Who Goes First?Deaf People and CRISPR Germline Editing
The development of CRISPR technology has catapulted the issue of germline editing to the forefront of a debate between the goals of medical advancement and promotion of human diversity. The US National Academy of Sciences and the National Academy of Medicine recommended in a joint report that germline editing should be tightly regulated and pursued only for "serious diseases." A follow-up statement from an international summit on human genome editing emphasized a more general point that "the risks [are] too great to permit clinical trials of germline editing at this time." Here we review their recommendations in the context of genetic deafness, a condition that historically has been viewed by the medical community as a pathology. Deafness does not meet the standard of "serious disease" for experiments with human germline editing, but there is a real concern that scientists may soon begin to do germline editing with deaf individuals because, as we will discuss, they are in many ways ideal subjects for a clinical study of CRISPR, though their condition is neither fatal nor debilitating. In light of this, we worry about the potential for medical overreach and expediency. Drawing from examples of living deaf communities around the world, we propose an expansive view of human diversity that recognizes the value of genetic, linguistic, and cultural diversity to the future health of humankind. [End Page 54]
Two years ago, the US National Academy of Sciences and the National Academy of Medicine released a report drafted by an international committee regarding the use of gene editing in humans (NAS 2017). Once a tedious and expensive process, gene editing has now become more accessible and cheaper using the new CRISPR technology, making the issue of its use more urgent and pressing. The committee cites general support for somatic nonheritable gene editing to correct for a serious disease already present in a human being, but more controversial is germline gene editing for heritable conditions. Performing gene editing that affects an individual's descendants is at present tightly controlled in the US, but around the world, scientists are pushing against the regulatory barrier. The committee includes a recommendation that germline gene editing, if it should ever be pursued in the future, should be limited to "serious diseases and conditions," which they define elsewhere in the report as those that are fatal or debilitating—though, as we shall argue, what is considered debilitating is open to debate. Their recommendations were echoed in a statement signed by the Organizing Committee of the Second International Summit on Human Genome Editing, that "we continue to believe that proceeding with any clinical use of germline editing remains irresponsible at this time" (NAS 2019, 132). The statement confirms what most biologists believe, that significant technical challenges remain in the use of CRISPR, aside from the fact that ethical and clinical standards for implementation have yet to be developed.
Genetic deafness encompasses different genetic conditions involving as many as 200 to 250 genes. The most frequent are those that are autosomal recessive and nonsyndromic, meaning that there are no obvious conditions other than deafness. These conditions are not fatal, but some may argue that the inability to hear is debilitating. Since CRISPR could eliminate at least some of the genes that confer deafness, it's crucial to debate openly and seriously about which types of conditions might qualify for germline gene editing, if in fact any qualify at all. In a review of the debate around gene editing, Coller (2019) cites members of the deaf community as "contending … that deafness is not a disability, but rather a manifestation of human diversity and that it is important to protect and preserve the deaf community's culture" (295). Shearer, Hildebrand, and Smith (2017), in an overview of hereditary hearing loss, include the term "Deaf culture" in their glossary, explaining that there are "members of the Deaf community in the US" who regard deafness as not "a pathology or disease to be treated or cured" (2). Their mention of cultural views of deaf people in the medical literature is novel, considering that earlier (or even current) texts describe deafness, and broadly hearing loss, as a pathology. The widespread acceptance and use of cochlear implants in young deaf children show that the prevailing medical and public view of deafness is that it should be repaired as early as possible, and eventually eliminated. [End Page 55]
For the authors of this article, the topic of gene editing is close to home professionally and personally. We live with genetic deafness, and the languages of our home are American Sign Language (ASL) and English. The first author, Carol Padden, is herself deaf, as are her parents and grandparents. ASL has been used in her home for at least three generations. Carol's daughter, the second author, Jacqueline Humphries, is hearing of two deaf parents (and grandparents) and is bilingual in ASL and English. Carol is a linguist by training, having received her PhD in Linguistics, and has published on sign language grammars and cultural practices in Deaf communities. She is professor and dean of social sciences at the University of California, San Diego. Jacqueline recently completed her PhD in molecular biology and works as a research scientist at a biotechnology company. In our family, we often discuss the scientific and ethical case for gene editing, as well as the cultural and social case for diversity. Fundamental to our discussions about gene editing is the view that genetic diversity and cultural diversity are constitutive of the other. Doing good science and medicine means understanding that humans are constituted by their cultural and social practices; culture is not a byproduct of human evolution, but phylogenetically, culture coemerged with the human cognition and language.
In our multigenerational experience with deafness, we share a version of cultural diversity that is aligned to the ways that genetic diversity has been described in the literature. Coller (2019) repeats an argument familiar to geneticists that it has been "vital for humans to have mutable DNA so that our species can improve its fitness to survive in the current environment or adapt to changing environments" (293). He describes mutations as not "errors … but rather an intrinsic and important aspect of our species' evolutionary success" (293).
In the medical literature, hearing loss is described as the most frequent sensory impairment appearing in newborns (Atik et al. 2015). In Europe and North America, most cases of deafness result from genetic inheritance. In other parts of the world, hearing loss more often results from acquired illnesses, but genetic causes account for many cases of hearing loss across all societies (Shearer, Hildebrand, and Smith 2017). One relatively common genetic cause of deafness involves mutations of the gene GJB2, which codes for the gap junction protein connexin 26. These mutations affect passage of ions through gap junctions and interfere with the transmission of auditory signals within the cochlea, resulting in hearing loss (Kemperman, Hoefsloot, and Cremers 2002). While multiple mutations in connexin 26 can result in deafness, one of the most common variants is known as 35delG. This variant is most prevalent among individuals of European descent, or from the Middle East and Turkey, and it is also found, but to a lesser extent, in those of Asian descent (Gasparini et al. 2000; Mahdieh and Rabbani 2009; Tekin et al. 2003). A substantial proportion of individuals with autosomal recessive nonsyndromic hearing loss have mutations in connexin 26. Because it is recessive, an individual must inherit two copies of the mutation, one from each [End Page 56] parent. Their parents may be deaf themselves, or each hearing parent may be a carrier of just one copy of the mutated variant. Researchers have observed high carrier frequencies of up to 1 in 35 for the 35delG variant of connexin 26 in certain populations (Gasparini et al. 2000). This raises the question of why carriers of mutations in connexin 26 are so common and widespread in different parts of the world.
The answer may lie in a phenomenon called "heterozygote advantage," where having one wild type—unmutated—copy and one mutated copy of a gene can present an advantage in certain conditions. A widely discussed example is a resistance to malaria among carriers of sickle cell anemia (Luzzatto 2012). In the case of genetic deafness, Meyer and colleagues (2002) observed that carriers or homozygotes of a certain mutation in GJB2 had thicker epidermis than family members without a mutant allele. Furthermore, individuals who were homozygous for the mutant allele had increased ion concentrations in their sweat. These changes might influence susceptibility to pathogens, providing a selective advantage to this mutation. Variation in the human gene pool is valuable, and the choice to remove such variation must be balanced against the severity of the condition it is intended to prevent. At present, there is not enough research on connexin-related conditions and heterozygote advantage—ironically, because deafness is not considered to be so debilitating that it warrants sizeable investment in this kind of research. In general, we think any potential of a heterozygote advantage in a given human population should give scientists serious pause before they explore germline editing to remove or reduce that advantage.
As we stated earlier, the immediate challenge in germline editing of deafness is the large number of genes involved in hearing. Further, individuals can carry different mutations in the same gene. As Atik and colleagues (2015) note, "this overwhelming genetic and clinical heterogeneity makes the identification of genetic etiology challenging, time consuming, and expensive" (2). Moving beyond identification to targeted editing of particular mutations is a challenge, because any gene editing procedure must protect against the chance of off-target edits, where edits are introduced in unintended places in the genome. It is currently difficult to develop a process that can reliably edit a single allele with no off-target effects, much less the more than 1,000 mutations associated with hearing loss.
But the enormity of the challenge is likely to do little to dissuade medicine from its goal of eradicating deafness. There is an abundance of nonheritable gene therapy studies directed at the cochlea of mice carrying mutant alleles that cause hearing loss in humans, including connexin 26 (Zhang et al. 2018). We believe that clinical attempts to carry out germline editing of certain forms of genetic deafness is imminent.
Indeed, as we were writing this article, the New Scientist reported that a Russian biologist announced plans to proceed with germline editing on behalf of 5 deaf couples, all of whom have recessive deafness caused by mutations in the [End Page 57] GJB2 gene (LePage 2017). The article quoted a bioethicist, Julian Savulescu, as warning against such an attempt: "The first human trials should start with embryos or infants with nothing to lose, with fatal conditions … you should not be starting with an embryo which stands to lead a pretty normal life." While we cringe at any reference to a human infant "with nothing to lose," as the right justification for experimentation, more to the point is why is this Russian biologist choosing to start human trials with deaf couples?
Why might scientists act against the recommendation of learned professional societies and committees and proceed with clinical studies involving deaf individuals, if in fact connexin-26 recessive deafness is not fatal or debilitating? We think it is because deaf people, particularly those who enjoy the benefits of multimodal language and are active participants in self-defining communities and cultures, are ideal clinical subjects. Those who have autosomal nonsyndromic recessive deafness are likely to be healthy and free of other inheritable disease. They are also more likely to marry other deaf people who share the same facility for communication, and thus they may end up with spouses who have the same recessive mutations. In their offspring, only one copy of the gene would have to be edited to prevent deafness. Deaf individuals who reach adulthood are active members of their society, and in developed parts of the world, they are generally educated and can give informed consent regarding clinical trials. Furthermore, adult couples with recessive deafness of the GBJ2 gene are not at all rare—indeed, in many different parts of the world, there are good numbers of them for a substantial clinical trial.
As a comparison scenario, consider cystic fibrosis (CF), which is a serious condition. If germline editing were to become feasible, CF could be among those diseases to attract support for such an intervention, because many individuals with CF die before age 30. However, those with CF who reach adulthood are actively discouraged from marrying and having families, because of the high risk of intimately being exposed to bacteria that might be resistant to the armamentarium of antibiotics an individual with CF must take to treat pulmonary infections. Further, men with CF are commonly infertile, because the CF affects the vas deferens. Compared to deafness, there would not be a readily available population of homozygous couples for clinical studies, so editing for CF is logistically far more difficult.
Recommendations or exhortations to wait for further discovery from learned societies may not be enough to dissuade scientists who believe germline editing is a valuable tool for the future. As generations of medical professionals through history have decided, deaf children and adults are in need of medical intervention, ranging from sterilization to experimental surgery. Now, potentially, they will be subjects for a new generation of CRISPR studies.
Some humanities scholars describe the rush to repair and eradicate any and all disability, including deafness, as resulting from an "ableist logic" which affirms [End Page 58] "a network of beliefs, processes and practices that produces a particular kind of self and body … that is projected as the perfect, species-typical and therefore essential and fully human." In this view, disability "is cast as a diminished state of being human" (Campbell 2012, 213). It is hard to counter these beliefs not just in medicine, but in nearly every institution of modern life: education, the labor force, and in public policy. Indeed, as disability scholars argue, the very fact of modernity necessitates a belief in repair and augmentation of the human body toward an ideal of perfection, even enhanced ability. The ubiquity of these ideas has been the focus of the emerging field of critical disability studies, whose project has been to counter the prejudice that has "negative social consequences for disabled people," as well as to "disfigure the non-disabled" in their impossible desire to attain what cannot be achieved (Hughes, Goodley, and Davis 2012, 313).
The breadth of the different ways that prejudice finds its way into the lives of deaf people, particularly about their sign language, is difficult to contain and impossible to review here (Humphries et al. 2017). Instead, we offer two compelling examples of deafness and the history of deaf people as expressions of cultural diversity—not contracting but expanding the abilities and capacities of human beings. Our examples come from those practices that are familiar to us in our lives—that is, how we live among deaf people and work with hearing people.
Sign language and signers have likely always been a part of human society, or at the very least when language became possible in the species. Modern-day sign language research is traced to the middle of the last century, when a small group of linguists became interested in whether human language could be found in a modality other than speech (Klima and Bellugi 1979; Stokoe, Casterline, and Croneberg 1965). Most of this work began with the easily accessible sign languages in the US and in Europe (at least those sign languages near colleges and universities where this research was undertaken). There are large communities of signers in both regions of the world, and they descend from durable institutions tied to educational systems for deaf children. There is no reliable measure of the numbers of users of ASL, but a reasonable estimate is that there are at least 500,000 primary users in North America, and many more who have learned it as a second language and are fluent in it. National sign languages, or popularly used sign languages that are broadly used within countries, exist all around the world. (An inventory of these sign languages can be found in glottolog.org.)
Recently, sign language linguists have turned their attention to smaller and very small communities of signers where new sign languages have emerged in recent time, within the current or last two or three generations of users (Zeshan and de Vos 2012). Many of these communities involve the recent appearance of deafness as a result of intermarriage between individuals who are carriers of autosomal-dominant or autosomal-recessive deafness. The scientific value of such communities is that they offer the opportunity to observe how human language emerges in as few as one to two generations of signers. These emergent forms [End Page 59] allow scientists to explore how the foundational elements of language, from its smallest whole unit of meaning—words—are combined to form propositions and sentences, allowing complex discourse (Kastner et al. 2014; Lanesman and Meir 2012; Sandler et al. 2005). The new interest in emerging sign language communities is part of a broader trend in science to study the dimension of time, or emergent phenomena, as a way to understand underlying principles and properties.
A number of studies on small village sign languages have been carried out in communities around the world where there is genetic deafness, in the Middle East, Turkey, Bali, Ghana, Algeria, Mali, Mexico, and Thailand (Aronoff et al. 2008; de Vos 2015; Ergin et al. 2018; Hou 2016; Lanesman and Meir 2012; Nonaka 2009; Nyst 2000, 2015). In one of the more well-known cases found in southern Israel in a Bedouin community, the sign language used in the community is traced to a family with deaf siblings in the 1930s who shared a mutation in the connexin 26 gene (Meir et al. 2010; Scott et al. 1998). Today there are at least 150 deaf adults and children in a community of 6,000; in addition to deaf signers, there are at least three times as many hearing family members who also use the sign language with their parents, children, siblings, and other relatives throughout the community. Other communities that have been studied are smaller, and some consist only of immediate family members, but in all of these cases, the common observation is how readily humans living together in a community will adapt to using language in a modality other than speech.
This fact is not so remarkable in the view of psychologists and linguists who study the human use of gesture in everyday life (Goldin-Meadow 2005; Kendon 2000; Padden et al. 2015). Gesture is often thought of as an accompaniment to spoken language, but it is now understood as integral to human language, as a way to extend spoken language's audible limitations: to say that something is located "over there," is to require pointing, or in order to describe detailed shapes and physical properties, gesture is indispensable. Goldin-Meadow (2006) has argued that the presence of gesture is not only to communicate, but also to organize and direct thought during speech. The deep relationship between gesture and thought has led some to argue that the emergence of the symbolic human being was not by speech alone but by gestural abilities as well (Kendon 2000).
The spontaneous emergence of sign language in modern-day communities of hearing and deaf people demonstrates that humans are naturally multimodal. It is mistaken to say that because speech appears in every human community, that humans are exclusively unimodal for speech. Humans will participate in the creation of more complex gestures leading to a sign language over time, if the conditions for sustained use persist long enough for it to flourish. In this sense, sign language is not a marker of lesser ability, used only by those who cannot use speech, but an indicator of an ability in humans to adapt to multiple modalities for different purposes. [End Page 60]
The argument for multimodality as the prevailing characteristic of human beings need not be limited to the examples of sign language communities but extends to all of us in our technologically infused lives. It is startling to deaf people who have grown up using subtitles and captions to watch hearing people also using captions in public places and at home. Captions are now easily available, generated either by human transcribers or by automation, for videos across all media platforms. At first, closed captions were required by law in the US to enable accessibility to the public airwaves by the disabled, but now nearly 40 years after they first appeared on television for spoken English, they are used by hearing people as well. Americans turn on captions while watching British television, or while jogging on the treadmill. Giant screens in Times Square feature captions for televised entertainment. Video clips on a website are often captioned so that viewers do not need to listen to them while scrolling down a page; they can be quickly read in the short length of time it takes to scan the page. Texting and email have taken over a large part of the interaction that used to take place on a telephone. To be sure, there has been remarkable expansion of audible communication, from podcasts to readily available music in all media, but the rise of visual and textual communication has been just as explosive. It can be said that building technological systems for the disabled has resulted in expanded abilities for the abled.
As we once noted in a book about the modern deaf community in the United States, we live in times of enormous possibilities and prejudices (Padden and Humphries 2009). The rapid rise of ASL classes in the US, not only in universities, but in high schools and even middle and elementary schools, is nothing short of astonishing. Young parents today use sign language vocabulary with their prelingual infants as a route for early language before their speech becomes more intelligible.
Returning to the subject of this essay, if the ultimate goal of medicine is to heal and improve lives, we harbor no illusion that medicine will ever view deafness as compatible with these goals, at least not in the present time. Though we are "the public" who should be heard in matters such as gene editing, our views—that deafness is one of many variations of being human and should be valued not pathologized—clash with those of many medical professionals, and probably many geneticists. There are too many parents, doctors, teachers, and other professionals who don't view deafness the way we do; the public is more likely to include them than us. Coller (2019) acknowledges as much: "healthcare professionals tend to overestimate the impact of some disabilities on life satisfaction of children and their families" (295).
To make a concluding point, we return to the report we cited earlier by the US National Academy of Sciences and the National Academy of Medicine, emphasizing below in italics the relevant phrases: [End Page 61]
Recommendation: Clinical trials using heritable genome editing should be permitted only within a robust and effective regulatory framework that encompasses
• the absence of reasonable alternatives;
• restriction to preventing a serious disease or condition;
• restriction to editing genes that have been convincingly demonstrated to cause or to strongly predispose to that disease or condition;
• restriction to converting such genes to versions that are prevalent in the population and are known to be associated with ordinary health with little or no evidence of adverse effects;
• the availability of credible preclinical and/or clinical data or risks and potential health benefits of the procedures;
• ongoing, rigorous oversight during clinical trials of the effects of the procedure on the health and safety of the research participants;
• comprehensive plans for long-term, multigenerational follow-up that still respect personal autonomy;
• maximum transparency consistent with patient privacy;
• continued reassessment of both health and societal benefits and risks, with broad ongoing participation and input by the public. (NAS 2017, 134–35)
We stand in alliance with various organizations and scientists who work at the front lines of ethical use of scientific knowledge about gene editing. They include orthodox Jewish communities that have instituted selected genetic testing they believe is compatible with their cultural practices, though there is disagreement among them about which genetic conditions should be tested, and how to avoid stigmatization of carriers (Ekstein and Katzenstein 2001; Raz and Vizner 2008). Another debate revolves around consanguinity in communities around the world where marriage between close relatives is commonly practiced though there is risk in increasing the incidence of certain genetic conditions (Bittles 2002). As members of such communities have explained, the overriding cultural consideration of consanguinity is the maintenance of close ties between families and within related families. The counterbalancing force in the gene editing debate will have to be "the public," which must include those of us who live the condition, in cultures around the world.
Complicating matters is that for all these communities, from deaf people to orthodox Jews and couples who are closely related, there are individuals who will support gene editing. Some deaf people sincerely do want to hear better. Some orthodox Jews want more and broader genetic testing to avoid having children with very rare medical conditions. The authors of the NAS report offer a suitable alternative to germline editing for parents: adopting a leftover embryo from a fertility clinic that does not have the mutation (NAS 2017, 113). But, because many parents desire to conceive genetically related children, this readily available alternative is not given the consideration it deserves. [End Page 62]
As we conclude, we repeat our abiding argument: there is an overriding concern that the debate around ethics and who should first benefit from new medical breakthroughs may fall in favor of expediency and urgency of scientific discovery. Deaf individuals may find themselves first in line for germline editing because their autosomal nonsyndromic recessive condition offers more clinical opportunities than those with debilitating diseases. For this reason, and in order to do good science at all times, we urge caution and care to understand how to promote and not diminish genetic and cultural diversity, both of which offer our best possibilities for living well into the future.