for disabled children; and for what such children’s futures may offer. Practitioners may be required to counsel patients on the basis of such knowledge or to refer their patients to alternative sources of relevant information, including literature or pamphlets in language at a level of patients’ comprehension. Accordingly, patients’ routine access to cffDNA testing may add genetic understanding and counseling to required teaching and training curricula within the reproductive healthcare field and be a criterion of approval for accreditation of training programs and for professional licensure. It has been observed in the USA that “[t]he introduction of noninvasive prenatal diagnosis would require the ill-prepared medical system to change how patients are counseled and how cases are managed” [16]. 6. Patients’ choices following test results When test results cannot afford pregnant patients reassurance that their fetuses can be born as healthy children, patients may face difficult and even agonizing choices. An initial difficulty is to determine parameters of a child’s “normal” genetic diagnosis and a normally healthy future. In popular culture, normality may be assessed by superficial criteria such as appearance and ability to function in society. For children, this may mean ability, for instance, to attend a regular school, play safely with other children, and participate in sports. Medical assessment may be more statistical or normative, and consider risk levels of dysfunction, recognizing that—in the course of nature—all individuals bear genetic traits that may condition them to functional limitations or that make them susceptible to impairments. These may or may not materialize, depending on factors such as diet or environment that they may or may not be able to regulate. Newer forms of genetic diagnosis and health prognosis may show that every individual departs to some extent from a statistical or biological genetic norm. A challenge is to have patients understand the contrast between the possibility that a child will suffer effects of its genetic inheritance or predisposition and the probability of such effects, the assessment of risk, the possibility—although apparently low in cffDNA testing—of a falsepositive test result, and uncertainties due to genetic variants of unknown clinical significance. An immediate key decision for patients, where they have the choice of legal abortion, may be whether or not to continue their pregnancies. If they have the option of legal abortion but—as a matter of principle—would not accept this procedure, the test may be of value to them to prepare for any diagnosed conditions
in the children they deliver. Where testing is available, for instance in high-risk pregnancies, it should not be denied to eligible patients on the ground that they will refuse to terminate their pregnancies. Denial would appear as discrimination on grounds of conscience or religion, and characterize the test as abortion opponents have claimed it is: as a “search and destroy” exercise. Practitioners who would not participate in services or counseling that includes an abortion option should not suffer sanction or discrimination but should refer their eligible patients to accessible practitioners who do not object, in accordance with FIGO ethical guidelines [11]. All appropriately trained practitioners should counsel patients who receive cffDNA test results showing a trisomy and/or a comparable genetic condition about its implications and the management of affected children. Counseling should be non-directive and designed to present available options for patients to consider. Where abortion would not be lawful, counselors have to consider whether patients have feasible options of traveling to another, more accommodating, country or jurisdiction. For instance, the European Court of Human Rights has upheld the restrictive abortion law in Ireland on the ground that Irish law allows its nationals to receive lawful abortion services in other countries [17]. In 2012, for instance, 3982 women providing home addresses in the Republic of Ireland had abortions in the UK [18]. Conflict of interest The author has no conflicts of interest. References [1] Morain S, Greene MF, Mello MM. A new era in noninvasive prenatal testing. N Engl J Med 2013;369(6):499–501. [2] Brock DJH, Scrimgeour JB, Steven J, Barron L, Watt M. Maternal plasma alpha feto protein screening for fetal neural tube defects. Br J Obstet Gynaecol 1978;85:575–81. [3] Mason JK. The Troubled Pregnancy: Legal Wrongs and Rights in Reproduction. Cambridge: Cambridge University Press; 2007. [4] American College of Obstetricians and Gynecologists Committee on Genetics, The Society for Maternal-Fetal Medicine. Committee Opinion Number 545, December 2012. Noninvasive Prenatal Testing for Fetal Aneuploidy; 2012. [5] Abrams L. Prenatal testing: earlier and more accurate than ever. The Atlantic; December 11, 2012. [6] Nicolaides KH, Syngelaki A, Ashoor G, Birdir C, Touzet G. Noninvasive prenatal testing for fetal trisomies in a routinely screened first trimester population. Am J Obstet Gynecol 2012;207(5):374 eI-374, e.6. [7] Norton ME, Rose NC, Benn P. Noninvasive prenatal testing for fetal aneuploidy: clinical assessment and a plea for restraint. Obstet Gynecol 2013;121:847–50. [8] de Jong A, Dondorp WJ, de Die-Smulders CEM, Frints SGM, de Wert GMWR. Noninvasive prenatal testing: ethical issues explored. Eur J Hum Genet 2010;18(3):272–7. [9] Kevles DJ, Hood L. The Code of Codes: Scientific and Social Issues in the Human Genome Project. Cambridge, MA: Harvard University Press; 1992. [10] Brownfield v Daniel Freeman Marina Hospital. 256 California Reporter 240 (Court of Appeal, Second District, Division 4, California, 1989); 1989. [11] FIGO Committee for the Study of Ethical Aspects of Human Reproduction and Women’s Health. Ethical Issues in Obstetrics and Gynecology. London: FIGO; 2012 25–7. [12] Ranous JS, Sherrin JJ. Busting the blessing balloon; liability for the birth of an unplanned child. Albany Law Rev 1975;39:221–51. [13] Gleitman v Cosgrove. 227 Atlantic Reporter 2d 689 (New Jersey Supreme Court, 1967); 1967. [14] Procanik v. Cillo. 478 Atlantic Reporter 2d 755 (New Jersey Supreme Court, 1984); 1984. [15] Dickens BM. Medical errors: legal and ethical responses. Int J Gynecol Obstet 2003;81(1):109–14. [16] Benn PA, Chapman AR. Practical and ethical considerations of noninvasive prenatal diagnosis. JAMA 2009;301(20):2154–6. [17] Westeson J. Reproductive health information and abortion services: standards developed by the European Court of Human Rights. Int J Gynecol Obstet 2013;122(2):173–6. [18] UK Government. Abortion Statistics for England and Wales, 2012. Department of Health; 2013 52 . Table 12a.
