Non-disclosing preimplantation genetic diagnosis: Questions, challenges and needs for guidelines

Consider This
Non-disclosing preimplantation genetic diagnosis: Questions, challenges and needs for guidelines


Robert Klitzman, M.D.

Professor of Psychiatry, Director, Masters of Bioethics Program
Columbia University
New York, NY

Consider This:

Non-disclosing Preimplantation Genetic Diagnosis (ND-PGD) is performed, but controversial, raising many questions.  It has been used when prospective parents at-risk for mutations highly associated with serious disease (especially Huntington’s disease [HD](1)), do not want to know their mutation-status, but wish to ensure that no mutation-containing embryos are transferred.  Physicians would then transfer only mutation-negative embryos, and not tell the patient whether any mutation-positive embryos were identified.  In 2002, Stern et al. described using ND-PGD successfully with 10 couples.1  

Pros and cons of non-disclosing PGD

Several advantages and disadvantages have been articulated.  Few individuals at-risk for HD want to learn their mutation-status.  Caused by an autosomal dominant mutation, the disease lacks treatment, and leads to debilitating neurological and psychiatric symptoms and death, generally in the 4th-5th decade of life.  Many at-risk individuals see a mutation-positive test result as a “death sentence,” and only 3%-21% of at-risk adults get tested (e.g. only 3-5% in Sweden).(2)

Though the patient may not be infertile, ND-PGD requires IVF, which has certain risks.  Yet many patients may see the procedure’s benefits as outweighing these dangers.  Misdiagnoses can also occur, but prenatal confirmatory tests can be performed.

Still, critics argue that the parents’ right to remain ignorant of their HD mutation status may have limits.(3) Erez et al. describe a woman seeking PGD for HD since her husband did not wish to know if he had the mutation, though his father had it.  Yet due to testing done during a prior pregnancy, the woman’s clinician knew that the husband was mutation-free.  These authors argue that PGD would be a “sham”, subjecting the women to IVF and PGD, when the parents could instead have conceived naturally.  Moreover, if all the embryos are mutation-positive, a clinic “could feel compelled” to still perform an embryo transfer, though not in fact transferring any.iii

            Providers, these critics aver, should instead perform so-called exclusion testing – assaying the affected prospective grandparent, if alive, along with the fetus, to determine if the fetus has inherited one of this grandparent’s fourth chromosomes.iii  But exclusion testing has limitations: discarding 50% of healthy embryos. Genetic recombination can also occur between chromosomes.  IVF/PGD would still be required, though unnecessary if the patient lacks the mutation.iii  In 2009, the Dutch government banned both ND-PGD and exclusion testing for HD.(3)  ASRM and other US professional organizations and other countries do not appear to have issued guidance concerning ND-PGD.

Weighing pros and cons

Yet these critics have ignored several key points.  Many patients at-risk of HD face extreme stresses, and don’t want to undergo testing.  The Genetic Information Non-discrimination Act covers certain health insurance plans, but not life, disability, or long-term care insurance.  Many at-risk individuals therefore worry about potential discrimination.  

While Erez argues that ND-PGD could lead to sham transfers, this possibility presents only one of several potential outcomes of ND-PGD, and should be weighed in the context of other possible outcomes, and not necessarily be grounds for universally prohibiting it.  Informed consent can specify that no embryos may be available for transfer – i.e., that none may be both viable and mutation-negative.  Thus, “sham” transfers need not occur.  In IVF, embryos may be unavailable for transfer for various reasons unrelated to PGD (e.g., none appearing healthy).

Moreover, consideration of whether ND-PGD should be banned must include the fact that HD-diagnosed adults can suffer severe depression and suicide.  Among HD patients, suicide rates may be 10 times the national average.(4) At-risk individuals who undergo, and see benefits of, testing still have increased risk of depression and suicide.(5)  Importantly, they are also unrepresentative of most at-risk individuals, constituting only a small, self-selected minority who have undergone extensive pre-test counseling, and then feel able to cope with mutation-positive results.

Hence, banning ND-PGD raises critical questions of whether the risk of severe depression and suicide in 50% of at-risk prospective patients (who would have to undergo testing if wishing to have a mutation-free child, and if ND-PGD were banned) outweigh the risks of IVF to the mother.  After thorough informed consent, at-risk women who want children without HD may accept risks of IVF/PGD, since otherwise, they would have to conceive naturally and abort mutation-positive fetuses.  These women may know that IVF/PGD may be unnecessary (because they may be mutation-negative), but want to avoid anguish/suicidality they might experience if they learned that they had the fatal mutation.  Yet their wish for ND-PGD, after understanding the risks and benefits, isn’t unreasonable.  Moreover, to ban them from being able to make this choice about their own bodies, and instead compel them to undergo an abortion if they want HD-free children, is overly paternalistic, violating principles of autonomy.

            Though IVF/PGD might still be performed if patients are mutation-negative, this one potential scenario needs to be viewed not in insolation as the only outcome, but in the context of these other possibilities.  Debates about whether to prohibit ND-PGD for all patients should thus consider all benefits and risks of offering this procedure to the full spectrum of potential patients – not only to one, mutation-negative subset.

Providers should nonetheless proceed carefully, obtaining robust informed consent to ensure that patients understand all the risks and benefits (e.g., that no embryos may be available for transfer –– which could, but need not suggest the mutation’s presence of in some embryos), and that if no embryos are transferred, clinicians won’t discuss how many embryos were created and/or tested.  Some providers may still decline to offer this procedure; but these issues should at least be considered. 

Challenges in performing ND-PGD

Providers performing ND-PGD may also face questions regarding who exactly knows the patient’s or embryos’ mutation status.  Patients may request variations of ND-PGD, that clinics may accept, ponder, or reject – e.g., disclosing results to the spouse, but not to the at-risk patient.   Clinicians also vary in definitions of “non-disclosing” – e.g., whether genetic information is kept from not only the patient, but various clinicians and/or the spouse.  Providers may know the patient’s mutation-status before the embryos are tested, while the patient doesn’t want to know.  Neurologists or other physicians caring for the patient’s family may know or suspect the patient’s mutation status, and communicate it to the reproductive endocrinologist, geneticist or genetic counselor, straining these providers, affecting how they counsel the patient, and decide whether to perform the procedure.  Geneticists or genetic counselors may not all have experience with the procedure. 

Physicians who decline requests for ND-PGD face questions, too, of whether instead to encourage patients to try to conceive naturally, and then test and possibly abort the fetus if it is mutation-positive; or to use donor gametes; adopt; or consult another doctor.  Patients may also debate what to do, or shift over time, based on their medical and reproductive experiences – attempting ND-PGD, but if unsuccessful, trying to conceive naturally and undergoing prenatal testing, and abortion, pending results.  Patients may thus seek guidance from providers.  Ideally, clinicians who decline such requests should also educate and counsel patients as to why.

Clinicians who strongly suspect or know the patient’s mutation-status can face challenges regarding both the content and process of interactions – e.g., whether to be non-directive when the patient refuses to obtain the information.  Clinical geneticists or genetic counselors may find neutrality and non-directiveness difficult – i.e., to present options even-handedly, without favoring one option.  Clinicians may also face questions of whether they should discuss or encourage possible neurological consultation for the affected patients. 


            Though some critics have argued that all ND-PGD should be banned, it can offer important benefits that outweigh the risks in certain situations; but nonetheless poses challenges, warranting careful attention.  While prior literature has mentioned several advantages and disadvantages, several additional considerations arise.  These challenges can be difficult, balancing potential benefits, risks and patients’ autonomy.  Patients themselves may be uncertain and/or vary over time.  Providers may face tensions, knowing or suspecting, but being unable to disclose patients’ mutation status.  Providers may thus perform, decline or discourage ND-PGD, and/or recommend neurological workups, counsel patients about the pros and cons, perform ND-PGD in varying ways, or remain unsure.  Physicians performing ND-PGD should obtain thorough informed consent that includes the fact that “no available embryos” may suggest that the patient has the mutation.  ND-PGD, though now used largely for HD, may over time be used for other untreatable, stigmatized conditions for which genes are identified.

Professional organizations (e.g., ASRM) can assist clinicians by examining and considering guidelines regarding whether, when, and how to approach these discussions, articulating “best practices”, including issues clinicians should discuss with patients.  Such guidelines could specify what factors to weigh (e.g., risks of severe depression and suicide among mutation-positive patients), and why, who exactly should know the information, and what the informed consent should include (e.g., that no embryos may get transferred due to reasons other than lack of mutation-free embryos).  Enhanced professional and patient education can also help.  Future research should examine how often patients request and/or providers perform ND-PGD, and how, who receives the genetic information, for what conditions, and with what outcomes.

            In sum, despite arguments to prohibit all ND-PGD, it may aid numerous patients, but can pose challenges that future guidelines, professional and patient education, and research, can and should address.  


Funding for this paper is supported by grant number UL1 RR024156 from the National Center for Research Resources (NCRR), The Greenwall Foundation, and the John Simon Guggenheim Memorial Foundation.  The author would like to thank Patricia Contino for her assistance with the preparation of the manuscript. 


(1) Stern HJ, Harton GL, Sisson ME, Jones SL, Fallon LA, Thorsell LP, et al. Non‐disclosing preimplantation genetic diagnosis for Huntington disease. Prenat Diagn. 2002 Jun 1;22:503-7.

(2) Wahlin TB. To know or not to know: a review of behaviour and suicidal ideation in preclinical Huntington's disease. Patient Educ Couns. 2007 Mar 31;65:279-87. DOI: 10.1016/j.pec.2006.08.009

(3) Erez A, Plunkett K, Sutton VR, McGuire AL. The right to ignore genetic status of late onset genetic disease in the genomic era; Prenatal testing for Huntington disease as a paradigm. Am J Med Genet A. 2010 Jul 1;152:1774-80. DOI: 10.1002/ajmg.a.33432

(4) Bird TD. Outrageous fortune: the risk of suicide in genetic testing for Huntington disease. Am J Hum Genet. 1999 May 31;64:1289-92. DOI: 10.1086/302388

(5) Quaid KA, Eberly SW, Kayson‐Rubin E, Oakes D, Shoulson I, Huntington Study Group PHAROS Investigators and Coordinators. Factors related to genetic testing in adults at risk for Huntington disease: the prospective Huntington at‐risk observational study (PHAROS). Clin Genet. 2017 Jun;91:824-831. DOI:10.1111/cge.12893 [Epub 2016 Nov 24]