Mitochondrial Replacement Techniques: Ethical, Social, and Policy Considerations (2016)

4

Regulation and Oversight of MRT in Humans

As discussed in Chapter 3, the committee believes that the constellation of ethical, social, and policy issues surrounding mitochondrial replacement techniques (MRT) does not preclude clinical investigations of these novel techniques, but does warrant proceeding with caution while adhering to a circumscribed set of conditions. Within the realm of the U.S. Food and Drug Administration (FDA) regulatory review and oversight, MRT entails challenges that would demand tailored regulatory review and oversight strategies. These challenges include (1) establishing a sufficient level of preclinical evidence, which would entail creating, manipulating, and possibly destroying human embryos for research purposes; and (2) drawing conclusions from very small sample sizes for first-in-human research that would challenge even well-established rare disease research methodologies, introduce challenges for protecting participant privacy, require long-term follow-up of potential offspring beyond traditional clinical investigation evaluation stages, and challenge decisions about when it would be appropriate to expand MRT.

This chapter reviews the trajectory of any potential preclinical and clinical investigations of MRT and the ethical, social, and policy considerations that would need to guide MRT investigations throughout the phases of research, the regulatory approval process, and postapproval. Addressed in turn are assessment of benefits and risks, submission of preclinical evidence prior to authorization of clinical investigations of MRT in humans, conditions for clinical investigations, principles and practices that should guide clinical investigations, extension of MRT research to female embryos, informed consent, and guiding principles for oversight.

ASSESSMENT OF BENEFITS AND RISKS

In making regulatory determinations, FDA must form a conclusion that a treatment or technology is safe, that it is effective for its intended use, and that its benefits outweigh its risks. In assessing effectiveness, FDA applies standards to ensure the highest possible scientific validity and integrity. FDA’s judgment about benefits and risks takes into account (but is not limited to) the subjective perspective and experience of the expected beneficiaries of the treatment or technology.

Those affected by mitochondrial DNA (mtDNA) diseases and at risk of transmitting them to their children are uniquely positioned to inform FDA’s understanding of the clinical and personal context of these diseases. Proponents of MRT sometimes describe the use of the techniques as either a preventive measure or a therapy for children with mtDNA disease. Because in vitro fertilization (IVF) techniques are required as part of the MRT process to create an embryo, MRT would not treat a preexisting person or prevent a likely medical condition in an already existing individual. With this in mind, MRT has at least two potential benefits.

The first potential benefit of MRT is subjective, and depends on how important it is to individuals at risk of transmitting mtDNA disease to have children who are genetically related by nuclear DNA (nDNA) (but not mtDNA) and thereby at significantly reduced risk of manifesting mtDNA disease. Another suggested benefit is the reduction in the number of children who would be born with serious mtDNA disease as a result of access to this reproductive technology. At present, women at known risk of passing on serious mtDNA diseases choose childlessness, adoption, or the use of donor oocytes or embryos (preimplantation genetic diagnosis [PGD] is not offered by most U.S. academic medical clinics due in part to the fact that it is not a reliable means of avoiding transmission of mtDNA disease, as discussed earlier). But others choose to have genetically related children through unassisted sexual reproduction, and some of the resulting children are likely to be affected by mtDNA disease of varying severity at some point in their lives. If MRT were available, some women would have a reproductive option that could allow them to reduce the likelihood of passing on the disease mutation, and if MRT were successful for them, the overall number of children born with a risk of serious mtDNA disease would be somewhat reduced.

Both of the above benefits are relevant in determining whether the risks of MRT are reasonable in relation to its potential benefits. However, one of those benefits would accrue primarily to the prospective parents and the other at a population level; neither would accrue to the children who would be born as a result of MRT and thus would not have existed (either with or without mtDNA disease) but for the performance of MRT. This complicates

assessment of the benefits and risks of MRT. Typically, assessment of benefits and risks in clinical medicine is performed in contexts in which the same individual both realizes benefits and bears risks. In biomedical research, by contrast, some individuals are asked to consent to bear risks voluntarily to enable potential benefits that would be enjoyed largely or even exclusively by others. MRT challenges both paradigms in that the child born would not exist but for the MRT procedure, and so could not be asked for or give consent to participating in the research that led to his or her existence. This scenario raises important challenges related to both consent (discussed later in this chapter) and assessment of benefits and risks. These challenges argue for an approach that entails weighing, first and foremost, the probability of significant adverse outcomes borne by the children born as a result of MRT against the benefits accruing to families desiring children who are related to them through their nDNA.

The irreversibility of MRT introduces additional complexity to assessment of the appropriate balance of benefits and risks, and supports the imposition of conditions that could minimize risk until there was a high degree of confidence in the success of the techniques. The irreversibility of the “intervention” also creates challenges relative to respecting the right of research subjects to withdraw from research. Similar issues of irreversible risk and long-term assessment of benefits and risks have been navigated in somatic cell gene transfer, as well as in cell-based interventions and some surgical therapies. Research investigations with these techniques must deal with the tension among subjects’ right to withdraw from research, the irreversible nature of the intervention, and the research goal of long-term collection of information (with some investigations following subjects until they die and seeking permission for autopsy as part of research participation). MRT shares many similarities with these other research areas, with the addition that a child—who did not and could not consent to long-term follow-up—would be born if the research were successful. As in other examples, children who become research subjects once they are born would retain the right to withdraw from further participation in the research protocol, and hence discontinue their participation in systematic evidence collection, even if they could not effectively “withdraw” from the intervention.

Broader challenges affecting the assessment and balancing of benefits and risks of MRT stem from the fact that there are five potential parties with interests affected by MRT: (1) individuals who provide gametes (oocytes or sperm) that are used to construct embryos, (2) the intended parents, (3) the gestational carrier (if needed), (4) the child born as a result of MRT, and (5) the potential future offspring of female children born as a result of MRT. Each of those parties could be exposed to risk directly or indirectly at some point in the preparation for and application and outcomes of MRT. Indeed, some of these risks are no different from those encountered in tech-

niques commonly used in clinical applications of assisted reproduction. In the research context, risks to all parties involved in the research process need to be considered, even if they technically do not qualify as “research subjects” under federal regulations. In the case of MRT, attempts to minimize risk and burden for one of the above parties could interact with risk for another. However, the health and well-being, and minimizing risk of harm to any future children born as a result of MRT warrants priority in balancing benefits and risks in the design of clinical investigations.

As a general matter, parents put their children at some risk, even when it entails no direct benefit to the child. For example, parents take children with them in a car when running errands, which increases the risk of injury with no benefit to the child. In the context of human reproduction, this issue has arisen repeatedly with the introduction of new technological advances. Notably, it was the subject of discussion when infertile couples sought to use IVF and its variants (e.g., intracytoplasmic sperm injection [ICSI]). Protecting the health and well-being of future children demands that their safety be the primary value in any assessments of the benefits and risks of MRT.

PRECLINICAL EVIDENCE TO SUPPORT MRT

FDA requires that sponsors submit preclinical evidence before clinical investigations in humans are authorized. Preclinical studies serve several purposes, including enabling researchers to characterize risk, optimize techniques, and establish that the prospect of clinical application warrants investigation in human subjects. The goals of preclinical research on MRT would be to determine whether the techniques were reasonably safe for initial use in humans and whether they exhibited effectiveness that would justify clinical applications and commercial development.

FDA has traditionally demanded some evidence of “feasibility and rationale for clinical use” (FDA, 2015) for gene and cell therapies, in addition to evidence of safety, before approving the move from preclinical research to clinical research in humans. In the case of MRT, the effect of the exposure would go beyond that of traditional gene and cell therapies in that a child would be born as a result of the procedure. As discussed in Chapter 2, proof-of-concept studies for MRT have been conducted to varying degrees in animals (see also Appendix B for more detailed information on MRT research conducted to date). The main outcome measures in animal studies of maternal spindle transfer (MST) and pronuclear transfer

(PNT), which include mtDNA carryover and resultant heteroplasmy levels, have been studied in mice, “mito mice” (which carry a large-scale mtDNA deletion), and rhesus macaques. Animal studies to date have found varying levels of mtDNA carryover, ranging from undetectable (Tachibana et al., 2009) to 39.8 percent in first-generation mice and 22.1 percent in secondgeneration mice in the only study that has evaluated the effects of MRT in second-generation animals (Wang et al., 2014).

FDA guidance indicates that the kind, duration, and scope of required preclinical evidence will vary with the duration and nature of the proposed clinical investigations (FDA Center for Biologics Evaluation and Research, 2013). As suggested by FDA’s Cellular, Tissue and Gene Therapies (CTGT) Advisory Committee, additional preclinical evidence for MRT would likely need to include studies of a sufficient number of animals (mothers and offspring) from various species, with evaluation of safety over the long term—through all developmental stages and possibly extending to multigenerational follow-up (Liang, 2015). Multigenerational follow-up might not be necessary for animal research to support first-in-human clinical investigations in males only given that the limitation to males would avoid transmission to future generations. Before MRT was extended to female embryos, however, animal studies of second, and perhaps third, generations would need to be performed to collect data on the techniques’ intergenerational safety and efficacy.

CONDITIONS FOR CLINICAL INVESTIGATIONS

Once safety had been established through preclinical research, MRT could be tested further through the transfer of embryos to intended mothers (or gestational carriers), with the intended result of the birth of children at greatly reduced risk of inheriting pathogenic mtDNA mutations known to cause severe disease. Ethically acceptable clinical investigations of MRT would depend on certain restrictions and conditions being in place for initial clinical investigations.

Restriction to Women at Risk of Transmitting Serious mtDNA Disease

A traditional factor in decisions regarding the initiation and conduct of clinical investigations is optimizing the balance of benefits and risks of the intervention to be investigated. Given the novelty, complexities, and uncertainties inherent in MRT, minimizing risk in the conduct of first-inhuman clinical investigations of these techniques would need to be weighted heavily in favor of the health and well-being of the child born as a result of MRT. Simultaneously, maximizing the benefit gained from undertaking MRT would be an important consideration.

As discussed in Chapter 2, mtDNA diseases are clinically heterogeneous; can have early or late onset; and can result in negative health outcomes across a wide range of severity, including early death. In the case of MRT, maximizing benefit would be inextricably linked to the expected natural history and severity of the mtDNA disease at risk of being transmitted. In this sense, benefit would be maximized in initial clinical investigations by preventing the transmission of those mtDNA diseases known to be the most severe. In addition, to appropriately manage the balance of benefits and risks in MRT clinical investigations, individuals who provided oocytes would need to be pretested to ensure that they were not carriers of known pathogenic mtDNA mutations. It remains possible that an oocyte provider’s gamete could harbor a pathogenic mutation not present or detectable in easily tested tissues, such as the provider’s blood, cheek swab, or urine sample. Still, such testing would minimize the likelihood that the mtDNA in the provided oocyte would introduce pathogenic mutations whose transmission to the child MRT would be designed to avoid. Once the embryo had been created via MRT, performing PGD to confirm the absence of mtDNA mutations would also help ensure that this scenario did not occur.¹

The desire to avoid life-threatening illnesses generally comports with societal values surrounding the use of reproductive technologies for preventing the transmission of inherited genetic disease. A 2005 survey of public opinion on new reproductive technologies indicated that among this group of U.S. survey participants, there was general agreement that it is appropriate to use reproductive technologies to avoid life-threatening illnesses with an early onset (Kalfoglou et al., 2005a,b). There was less support for the use of reproductive technologies when the disease to be avoided was less severe, non-life-threatening, or characterized by adult onset of symptoms. However, survey participants were largely sensitive to individual perceptions of disease severity, quality of life, and suffering caused by a particular disease, noting that these are extremely personal concepts.

Inclusion criteria for women carrying mtDNA mutations who wanted to participate in MRT research would need to reflect the societal value of avoiding life-threatening illnesses. The research would need to be limited to women who otherwise were at risk of transmitting a serious mtDNA disease, and to cases in which the mutation’s pathogenicity was not disputed, and the clinical presentation of the disease caused by the mutation was predicted to be severe and characterized by early mortality or substantial

_________________

¹ As previously described, PGD may not be a reliable method for preventing mtDNA disease transmission in women at known risk of transmitting mtDNA disease because of limitations related to the complexities of mitochondrial genetics. With the advent of increasingly sensitive and accurate sequencing technologies, however, PGD is expected to be a reliable technique for determining the efficacy of MRT prior to embryo transfer.

impairment of basic function. In addition, the primacy of the interests of the child dictates that selection of oocyte providers would need to include genetic testing to confirm that the provider’s oocytes harbored no known pathogenic mtDNA mutations.

Health of the Gestational Carrier

The overall health status of the woman who would carry the pregnancy of the child born as a result of MRT (i.e., the intended mother or a gestational carrier, if needed) would need to be a key consideration in the design of inclusion criteria for potential clinical investigations of MRT. In keeping with the principle of minimizing risk to favor the health and well-being of the future child, inclusion criteria for a gestational carrier would need to be based on minimizing the risk of adverse health effects to the future child while also taking into consideration the impact of carrying the pregnancy on the health of the gestational carrier. If the intended mother planned to carry the pregnancy, her medical history and available evidence on pregnancy and mtDNA disease would make it possible to determine whether she would be able to complete the pregnancy without significant risk of adverse consequences to her health or that of the future child. If a gestational carrier were being used, she, too, would need to be healthy enough to carry the pregnancy to term and not present with any known risk factors for serious adverse conditions in the future child. The committee notes that a 2014 meeting of the FDA’s CTGT Advisory Committee included significant discussion on the topic of gestational carriers in MRT clinical investigations. Any future decision about the suitability of including gestational carriers in the research phase of MRT would be informed by the agency’s internal and external experts.

Initial Restriction to Males

As explained in Chapter 3, MRT in male embryos would not constitute heritable genetic modification. Because of the scientific uncertainties associated with these novel techniques and because MRT in female embryos would have the effect of creating heritable genetic modification, an appropriately cautious approach to MRT research in the United States would need to include restricting initial first-in-human clinical investigations to male embryos. This restriction would be justifiable in two regards.

First, unforeseen consequences of MRT—for example, health issues due to cellular manipulation, mtDNA-nDNA incompatibility, or failure to eliminate mtDNA disease—could become apparent in the first generation. By restricting initial investigations to males, these issues could be addressed in the first generation, without the risk of their affecting future generations.

Second, certain issues would arise only in female offspring of MRT—for example, the potential that future children could inherit a higher level of pathogenic mtDNA molecules relative to the first generation (Bredenoord et al., 2015). Although issues arising only in female offspring could not be resolved as long as MRT produced only male offspring, performing MRT initially only in males would allow preclinical research on intergenerational effects to continue while at the same time allowing families to use MRT to have male children with a significantly reduced risk of mtDNA disease.

While there is ethical debate about the acceptability of sex selection, the restriction recommended by the committee is predicated not on selection of one sex over another, but on the need to proceed slowly and to prevent potential adverse and uncertain consequences of MRT from being passed on to future generations. Bredenoord et al. (2015) observe that sex selection for medical reasons is generally accepted and relevant to the case of MRT, noting that PGD was initially introduced to select female embryos so as to avoid X-linked disorders (Handyside et al., 1990). The authors further note that sex selection for nonmedical reasons is regarded by many people as “morally problematic,” but that sex selection in the context of MRT would be health-related and represents a use that even those uncomfortable with sex selection would often find compelling. Appleby (2015) also suggests that an initial restriction of the use of MRT to males would be a “worthwhile limitation” because the research could “provide additional confidence” that MRT would be safe for the creation of females and subsequent generations.

In contrast, the report of the Nuffield Council on Bioethics on MRT in the UK context argues that restriction to male embryos would be unacceptable because it would result in an “experimental” group of male children, and “the boys born would need to be monitored throughout their lives and deemed healthy before females could be conceived in this way: they would in effect be experiments” (Nuffield Council on Bioethics, 2012, pp. xvi, 86). However, the context in which the committee undertook this study was focused on the prospect of initial investigation of MRT, and such initial investigation, if successful, would result in the first offspring born through MRT. The committee does not choose to characterize these children as “experimental”; however, any such births would in fact be part of an investigational context in which the first humans were produced following use of a novel technique. This unique combination of characteristics (novel technique; unique ethical, social, and policy concerns; and first-in-human use) argues for careful and responsible initial steps that would avoid risks to the extent possible and minimize them when they could not be avoided. In this framing, the limitation to male embryos would be a matter of responsible clinical investigation focused on reducing a significant risk rather

than a matter of sex preference. Notably, in at least one reported instance, PGD for preventing the transmission of mtDNA disease utilized selection of male embryos for the purpose of “avoid[ing] inheritance of the mutation in the third generation” (Treff et al., 2012, p. 1237).

Restricting initial investigations to male embryos would admittedly limit the potential benefits of the research because the research would yield no information about the effects of heritable transmission of mtDNA. An initial restriction to males also would mean that all embryos created through MRT would need to be tested for sex chromosome determination, with female embryos being frozen, donated for research, or discarded. In hypothetical but foreseeable instances in which the only MRT embryos suitable for transfer were female, some intended mothers would be unable to complete the study protocols. In addition, families who wished to have only female offspring or who were uncomfortable with sex selection would not be eligible for initial investigations.

While there are real issues related to limiting initial investigations of MRT to male embryos, the committee believes the trade-offs involved are necessary and justifiable to effectively eliminate the risk of introducing deleterious heritable genetic modifications, and are consistent with eligibility criteria, design features, and research staging used for clinical investigations in other realms of medical innovation.

Expertise of Investigators and Centers

Most MRT approaches contemplated at present would involve highly intricate micro-manipulations of human gametes and/or embryos. Use of the techniques would therefore require operator skill, which evolves over time, varies from one individual to another, and resists specification in a protocol. The inability to standardize interventions makes it extremely challenging to evaluate them. In this respect, MRT studies would face design issues similar to those encountered in surgery. The difficulty of the MRT process was described at the meeting of FDA’s CTGT Advisory Committee in February 2014. Evan Snyder, summarizing responses to a question about how FDA should control the production process for MRT, said the process “requires an enormous amount of skill,” and it “should only be done by specialists who have been qualified, and in specialized centers, at least initially.” He noted that “every stage in the manufacturing process needs to be monitored—the operators, the equipment, the preparations” and that ongoing quality control would be needed at each one of those stages.

In draft plans for the regulation and licensing of MRT, the United Kingdom’s Human Fertility and Embryology Authority (HFEA) (HFEA, 2015) proposed that MRT be restricted to clinics licensed specifically to perform

it. The licensing application would include evidence of the competence of the staff and the appropriateness of the premises for performing MRT. Specifically, all staff would have to be “suitably qualified, trained and assessed as competent for the tasks they perform,” as evidenced by information on staff experience in performing micro-manipulation on oocytes or embryos, as well as specific experience carrying out MRT and any other relevant information (HFEA, 2015, p. 4). The application also would require evidence of “suitable validation of equipment and processes” (HFEA, 2015, p. 4).

mtDNA Haplogroup Matching

As discussed in Chapter 2, MRT could entail some risk of adverse health effects related to nuclear-mitochondrial genome incompatibilities arising from the artificial combination of nDNA and mtDNA from genetically distinct lineages. This risk remains a significant subject of scholarly debate, even among experts in mitochondrial biology (IOM, 2015). With regard to the design of potential MRT clinical investigations and in keeping with the principle of minimizing risk to children born as a result of MRT, should FDA’s review of the preclinical data package reveal compelling evidence that mtDNA haplogroup matching between potential oocyte providers and intended mothers might mitigate the risk of mtDNA-nDNA incompatibilities resulting from MRT, such matching would be a reasonable inclusion criterion for initial investigations. Depending on the degree of match required, a decision to require haplogroup matching would most likely decrease the pool of oocytes provided by individuals for each procedure and thus potentially limit the overall probability of its success. Therefore, FDA would need to weigh requiring haplogroup matching as a means of mitigating risk against the potential effect of a decreased pool of available oocytes.

To the extent that mtDNA contributes to one’s sense of identity as associated with ancestry, haplogroup matching could have the benefit of retaining these ties. Haplogroup matching could restore the ancestral link to the intended mother’s lineage in a child born as a result of MRT. Yet while the notion of retaining ancestral and kinship ties might be a significant value for some people, this consideration is secondary to the principle of minimizing risk to future generations in crafting MRT clinical investigations, in order to maximize the possibility of the safest and most efficacious outcome.

• Initial safety is established, and risks to all parties directly involved in the proposed clinical investigations are minimized. Because attempts to minimize risk and burden for one of the parties could interact with risk for another, minimizing risk to future children should be of highest priority.
• Likelihood of efficacy is established by preclinical research using in vitro modeling, animal testing, and testing on human embryos as necessary.
• Clinical investigations are limited to women who otherwise are at risk of transmitting a serious mitochondrial DNA (mtDNA) disease, where the mutation’s pathogenicity is undisputed and the clinical presentation of the disease is predicted to be severe, as characterized by early mortality or substantial impairment of basic function.
• If the intended mother at risk of transmitting mtDNA disease is also the woman who will carry the pregnancy, professional opinion informed by the available evidence determines that she would be able to complete a pregnancy without significant risk of serious adverse consequences to her health or the health of the fetus.
• Intrauterine transfer for gestation is limited to male embryos.
• Clinical investigations are limited to investigators and centers with demonstrated expertise in and skill with relevant techniques.
• FDA has reviewed mtDNA haplogroup matching and if compelling, considered it as a means of mitigating the possible risk of mtDNA-nuclear DNA (nDNA) incompatibilities.

RESEARCH IN HUMAN OOCYTES AND EMBRYOS

In addition to animal investigations, preclinical research on MRT would likely entail extensive experimentation on human gametes and embryos with no intention of performing intrauterine transfer to establish a pregnancy in a woman. Such research might be necessary to learn about and optimize the physical manipulations of oocytes and embryos required for MRT, establish optimal timing for applying the techniques in gamete provider and intended mother gametes, and provide a better understanding of the appropriate application of reagents to achieve desired effects.

Initial published studies of the safety and efficacy of MRT in the human system have tested the techniques in human oocytes provided by healthy volunteers (Tachibana et al., 2013) or parthenogenetically activated oocytes

(Paull et al., 2013) in the case of MST, or in abnormally fertilized zygotes (Craven et al., 2010) in the case of PNT. The United Kingdom’s Newcastle Group is currently engaged in research aimed at comparing MST in human oocytes with PNT in normally fertilized human zygotes.

FDA performs in-depth review of in vitro and animal studies (i.e., preclinical studies) before granting permission to commence clinical investigations with human subjects. In the case of MRT, it would be important to accumulate sufficient preclinical data on how the manipulation of gametes or embryos might affect the resulting embryos so as to reduce the risk of harm to children born as a result of MRT during clinical investigations. Preclinical research involving embryos of varying quality that would not be transferred would likely be necessary to produce the data necessary to protect future children.

MRT would involve the creation, manipulation, and possible destruction of embryos not only in the preclinical research phase but also during clinical investigations and perhaps in clinical use. As discussed in Chapter 3, clinical use of MRT, even at its most efficient and successful, might require the creation of multiple embryos to produce a viable pregnancy leading to the birth of a child. The creation, manipulation, and destruction of embryos have long been controversial in the United States. Various perspectives exist on the “moral status” of the embryo, with some considering it to be a human being and thus entitled to the same protections. The creation of embryos specifically for research is particularly controversial, as the committee heard from presenters and public commenters during its public sessions (Darnovsky, 2015; Fitzgerald, 2015; Zoloth, 2015). It was the subject of discussion by the 1994 Human Embryo Research Panel and has been the focus of debates about research on nuclear transfer of somatic cells and on embryonic stem cells. In addition to these ethical debates, federal funding for research on embryos is restricted by the Dickey-Wicker amendment (see Chapter 2). While the creation of human embryos solely for research purposes is not prohibited under federal law in the United States (although some states are more restrictive), there are significant restrictions on the federal funding of such research. Even an agency request that data from such research be submitted in support of an Investigational New Drug (IND) application to start first-in-human research may well be controversial.

For both preclinical and clinical investigations of MRT, researchers would need to procure oocytes or embryos. If preclinical research required

the procurement of oocytes containing abnormal mtDNA, women with mtDNA mutations would be exposed to the oocyte stimulation and retrieval process, but without the benefit of potentially creating a child via MRT.

In part as a result of these types of ethical concerns surrounding the procurement of oocytes or embryos and the creation, manipulation, and destruction of embryos, the National Institutes of Health (NIH) developed guidelines in 2009 that articulate detailed standards for “ethically responsible” procurement of embryos for NIH-funded human embryonic stem cell (hESC) research (NIH, 2009).² These guidelines require, for example, that hESCs be derived from embryos that were created for reproductive purposes and are no longer needed, that all options for the embryos were explained to the potential donor, and that no payments were offered for the donated embryos. The guidelines also require a clear separation between the decision to create embryos for reproductive purposes and the decision to donate the embryos for research, as well as a detailed informed consent process. These standards, while specific to NIH and to the hESC context, address many of the same ethical, social, and policy issues that could arise in the provision of gametes or embryos for MRT.

_________________

² While the hESC research is federally funded, the procurement of embryos and the derivation of the stem cells are not federally funded.

PRINCIPLES AND PRACTICES TO GUIDE CLINICAL INVESTIGATIONS

A range of criteria would need to be satisfied for MRT investigations to be ethically acceptable and scientifically valid. To this end, such investigations would need to be guided by the principles and practices detailed below.

Health and Well-Being of Future Children

Given the novelty and unknown potential risks of MRT, clinical investigations would need to proceed with caution, with the health and well-being of the potential child being considered at every step. The balance of benefits and risks would fluctuate as investigations moved from initial stages into studies involving greater risk (e.g., different techniques) or less benefit (e.g., populations with less severe mtDNA disease). The conditions for initial investigations laid out in Recommendation 1—including restriction to serious mtDNA disease, a healthy gestational mother, initial restriction to male embryos, and expertise of investigators—represent an attempt to prioritize the minimization of risks to future children. As data accrued on the benefits and risks of MRT, these data would need to inform the assessment of benefits and risks for potentially less beneficial or riskier investigations. If initial investigations showed that the risk associated with MRT was low, (e.g., there were no short- or long-term detrimental effects for the resulting child), it might be appropriate to offer MRT to mothers at risk of passing on a less severe mtDNA disease, always prioritizing the health and well-being of future children in the balancing of benefits and risks. This cautious, staged approach would need to be taken in the design of initial and subsequent investigations—for example, in determining the eligibility of intended mothers, numbers of participants, and pacing of investigations.

Standardized Study Designs

Clinical investigations aim to establish three core elements: the optimal conditions for applying an intervention, its safety, and its efficacy. When studies are standardized and outcomes can be compared, establishing these elements is facilitated. However, efforts to standardize MRT studies would face a number of challenges.

Isolating the causal effects of any intervention requires standardizing treatments and populations so that outcomes can be compared. In the case of MRT, the “treatment” would be complex in that it would involve highly intricate manipulations of human gametes and/or embryos. As noted above,

treatment therefore would require operator skill, which evolves over time, varies from one individual to another, and resists specification in a protocol. Standardizing eligible patients to ensure a homogeneous and comparable population of study subjects could be difficult as well, as mtDNA diseases are a heterogeneous mix of genotypes and phenotypes—both of which can be highly unpredictable. Yet investigations might need to specify eligibility criteria based on genotypes to enable researchers to disentangle the effects of MRT from those that might otherwise arise randomly or as a result of variation in disease processes.

Outcomes might also be difficult to specify. Certain outcomes—such as the level of heteroplasmy in a defined set of tissues—might offer reasonably straightforward study endpoints. However, some potential effects of MRT—such as onset or severity of a disease or condition—might be difficult to detect without years of observation among very large populations. Because of the rarity of mtDNA disease, it might be difficult to recruit a sufficiently large sample to detect unintended effects of the intervention.

Finally, it might be challenging to use comparators in studies of MRT. Comparators enable researchers to isolate the effects of a treatment from those of other factors, such as the natural course of disease. Testing MRT using randomized designs would require some means of delivering sham interventions to a set of oocytes or zygotes. Yet, because MRT would involve many different types of manipulations, the choice of sham comparators would be far from obvious. Investigators might design a sham whereby MRT would be withheld from some women, who would instead be offered the usual standard of care (e.g., PGD). A more aggressive approach would be to perform MRT by transferring nDNA from the intended mother into another oocyte or zygote from the intended mother (rather than from an oocyte or zygote provider). Both of these options would present inferential and ethical problems. The former would (primarily) test the effects of oocyte or zygote manipulation, but would tell little about the added risk of the particular manipulations used in MRT. The latter approach would be unethical because performing MRT on oocytes or zygotes with parental (and pathogenic) mtDNA would subject future children to the risks of MRT with none of the potential benefits.

Despite these challenges, it would be essential to attempt to standardize clinical investigations of MRT to the extent possible. In addition, it might be beneficial for FDA to incorporate data from research or clinical practice outside of the United States to enhance the quality of the assessment of benefits and risks. The UK regulations allowing MRT as a clinical procedure went into effect at the end of 2015; FDA could utilize any data available from these procedures or MRT procedures performed in other countries.

Identity, Kinship, and Ancestry

As discussed in Chapter 3, MRT has implications for identity, kinship, and ancestry. The genetic contribution of three individuals might give children born as a result of MRT a unique perspective on their sense of self, to whom they are related and how, and their origins and lineage. While traditional oocyte or sperm provision raises similar issues, MRT is distinct from these procedures in that resulting children would be genetically related to three individuals. Clinical investigations of MRT would therefore need to include study of the potential psychological and social effects of MRT on notions of identity, kinship, and ancestry.

• The health and well-being of any future children born as a result of clinical investigation protocols of MRT should have priority in the balancing of benefits and risks with respect to the design of investigations, eligibility of prospective mothers, numbers of participants, and pacing of investigations.
• Study designs of clinical investigation protocols of MRT should be standardized to the extent possible so as to minimize the number of variables and enable valid comparisons and pooling of outcomes across groups.
• Data from research or clinical practices outside FDA jurisdiction should be incorporated into FDA’s analysis to enhance the quality of the assessment of benefits and risks.
• Clinical investigations should collect long-term information regarding psychological and social effects on children born as a result of MRT, including their perceptions about their identity, ancestry, and kinship.

EXTENSION OF MRT RESEARCH TO FEMALE EMBRYOS

As discussed above, restricting the first investigations of MRT to male embryos would initially eliminate the risk of deleterious health effects resulting from the introduction of heritable genetic modifications, allow time for evidence to be collected on safety and efficacy in the first generation of male children born as a result of MRT, and provide greater understanding of the effects of genetic modification via MRT. Regardless of how safe or efficacious MRT was found to be in clinical investigations with male embryos, however, moving to female embryos would introduce the additional ethical, social, and policy issues raised by heritable genetic modification of germ cells.

In addition to issues raised in Chapter 3 regarding heritable genetic modification, the use of MRT to create and transfer female embryos would raise novel questions in the research context. Any assessment of benefits and risks would need to take into account the risks of introducing unforeseen or unintended mtDNA mutations or unexpected effects of mtDNA and nDNA genome combinations that would be experienced not only by the immediate female offspring born as a result of MRT but also by all of their prospective progeny into the future. As recommended by the committee, these potential intergenerational risks could be avoided by limiting initial clinical investigations of MRT to male embryos. However, important information and potential benefits would be gained from eventually transferring female MRT embryos, including understanding the effects of heritable genetic modification on reproduction and the health of offspring eventually born to women who were born as a result of MRT. Significantly, transfer of female embryos would minimize the risk of passing on pathogenic mtDNA mutations that might otherwise be faced by all maternal members of a family’s lineage over generations, effectively preventing mtDNA disease in future generations of families known to be at high risk. Giving families the ability to bear female children is also a value to be respected and one that could be served only by transferring female embryos. The question is when in the course of investigation of MRT it would be acceptable to move to transferring female embryos. The committee identified three general criteria that would need to be satisfied before moving forward with MRT for female offspring: (1) compelling evidence of safety and efficacy in male embryos; (2) preclinical animal research showing evidence of intergenerational safety and efficacy; and (3) the existence of a shared framework concerning the acceptability of, and moral limits on, heritable genetic modification.

Compelling Evidence of Safety and Efficacy

Moving to transferring female embryos would constitute an important additional step in MRT human investigations. In addition to sharing the characteristics of MRT with male embryos, MRT involving female embryos would introduce intergenerational effects, whether those effects were positive or negative. Among the most significant concerns regarding heritable genetic modification resulting from MRT is the inability to limit unintended deleterious genetic effects to the individual born via MRT. A female born as a result of MRT who carried pathogenic mtDNA mutations could pass them on, whereas similarly affected males could not. Therefore, the committee’s view is that sufficiently robust evidence of the safety and efficacy of MRT in males would be necessary before introducing the additional risks associated with the potential intergenerational effects that would accompany transferring female embryos, regardless of how long it took to collect this evidence. Sufficiently compelling evidence that would reach the level of confidence envisioned by the committee would come from experience with numerous male children followed at least during their early childhood years. While the threshold for sufficient evidence might be difficult to gauge before first-in-human investigations began, FDA could consider establishing a minimum threshold to be met before moving to MRT in female embryos. For example, should FDA ever come to the point of granting a license for the application of MRT in male embryos, it would be on the basis of evidence suggesting that certain major risks could be excluded. This evidence appears likely to be relevant for both male and female embryos. The agency could link judgments about initiating investigations in female embryos to the grant of licensure in male embryos.

Preclinical Data on Intergenerational Effects

Clinical investigations of MRT in males would generate data on safety and efficacy only in the first generation, that is, the children born as a result of MRT. Data on the effects of MRT in subsequent generations could only be generated by transferring female embryos, allowing time for these females to reach sexual maturity and choose to reproduce, and then assessing the health and well-being of these subsequent offspring. Because these data could not be collected through MRT entailing the transfer of male embryos, sufficient preclinical evidence from animal models regarding intergenerational safety and efficacy would need to be gathered before clinical investigations of MRT involving female embryos were allowed.

Shared Framework on Heritable Genetic Modification

If and when sufficiently compelling evidence of safety and efficacy from experience with male MRT offspring and preclinical data on intergenerational effects were obtained, moving to transferring female embryos would remain a controversial step in that it would entail heritable genetic modification. As articulated elsewhere in this report, the committee views heritable genetic modification via MRT as distinct in relevant and important ways from modification of nDNA—distinctions that would inform the acceptability of going forward with female embryos if safety and efficacy criteria for MRT had been established and met. A productive public discussion and process has been initiated to establish a shared framework with respect to whether heritable genetic modification is acceptable and if so, under what circumstances and for what purposes. The committee believes its analysis can aid this ongoing discussion and that any decision about moving forward with MRT with female embryos should be informed by this discussion. Therefore, the committee recommends that if and when compelling evidence of safety and efficacy is established, a decision to move forward with transferring female embryos should be consistent with the established shared framework in effect at that time concerning the acceptability of techniques that result in heritable genetic modification of human embryos.

• clear evidence of safety and efficacy from male cohorts, using identical MRT procedures, were available, regardless of how long it took to collect this evidence;
• preclinical research in animals had shown evidence of intergenerational safety and efficacy; and
• FDA’s decisions were consistent with the outcomes of public and scientific deliberations to establish a shared framework concerning the acceptability of and moral limits on heritable genetic modification.

INFORMED CONSENT

Informed and voluntary consent of those deemed research participants in MRT clinical investigations would be required pursuant to federal guidelines and applicable state laws and institutional practices. As noted earlier, five potential parties could have interests affected in the course of the MRT process: (1) individuals who provide gametes (oocytes or sperm) used to

construct embryos, (2) the intended parents, (3) the gestational carrier (if needed), (4) the child born as a result of MRT, and (5) any potential future offspring of the child born as a result of MRT. Each of these parties has rights and interests deserving of protection, although they all might not necessarily be recognized as research subjects from a regulatory perspective (in accordance with federal or state regulations or institutional requirements or practices). MRT necessarily would involve an oocyte provider, but depending on the family structure of the intended mother pursuing MRT, sperm might be provided by either the intended father or another individual. The consent process would need to ensure that information about the MRT process was adequately disclosed and comprehended and that any decisions to participate in the MRT process were voluntary.

The research community has debated the utility of the informed consent process for years. Nonetheless, the complexities and uncertainties associated with MRT suggest that the consent process holds significant potential to provide a thoughtfully designed structure for what is ultimately a highly valuable and critical component of research. The consent process would need to ensure that those participating in MRT research understood what their participation entailed and that it was voluntary. While MRT presents a number of challenges to the consent process, as described below, efforts to develop best practices could provide a foundation for consent processes appropriate to novel reproductive technologies (Aldoory et al., 2014).

Individuals Who Provide Gametes

For women providing their oocytes for MRT research purposes, the process, and thus the implications for consent, would vary depending on the technique used. In the case of MST, the nuclear chromosomes would be removed from the provider’s oocyte and replaced with the intended mother’s nuclear chromosomes. Thus, in MST, the provider’s oocyte would be the focus of the MRT manipulation. This reconstructed oocyte would subsequently be fertilized by sperm from the sperm provider (either the intended father or another individual). In PNT, both the provider’s oocyte and the intended mother’s oocyte would be fertilized in vitro with sperm from the sperm provider to create two embryos. The pronuclei would be removed from both embryos, and the oocyte provider’s pronuclei would be replaced by the pronuclei of the intended mother to create a reconstructed embryo. Thus, in PNT, the provider’s fertilized egg (zygote) would be the focus of the MRT manipulation. If medically acceptable, the intended mother would likely gestate the embryo; if not, a gestational carrier could be used.

The sperm provider’s sperm would be used to fertilize one oocyte in the case of MST or two oocytes in the case of PNT to create embryos (one of which would be discarded after the pronuclei had been removed in the

case of PNT). Thus, in both MST and PNT, material in which the sperm provider had an interest (his sperm or a zygote fertilized with his sperm) would be the focus of the MRT manipulation.

MRT Research Procedures and Applications

As part of any clinical investigation of MRT, consent would need to be obtained for the series of interventions necessary to stimulate and collect oocytes (including both surgical and postsurgical procedures) and create embryos (if PNT were the technique being used), as well as for the intended use and disposition of a provider’s oocytes, and potentially embryos. Short- and long-term risks of oocyte retrieval and any unknowns associated with these data would need to be explained to individuals providing oocytes.³ Sperm providers would also need to be involved in the consent process for the use of their gametes to create embryos and the use and disposition of remaining sperm or embryos.

Individuals who provided gametes (both oocytes and sperm) would also need to be given the opportunity to understand the ethical, social, and policy issues associated with MRT research and the role their tissues would play in the research process. It would be necessary to explain to gamete providers the degree to which their gametes (or embryos) would be used for research purposes and stored indefinitely or destroyed.

Incidental Findings

Depending on the diagnostic techniques used to evaluate gamete providers and their gametes in the MRT research context, the consent process would need to include consideration of the possibility that the research would yield incidental findings with clear implications for participants’ reproductive or other health care decisions (for example, if the gamete provider or his or her gametes were to undergo tests that revealed a particular genetic trait or mutation that would affect such decisions). Mechanisms for delivering these findings to gamete providers would have to be determined. MRT researchers and institutions would have to be informed by existing guidance documents on how, when, and to whom such incidental findings are to be reported throughout the course of research (e.g., Presidential Commission for the Study of Bioethical Issues, 2013). For instance, if a

_________________

³ Some scholars have suggested a new category to address individuals who provide gametes. In the example of women providing oocytes for stem cell research, Magnus and Cho (2005) recommend the term “research donor” as distinct from “research subject” to signify that the risk incurred by women providing oocytes for research comes from the procurement of materials for research and not the actual research itself.

child born as a result of MRT were found to have a novel, unpredicted mtDNA disease, it would be necessary to consider whether to inform the oocyte provider of this result, as it could affect her health or the health of her children.

Incentives and Potential Financial Gain

The appropriate compensation of women and men who provide their gametes for research has been the topic of ethical and legal analysis in other clinical contexts. Some have suggested that individuals who provide gametes should receive financial reimbursement only for out-of-pocket costs or direct expenses incurred as a result of the procedures, as determined by an institutional review board (IRB) (IOM, 2013; NRC and IOM, 2010), thereby avoiding ethical issues because the gamete provider would derive no financial gain from participating in the research and would not be vulnerable to arguably undue levels of enticement. On the other hand, those who provide their gametes in the context of infertility treatments often receive financial compensation reflective of the time, inconvenience, and discomfort associated with screening, ovarian stimulation, and oocyte retrieval (ASRM, 2007); therefore, some suggest banning payments to gamete providers in the context of research would be unfair (Lo and Parham, 2009). Indeed, the International Society of Stem Cell Research’s recent recommendations regarding compensation of oocyte providers suggest it is appropriate to compensate for an oocyte provider’s time, effort, and inconvenience (Haimes et al., 2013). Practically speaking, moreover, finding oocyte providers in the absence of compensation is a notable challenge (Egli et al., 2011).

Any increased demand for provider oocytes and sperm resulting from the initiation of MRT research, however small, would have the potential to put some women and men of low socioeconomic status at risk for arguably undue enticement to donate gametes. In lieu of banning payments, which might be criticized as being paternalistic, the literature suggests there are opportunities to strengthen protections for all gamete providers, including those of low socioeconomic status (IOM and NRC, 2007; Lo and Parham, 2009; Lomax et al., 2007). It would be important for MRT researchers and institutions, in consultation with local review committees or a central IRB, to consider current guidance and emerging best practices in determining appropriate compensation for gamete providers, taking into account the demands placed on a gamete provider by an MRT research protocol. It would be necessary as well to give special attention to crafting a compensation and recruitment strategy that would not place women and men of low socioeconomic status in a position of being unduly enticed to provide gametes against their better judgment.

Provider Contact

An additional set of issues that would need to be part of a consent process for MRT clinical research participation or as a separate agreement with the intended parents relates to the social involvement, if any, of a gamete provider with the future child and his or her family. It is possible that a child born as a result of MRT would have an interest later in life in contacting his or her mtDNA or sperm provider. The consent process would need to include agreement among all parties as to whether gamete providers would remain anonymous to the child, or contact in the future would be possible, subject to applicable state laws (which might, in some cases, require open adoption in which future contact would be possible).

Management of Residual Gametes and Embryos

The consent process for MRT would need to include discussion of the disposition of any remaining gametes and embryos. Multiple attempts at MRT for establishment of a pregnancy could be necessary, so it would be important for gamete providers to know how any remaining gametes or embryos would be managed. In particular, if gametes or embryos were to be cryopreserved, a clear understanding of their longer-term management would be needed. For example, who would be responsible for storage costs, and who would make decisions about the use or destruction of the oocytes or embryos or their donation to other couples or for research purposes if they were no longer needed for the MRT investigation? In some cases, state laws or precedent cases could limit the gamete providers’ options, and this, too, would need to be explained to them.

Intended Mother and Intended Father (if applicable)

Consent Components Applicable Specifically to the Intended Mother

The consent process for an intended mother considering MRT in first-in-human investigations would likely be “a difficult and long-term process” (FDA Cellular Tissue and Gene Therapies Advisory Committee, 2014) requiring many conversations over time. As a participant in a first-in-human clinical investigation, the woman would assume the risk associated with the lack of prior information on the safety and efficacy of MRT in humans. The complexity of MRT also would present a psychological challenge to the intended mother in the form of overlapping uncertainties. For instance, she would have to weigh the benefits and risks, and the uncertainties, associated

with IVF,⁴ with MRT itself (for which no data from a born human exist), and with the role of mtDNA in human development (evolving data), as well as any potential health risks associated with transmitting this novel genetic combination (no data available) (Bonnicksen, 1998).

In addition to the benefits, risks, and uncertainties of MRT itself, the informed consent process for an intended mother who would be gestating the embryo would need to include discussion of the aspects of (1) the procedures involved in MRT (including PGD testing in embryos and prenatal tests in fetuses, such as chorionic villus sampling, amniocentesis, or cell-free DNA screening); and (2) the genetic testing processes, and their potential limitations, that would accompany the procedures. In addition, an intended mother would need to be made aware of the potential that a child with significant disability could be born, and of the difficult decisions she might face regarding pregnancy termination if prenatal diagnostic testing revealed genetic or developmental anomalies or other adverse outcomes. Follow-up conversations as part of the informed consent process could help ensure that research participants had adequate information about the testing procedures (including information about each test’s specificity, sensitivity, accuracy, risks, benefits, and limitations) used throughout the MRT process (McGowan et al., 2009).

Consent Components Relevant to Both the Intended Mother and the Intended Father (if applicable)

If intended mothers sought to avail themselves of MRT but did not have a male partner, there would not necessarily be an intended father. In those cases, an individual who provided sperm would be involved, for whom the applicable consent principles are described above. In many cases, however, there would likely be an intended father or co-parent, and although he or she might not also be a gamete provider, the legal and social role in raising the resulting child would make the following components of the consent process relevant to the intended mother and intended father or co-parent.⁵

Alternatives The consent process for MRT clinical research would need to include a thorough discussion of the alternative means of becoming a parent that would avoid the transmission of mtDNA disease, including their advantages and disadvantages. The discussion of these alternatives would

_________________

⁴ IVF success rates vary from 1.2 to 23.8 percent based on the presenting indication and the treatment approach pursued (CDC et al., 2015).

⁵ In the hypothetical case of a female co-parenting couple, in which one woman was the nDNA contributor (intended mother) and the other the mtDNA contributor (oocyte provider), the identification of one or both as “legal” mother would pose novel questions for the state courts.

need to be supported by a range of advisors and counselors to help inform the intended parents and answer their questions about potential participation in MRT research.

Research restrictions If protocols for first-in-human investigations required that only male embryos be transferred, it would be necessary to inform intended parents of the possibility that the oocyte retrieval and MRT process could result in only female embryos without pathogenic mtDNA and therefore otherwise suitable for transfer. Should this be the case, the intended parents would need to understand that they would be unable to have an embryo transferred as part of the research process.

Long-term follow-up The important role of long-term follow-up of any children resulting from MRT research would need to be highlighted in the consent process. It is possible that decades of observation would be necessary to detect subtle effects of such factors as epigenetic changes or variable levels of heteroplasmy in certain tissues. Lengthy observation periods also would be appropriate as a means of maximizing information gained from the small study samples that would be likely. Sponsors of clinical investigations of MRT would need to have a plan and a budget to support such long-term follow-up of any resulting offspring. Extended periods of regular, potentially invasive and intensive observation could add to the burdens of children resulting from MRT. Such follow-up could be especially burdensome if the children were otherwise healthy, because invasive monitoring would not be therapeutic for them.

Long-term follow-up would be necessary to determine whether there were issues with residual pathogenic mtDNA molecules (i.e., MRT did not effectively prevent the transmission of mtDNA disease), whether the manipulation of oocyte or zygote and the process of mtDNA replacement adversely affected the subsequent child, and whether there were any effects of possible mtDNA-nDNA mismatch. In the initial research phases, follow-up would likely include evaluations in infancy and early childhood to determine whether gross anomalies, developmental disabilities, mtDNA mutations, or signs or symptoms of mtDNA disease were present. It could be necessary to evaluate children to the point of sexual maturity to confirm that the reproductive system had not been adversely affected. Decades-long observation of children born as a result of MRT would be necessary to determine whether there were late-onset effects of MRT, and intergenerational follow-up would be necessary to track the health and well-being of subsequent generations if female embryos were transferred. While continued assent (for children) and consent (when they reach the age of consent) to this type of follow-up cannot be mandated, the intended parents would need to be well informed from the outset as to why long-term follow-up was crucial

for MRT research and to understand that it would be an important part of the child’s experience as part of the research protocol.

Privacy The first individuals participating in clinical investigation of MRT could be targets of intense media scrutiny. While research centers are required to institute measures to protect the confidentiality of health information and the personal identity of research participants, participants can choose to maintain their privacy or to make themselves known to the media. Assisted reproductive technologies (ARTs) are frequently the subject of media and public interest. Therefore, although the clinical research context offers protections for patient privacy, special attention would need to be paid to preparing prospective research participants, investigators, research institutions, and their staffs and media departments for the likelihood of high-profile attention, wanted or unwanted, associated with an MRT investigation. The ethics review committee(s) charged with evaluating any initial MRT clinical investigation protocols, and the associated consent processes, would play an important role in ensuring that provisions for protecting the privacy of research participants were adequate and that the relevant parties were appropriately informed and prepared for any unintentional and inadvertent disclosures.

Gestational Carrier (if needed)

As discussed earlier in this chapter, if the intended mother were unable to carry or were at high risk for complications associated with carrying a pregnancy, a gestational carrier might be deemed appropriate or necessary. Any such gestational carrier would need to have a clear understanding of the potential benefits, risks, and uncertainties associated with participation in the MRT process. For instance, gestational carriers would need to be made aware of the potential that a child with significant disability could be born, and of the difficult decisions she might face regarding pregnancy termination if prenatal diagnostic testing revealed genetic or developmental anomalies or other adverse outcomes.

Child Born as a Result of MRT

Consent by intended parent(s) to a process that would result in the birth of a child through MRT could not fully protect the interests and welfare of future children. As mentioned elsewhere in this report, protecting the health and well-being of future children born as a result of MRT needs to be the cornerstone for all assessments of MRT and balancing of its benefits and risks, including decisions surrounding the adequacy of

preclinical studies, justification for clinical investigations, and the design of first-in-human investigations.

As noted above, once a child had been born, investigators would need to obtain parental permission for such research-related procedures as blood sampling or tissue biopsy of a newborn. To meet this need, staged parental permission could be implemented, as was planned for the National Children’s Study, thereby avoiding a long and complex consent process for future interventions during enrollment in the study and allow parents to make decisions as they might arise (IOM, 2008).

The Nuffield Council on Bioethics (2012, pp. xvi, 88) argues that, in MRT clinical research, “consent to follow up would need to be included as a mandatory part of parental consent to participation in the trial.” Consent to participate in research is an ongoing process, not a one-time event or a signature on a consent document. A participant’s right to withdraw, without penalty, is recognized as a critical element of participation in clinical research. This right can create difficulties in the conduct of research, such as MRT, that requires long-term monitoring and follow-up. Optimal follow-up for MRT research could be decades long to make it possible to assess effects that might appear later in life or to monitor the health of offspring born to children born as a result of MRT. Parents of children born as a result of MRT would be asked to provide ongoing permission for the child’s participation in follow-up evaluations, even though they would have the right to decline, just as they could withdraw from the study at any time. Similarly, children born as a result of MRT would be asked to assent or consent (at appropriate ages) to further involvement in the research, which they would be free to decline at any time.

It is ethically permissible, within limits, to try to persuade research participants, including children who have reached the age of consent, to continue to participate in research. In fact, in the committee’s view, MRT research participants would have an ethical—though not legal—duty to remain involved in follow-up activities for their own benefit as well as that of other potential future users of MRT. It is reasonable for clinical researchers to use pre-enrollment consent discussions, as well as postprocedure discussions, to strongly encourage individuals to participate in follow-up. Reimbursement for costs and modest incentives, such as access to personalized medical services or general recognition and praise, are justifiable in some circumstances (Grant and Sugarman, 2004), although the individual’s eventual decision must be respected. No coercion or other efforts that undermine voluntary decision making are acceptable, either to encourage initial participation in or discourage withdrawal from research.

Protection of Future Generations

MRT clinical investigations that entailed creating and transferring female embryos for gestation would raise issues related to the transmission of heritable genetic modification to future generations. Other ARTs also involve heritable genetic modification, although without the unique combination of characteristics associated with MRT. For instance, prospective parents might use the assistance of a woman who provided her oocyte to have a child genetically related to one of those parents, and that new combination of genetic material would be passed on to future generations. As described throughout this report, however, MRT is unique in that in females it would result in a potentially heritable genetic modification comprising DNA from two women of different maternal lineage. This transmission of heritable genetic modifications to future generations as a result of MRT would constitute uncharted territory for any consent process. Evaluating the risks that MRT could pose to future generations is important from ethical, social, and policy perspectives; however, clinical investigations have no mechanism for seeking consent from future generations. Thus the potential effect of MRT on future generations needs to be a key consideration in broader policy discussions and research oversight related to MRT, becasue it cannot be addressed in the consent process for MRT clinical investigations.

• The nature of the MRT consent process for intended parents would need to reflect a research protocol that had been crafted with the health and well-being of future children in mind.
• Once a child had been born as a result of MRT and reached the applicable age, it would be necessary to carry out a more traditional process of parental permission and child assent, and eventually consent by the child, for participation in ongoing research assessments.

• For individuals who provide gametes, consent processes should reflect

• the range of MRT procedures contemplated for preclinical and/or clinical investigations and the general ethical, social, and policy considerations surrounding MRT;
• the management of incidental findings, should they arise;
• appropriate compensation, with sensitivity to socioeconomic status;
• the prospect of future contact between individuals who provided their gametes and children born as a result of MRT; and
• the management of residual eggs and embryos.

• For intended parents, consent processes should reflect

• information on the MRT research protocol, with focus on the implications for the health and well-being of resulting children;
• alternative ways of becoming parents that can avoid maternal transmission of mitochondrial DNA (mtDNA) disease;
• the management of and potential restrictions on access to embryos created through MRT (e.g., if initial investigations are limited to male embryos);
• preimplantation and prenatal genetic diagnostic tests that would be incorporated into clinical investigation protocols;
• the importance of long-term follow-up and how it would be part of the experience of any child born as a result of MRT; and
• the challenges of maintaining patient privacy given intense media interest in MRT.

• For children born as a result of MRT, consent processes should reflect assent (and eventual consent) for monitoring and research procedures to be performed after birth, up to and including seeking informed consent from the children upon their reaching the legal age of consent.

GUIDING PRINCIPLES FOR OVERSIGHT

Although MRT is in some ways similar to other reproductive technologies, it has a unique combination of characteristics that raises a novel collection of ethical, social, and policy issues. Because of this unique combination of characteristics, MRT would require special considerations across the trajectory of regulation and oversight—from preclinical studies to authorization of an IND, potential approval for clinical use, and postmarketing surveillance. These considerations could be addressed through the following guiding principles for oversight.

Transparency

MRT would entail sensitive and controversial procedures of great interest to many people, particularly those at risk of transmitting mtDNA disease to their offspring. FDA and other regulatory authorities would need to take every opportunity to inform the public and key stakeholders about all aspects of MRT within the constraints of legal obligations regarding confidentiality. The information to be shared could include preclinical work supporting regulatory decision making and relevant emerging scientific developments, as well as decisions regarding clinical investigations, approvals, and postmarket studies. Regulatory authorities would need to promote transparency by utilizing forums that would permit an exchange of information between the agency and the public; this process could employ existing venues such as public bioethics commissions, FDA advisory committee meetings and public workshops, or meetings of the NIH Recombinant DNA Advisory Committee (RAC). In addition, FDA would need to encourage sponsors to voluntarily waive confidentiality concerning protocol design and reporting of deidentified results whenever possible, while always maintaining the privacy of the individuals participating in the research.

Public and Patient Engagement

Because MRT is currently controversial, the question arises of how the public and key stakeholders can inform regulatory decision making. FDA’s decision-making process, however technical and preoccupied with assessment of benefits and risks, is ultimately informed by value judgments concerning such issues as clinical need and the availability of viable alternatives. In the case of MRT, larger debates about the ethics of reproductive interventions fall outside the agency’s mandate and core competencies. In the United States, scientific and political issues in this area are for the most part addressed separately, with political issues being resolved largely by legislation and technical issues by regulatory agencies. In the United Kingdom, by contrast, both are managed by the HFEA. However, given the nature of the issues raised by MRT and the subjectivity of elements of the assessment of its benefits and risks (e.g., the importance to prospective parents of having a genetic link to their offspring), public engagement in FDA’s decision-making process could be beneficial.

Other novel and controversial technologies have undergone similar public and patient engagement. Until 2014, for example, all gene-editing experiments were reviewed by the RAC, which holds public reviews and discussions when necessary. National-level bioethics commissions have been convened to address ethical, social, and policy considerations on such topics as cloning and stem cell research. And FDA’s Patient-Focused Drug

Development Initiative systematically gathers patient perspectives to inform the assessment of a product’s benefits and risks. The regulatory process for MRT would likewise need to incorporate the views of the public and patient populations through such mechanisms as periodic reports to the public, opportunities for public meetings, and ongoing exploration of the views of relevant stakeholders. In particular, FDA would need to encourage the participation of those affected by mtDNA diseases and at risk of transmitting them to their children, who are uniquely positioned to inform the agency’s understanding of the clinical and personal context of these diseases. Mitochondrial disease patient advocacy groups and mitochondrial medicine physicians and medical societies could also play a role in informing FDA in this regard.

Partnership

FDA would need to take full advantage of partnerships with regulatory bodies in other countries where MRT research or clinical investigation is occurring so information from this research could be pooled. During prelicensure stages, the goal is to increase the quality of protocol design and regulatory decision making while reducing redundancy and risk to patients and research participants. This goal is especially important for MRT given the rarity of mtDNA diseases and the small number of patients that would be research participants. The UK regulations allowing MRT went into effect in October 2014; FDA could use any data available from these procedures, or from MRT procedures performed in other countries, to improve its assessment of benefits and risks. If MRT were approved and entered clinical use in the United States, these partnerships would need to be maintained to enhance long-term postmarket surveillance and the use of risk management tools. FDA also would need to consider partnering with other federal agencies to take advantage of their expertise, such as that in regulatory science research at NIH or that in public health monitoring at the U.S. Centers for Disease Control and Prevention (CDC).

Maximizing Data Quality

The importance of standardization of study designs for research on MRT was discussed earlier in the section on principles and practices to guide clinical investigations. Given the likely small numbers of MRT re-

search participants, it would be critical to standardize study designs so the highest-quality data could be collected and pooled in support of the regulatory decision-making process. Despite the challenges of standardizing MRT studies discussed earlier, FDA would need to require, to the extent possible, that sponsors have adequate resources, use appropriate designs, and plan studies that would enable cross-referencing and pooling of data.

Another aspect of data quality is the periodic review and evaluation of study data for monitoring of safety and study conduct. A data safety monitoring board (DSMB) would be needed to provide this independent review of MRT investigations. The DSMB could also play a role in reviewing prespecified stopping criteria for enrollment in and implementation of MRT clinical investigations—important for supporting the integrity of clinical investigations and the safety of research participants. Long-term follow-up would need to continue, however, even if stopping criteria were employed to prevent further enrollment in or implementation of MRT clinical investigations.

Circumscribed Use

Given the novelty of MRT, its possible intergenerational effects, and the fact that the persons most affected—future children—would lack a role in making the decision to proceed, FDA would need to restrict approval to studies involving women with mtDNA disease with a compelling clinical need. Future proposals to broaden the use of MRT for other indications (e.g., to treat idiopathic or age-related infertility) would need to be subject to fresh ethical analyses, including public discussion and debate through such mechanisms as those discussed elsewhere in this report.

If MRT were approved for clinical use in women with known pathogenic mtDNA mutations, FDA would need to use all of the tools at its disposal to control off-label uses of MRT beyond those indications and settings for which it had been tested and approved (see Chapter 3 for discussion of circumscribed use and issues regarding treatment versus enhancement applications of MRT). These tools could include mechanisms such as postapproval studies or a Risk Evaluation and Mitigation Strategy (REMS), as well as enhanced surveillance to detect adverse events (see Chapter 2).

Long-Term Follow-Up

Because the risks and benefits of MRT would make themselves known over time, FDA would need to require as a condition of approval that sponsors design, fund, and commit to long-term monitoring. In addition, FDA would need to emphasize the adverse event reporting obligations of sponsors and MRT providers, be committed to timely analysis of postmarket data, and take advantage of long-term data available from other countries.

This committee was not tasked with defining a specific period for long-term follow-up. Moreover, FDA would have to make such a determination on a case-by-case basis in close consultation with MRT researchers. However, the committee offers the following set of points to be considered by FDA in determining sufficient or optimal follow-up:

• Ability to identify any major medical consequences—There would have to be reasonably high confidence in ruling out procedure-related events that would be of major medical consequence to the child born as a result of MRT or would be transmissible to future generations. Such a standard would likely favor extending the period of monitoring to sexual maturity, so that gametic tissues could be studied. As used here, “major medical consequence” denotes medical events that would substantially compromise age-adjusted activities of daily living.
• Feasibility—Owing to the novel nature and implications of MRT, researchers would have to be expected to go to extraordinary lengths to fund and implement plans for follow-up. However, the committee also recognizes that long-term follow-up activities would be likely to present major logistical and budgetary challenges, and that it would not be desirable for the heavy demands of implementing the ideal follow-up protocol to forestall further innovative activity in this arena. Accordingly, it would be reasonable for review bodies to consider feasibility in establishing expectations for follow-up.
• Periods of less intensive monitoring—It might be reasonable for researchers to plan for intensive follow-up during the early years and perhaps into sexual maturity, and for less intensive follow-up to be allowed once most of the major concerns (e.g., birth defects, mtDNA disease, sterility) had been ruled out.

• Transparency: Regulatory authorities should maximize timely public sharing of information concerning the MRT activities and decisions within their jurisdiction. FDA should encourage sponsors to commit to depositing protocols and deidentified results in public databases.
• Public engagement: Regulatory authorities should incorporate ongoing exploration of the views of relevant stakeholders into the overall plan for review and possible marketing of MRT and should support opportunities for public meetings to gather these views.

• Partnership: FDA should collaborate with other regulatory authorities within and outside the United States to improve the quality of the data available for the assessment of benefits and risks.
• Maximizing data quality: FDA should require that sponsors have adequate resources, use appropriate designs, and plan studies that enable cross-referencing and pooling of data for assessments of benefits and risks.
• Circumscribed use: FDA should use the means at its disposal to limit the use of MRT to the indications, individuals, and settings for which it is approved, and should engage the public in a fresh ethical analysis of any decision to broaden the use of MRT.
• Long-term follow-up: FDA should require that sponsors design, fund, and commit to long-term monitoring of children born as a result of MRT, with a plan for periodic review of the long-term follow-up data.

REFERENCES

Aldoory, L., K. E. B. Ryan, and A. M. Rouhani. 2014. Best practices and new models of health literacy for informed consent: Review of the impact of informed consent regulations on health literate communications. http://iom.nationalacademies.org/Activities/PublicHealth/HealthLiteracy/~/media/Files/Activity%20Files/PublicHealth/HealthLiteracy/Commissioned-Papers/Informed_Consent_HealthLit.pdf (accessed September 27, 2015).

Appleby, J. B. 2015. The ethical challenges of the clinical introduction of mitochondrial replacement techniques. Medical, Health Care, and Philosophy 18(4):501-514.

ASRM (American Society for Reproductive Medicine). 2007. Financial compensation of oocyte donors. Fertility and Sterility 88(2):305-309.

Baylis, F. 2013. The ethics of creating children with three genetic parents. Reproductive Bio-Medicine Online 26(6):531-534.

Bonnicksen, A. 1998. Transplanting nuclei between human eggs: Implications for germ-line genetics. Politics and the Life Sciences 17(1):3-10.

Bredenoord, A. L., W. Dondorp, G. Pennings, and G. De Wert. 2010. Avoiding transgenerational risks of mitochondrial DNA disorders: a morally acceptable reason for sex selection? Human Reproduction 25(6):1354-1360.

CDC (U.S. Centers for Disease Control and Prevention), ASRM, and SART (Society for Assisted Reproductive Technology). 2015. 2013 Assisted reproductive technology fertility clinic success rates report. Atlanta, GA: CDC.

Craven, L., H. A. Tuppen, G. D. Greggains, S. J. Harbottle, J. L. Murphy, L. M. Cree, A. P. Murdoch, P. F. Chinnery, R. W. Taylor, R. N. Lightowlers, M. Herbert, and D. M. Turnbull. 2010. Pronuclear transfer in human embryos to prevent transmission of mitochondrial DNA disease. Nature 465(7294):82-85.

Darnovsky, M. 2015. Comments by Marcy Darnovsky, Ph.D., Executive Director, Center for Genetics and Society. Public Workshop of the Committee on Ethical and Social Policy Considerations of Novel Techniques for Prevention of Maternal Transmission of Mitochondrial DNA Diseases, March 31, Washington, DC.

Egli, D., A. E. Chen, G. Saphier, D. Powers, M. Alper, K. Katz, B. Berger, R. Goland, R. L. Leibel, D. A. Melton, and K. Eggan. 2011. Impracticality of egg donor recruitment in the absence of compensation. Cell Stem Cell 9(4):293-294.

FDA (U.S. Food and Drug Administration) Cellular Tissue and Gene Therapies Advisory Committee. 2014. Cellular, Tissue and Gene Therapies Advisory Committee Meeting Summary Minutes, February 25, Gaithersburg, MD.

FDA Center for Biologics Evaluation and Research. 2013. Guidance for industry: Preclinical assessment of investigational cellular and gene therapy products. http://www.fda.gov/BiologicsBloodVaccines/GuidanceComplianceRegulatoryInformation/Guidances/CellularandGeneTherapy/ucm376136.htm (accessed August 26, 2015).

FDA Center for Biologics Evaluation and Research. 2015. Guidance for industry: Considerations for the design of early-phase clinical trials of cellular and gene therapy products. http://www.fda.gov/downloads/BiologicsBloodVaccines/GuidanceComplianceRegulatoryInformation/Guidances/CellularandGeneTherapy/UCM359073.pdf (accessed January 8, 2016).

Fitzgerald, K. 2015. Ethical and social implications of MRT. Presentation to the March 31-April 1 Public Workshop of the Committee on Ethical and Social Policy Considerations of Novel Techniques for Prevention of Maternal Transmission of Mitochondrial DNA Diseases, Washington, DC. http://iom.nationalacademies.org/Activities/Research/MitoEthics/2015-MAR-31.aspx (accessed August 26, 2015).

Grant, R. W., and J. Sugarman. 2004. Ethics in human subjects research: Do incentives matter? The Journal of Medicine and Philosophy 29(6):717-738.

Haimes, E., L. Skene, A. J. Ballantyne, T. Caulfield, L. S. Goldstein, I. Hyun, J. Kimmelman, J. S. Robert, B. E. Roxland, C. T. Scott, J. H. Solbakk, J. Sugarman, P. L. Taylor, and G. Testa. 2013. ISSCR Committee Forum: Position statement on the provision and procurement of human eggs for stem cell research. Cell Stem Cell 12(3):285-291.

Handyside, A. H., E. H. Kontogianni, K. Hardy, and R. M. L. Winston. 1990. Pregnancies from biopsied human preimplantation embryos sexed by Y-specific DNA amplification. Nature 344(6268):768-770.

HFEA (Human Fertilisation and Embryology Authority). 2015. Regulating mitochondrial donation: Seeking expert views. London, UK: HFEA.

IOM (Institute of Medicine). 2008. The National Children’s Study research plan: A review. Washington, DC: The National Academies Press.

IOM. 2013. The California Institute for Regenerative Medicine: Science, governance, and the pursuit of cures. Washington, DC: The National Academies Press.

IOM. 2015. Science and Medicine Panel Discussion. Public Workshop of the Committee on Ethical and Social Policy Considerations of Novel Techniques for Prevention of Maternal Transmission of Mitochondrial DNA Diseases, March 31, Washington, DC.

IOM and NRC (National Research Council). 2007. Assessing the medical risks of human oocyte donation for stem cell research: Workshop report. Washington, DC: The National Academies Press.

Kalfoglou, A. L., T. Doksum, B. Bernhardt, G. Geller, L. LeRoy, D. J. H. Mathews, J. H. Evans, D. J. Doukas, N. Reame, J. Scott, and K. Hudson. 2005a. Opinions about new reproductive genetic technologies: Hopes and fears for our genetic future. Fertility and Sterility 83(6):1612-1621.

Kalfoglou, A. L., J. Scott, and K. Hudson. 2005b. PGD patients’ and providers’ attitudes to the use and regulation of preimplantation genetic diagnosis. Reproductive Biomedicine Online 11(4):486-496.

Liang, W. 2015. Mitochondrial manipulation technologies: Preclinical considerations. Presentation to the March 31-April 1 Public Workshop of the Committee on Ethical and Social Policy Considerations of Novel Techniques for Prevention of Maternal Transmission of Mitochondrial DNA Diseases, Washington, DC. http://iom.nationalacademies.org/Activities/Research/MitoEthics/2015-MAR-31.aspx (accessed August 26, 2015).

Lo, B., and L. Parham. 2009. Ethical issues in stem cell research. Endocrine Reviews 30(3): 204-213.

Lomax, G. P., Z. W. Hall, and B. Lo. 2007. Responsible oversight of human stem cell research. The California Institute for Regenerative Medicine’s Medical and Ethical Standards. PLoS Medicine 4(5):e114.

Magnus, D., and M. K. Cho. 2005. Issues in oocyte donation for stem cell research. Science. 308(5729):1747-1748. http://www.sciencemag.org/content/308/5729/1747.long (accessed December 6, 2015).

McGowan, M. L., C. Burant, R. Moran, and R. Farrell. 2009. Patient education and informed consent for preimplantation genetic diagnosis: Health literacy for genetics and assisted reproductive technology. Genetics in Medicine 11(9):640-645.

NIH (National Institutes of Health). 2009. National Institutes of Health guidelines for human stem cell research. Bethesda, MD: NIH.

NRC (National Research Council) and IOM. 2010. Final report of the National Academies’ Human Embryonic Stem Cell Research Advisory Committee and 2010 amendments to the National Academies’ guidelines for human embryonic stem cell research. Washington, DC: The National Academies Press.

Nuffield Council on Bioethics. 2012. Noveltechniques forthe preventionofmitochondrialDNA disorders: An ethical review. http://nuffieldbioethics.org/wp-content/uploads/2014/06/Novel_techniques_for_the_prevention_of_mitochondrial_DNA_disorders_compressed.pdf (accessed December 28, 2015).

Paull, D., V. Emmanuele, K. A. Weiss, N. Treff, L. Stewart, H. Hua, M. Zimmer, D. J. Kahler, R. S. Goland, S. A. Noggle, R. Prosser, M. Hirano, M. V. Sauer, and D. Egli. 2013. Nuclear genome transfer in human oocytes eliminates mitochondrial DNA variants. Nature 493(7434):632-637.

Presidential Commission for the Study of Bioethical Issues. 2013. Anticipate and communicate: Ethical management of incidental and secondary findings in the clinical, research, and direct-to-consumer contexts. http://bioethics.gov/sites/default/files/FINALAnticipateCommunicate_PCSBI_0.pdf (accessed December 28, 2015).

Tachibana, M., M. Sparman, H. Sritanaudomchai, H. Ma, L. Clepper, J. Woodward, Y. Li, C. Ramsey, O. Kolotushkina, and S. Mitalipov. 2009. Mitochondrial gene replacement in primate offspring and embryonic stem cells. Nature 461(7262):367-372.

Tachibana, M., P. Amato, M. Sparman, J. Woodward, D. M. Sanchis, H. Ma, N. M. Gutierrez, R. Tippner-Hedges, E. Kang, H.-S. Lee, C. Ramsey, K. Masterson, D. Battaglia, D. Lee, D. Wu, J. Jensen, P. Patton, S. Gokhale, R. Stouffer, and S. Mitalipov. 2013. Towards germline gene therapy of inherited mitochondrial diseases. Nature 493(7434):627-631.

Treff, N. R., J. Campos, X. Tao, B. Levy, K. M. Ferry, and R. T. Scott, Jr. 2012. Blastocyst preimplantation genetic diagnosis (PGD) of a mitochondrial DNA disorder. Fertility and Sterility 98(5):1236-1240.

Wang, T., H. Sha, D. Ji, Helen L. Zhang, D. Chen, Y. Cao, and J. Zhu. 2014. Polar body genome transfer for preventing the transmission of inherited mitochondrial diseases. Cell 157(7):1591-1604.

Zoloth, L. 2015. Ethical and social implications of MRT. Presentation to the March 31-April 1 Public Workshop of the Committee on Ethical and Social Policy Considerations of Novel Techniques for Prevention of Maternal Transmission of Mitochondrial DNA Diseases, Washington, DC. http://iom.nationalacademies.org/Activities/Research/MitoEthics/2015MAR-31.aspx (accessed August 26, 2015).