Translational research—the need of a new bioethics approach

Translational science is a novel concept, whose main purpose is to categorize practical, outcome-oriented research [1]. In healthcare, translational research can be viewed as research on human specimens, whose findings may inform basic science research and lead to a transfer of the results towards clinical therapeutics and novel healthcare policies; its definition seems however to be an ever evolving phenomenon; if initially it referred to the bench-to-bedside enterprise of using information from basic sciences to produce new treatment alternatives for patients [2], nowadays it is defined by a process that starts with fundamental research (genes, molecular processes, biochemical pathways) and ends at a macro level (social healthcare, access to healthcare, access to education, and so on) [3]. Moreover, some authors considered that translational process should not start with fundamental research, as this approach rarely succeeds, but rather from clinical medicine—the so-called bedside to bench to bedside approach. A classical example to show the usefulness of this approach is represented by the development of balloon angioplasty by Gruentzig [4]. A more recent example is represented by T cell therapy—T cells have various roles in cell-mediated immunity, being able, among others to differentiate between healthy and abnormal (including neoplastic) cells, and is involved in HIV infection. This clinical knowledge has recently been brought to the “bench” by a research team lead by dr. Marson, who, by editing the genome of human T (CD4+) cells using a CRISPR/Cas9 ribonucleoprotein delivery method, caused up to 40 % of the cells to lose high level cell surface expression of CXCR4 [5], which could open new therapeutic possibilities in oncology, autoimmunity or infectious diseases.

For the purposes of this article we will try to define the translational process as containing three main elements: phases, gaps, and data transfer; therefore, when analysing ethical issues determined by translational research we will consider them as appertaining to either an element of the translational process (phase, gap, data transfer), or the process of translational research itself (see Fig. 1—modified after [3, 6, 7], and Table 1 for some examples of ethical issues depending on the element of the translational process).

Initially translational research was considered as containing two main phases: T1—bench-to-bedside, in which new discoveries from the laboratory could be translated towards clinical research (proof of concept, phase I and II clinical trials) and T2, in which these applications were translated in clinical practice (phase III clinical trials, studies of clinical efficacy and development of clinical guidelines). Recently the process was further developed by the addition of other stages; therefore, the most recent models contain either four or six translational phases [3, 6] and may or may not include the bedside-to-bench approach. For the purpose of this article, we will use the six-phase approach shown in Fig. 1. The six-phase approach considers that translational research starts with fundamental research and ends with developing and implementing social/economical policies determined by the results obtained in the process.

Between each adjacent set of phases is a gap, or roadblock, which refers to the lack of funding/support needed for the progression to the next [8]. The most important gap is between T0 and T1, called valley of death because most fundamental research studies do not go beyond it (therefore to not enter the clinical phases). According to Pienta these translational gaps are dependent upon the successful management of four main risks: scientific, intellectual property, regulatory, and market [8].

In the translation process, the transfer of information from one study to the next one characterizes data transfer [8]. These transfers occur either within phases or between phases (usually adjacent, but not always). The type of data transferred depends on the location of the transfer on the translational research process. For example the data that is transferred in within the first two phases is usually newly created data, which stays at the basis of the whole translational research process. The information created in this stage is structured in a way that can be easily transferred toward clinical research. For example, suppose that an investigator discovers a new heart receptor for angiotensin in cell cultures. This information, which is newly created, is transferred to another preclinical investigator, who might test its clinical consequences on animal models. The created data from this study is then transferred to a pathologist who will test the effect of the receptor on the pathological human heart. The data is then transferred to another researcher who might be involved in developing a drug that could be helpful for that specific cardiac pathology. All this transfer of newly created data happens in Phase 0. Next, the information could be transferred to a clinical investigator who will try a Phase 0 study on humans with the above-mentioned drug. Each data transfer is directed toward a certain group of individuals (that can be anything from a small group of researchers to a major population group). The size of the group that receives the transmitted data increases in size with the advancement of translational process. For example, the target group for transferring newly created knowledge (during fundamental research) is represented by investigators involved in basic and subsequently clinical research, whose number is usually very small. When going from Phase 2 to Phase 3 the knowledge will be transferred to clinicians, patients, epidemiologists, and so on, a much larger target group. Data transfer is usually done step by step in translational research. However, there are instances in which the process can skip one or more phases. For example, detecting a gene that is often associated with cancer might lead directly to dissemination of the information, or even public health initiatives, without going through T1 and T2 phases.

According to the Merriam-Webster dictionary, bioethics is defined as a discipline dealing with ethical implications of biological research and applications, especially in medicine. This definition is very broad, as it includes biological research (medical research, animal research, environment research), and medical practice. However, it is limited when trying to analyse recent developments in medical research, in which non-biological components are starting to have a bigger importance—information technology (IT) development, nanotechnologies, engineering, and so on (see Table 2 for an example). These developments lead to two important questions that must be answered before defining the scope of translational bioethics: (1) Should we analyse translational research ethics in direct relation to the steps/phases, or should we develop a more general set of bioethical guidelines guiding the entire translational research process?, and (2) Should we consider translational research as a subtype of biomedical research, or should we develop a new set of bioethical tools specifically directed to the often transdisciplinary characteristics of translational research?

The purpose of this article is to analyse whether a bioethics approach based on current, healthcare centred principles, is enough to allow a comprehensive model for the morality of translational research.

Most articles dealing with translational bioethics limit their analysis to one or few steps of the translational research process. For example Kimmelman [9] analysed the biases generated by risk assessment of clinical trials by the Ethics Review Boards. Shapiro and Layde analysed case scenarios in different phases of translational research, in order to identify the critical role of bioethics in each stage of clinical and translational research [10]. Bærøe considered that translational ethics includes two main discussion topics: (1) the ethics of translational biomedical research, in which we must analyse the various practical scenarios derived from the translational research, and (2) the internal gap between theory and practice within the field of ethics itself [11]. Hyun and Kimmelman made a phase oriented ethical analysis in translational gene transfer research [12]. The first topic is a typical step-by-step analysis of research ethics issues, while the second is a more general issue, recognized by some authors in the last years, related to the development of a rift between theoretical bioethics and its implementation in medical practice [13, 14]. The translational research process contains however not only phases but also gaps between phases, and transfer of knowledge in which the results are translated from one phase to another, translation that could also pose ethical issues. To illustrate this, we will present a situation generated by translating information from Phases 1 and 2 to Phase 5 from Table 3, in which is detailed the translational process involving in vitro fertilization with sperm procured from the deceased [15]. Recently a series of articles was published that supported the idea of using a presumed consent model for this technique meaning that, in the absence of a refusal from the deceased husband before death, the wife could use the posthumously harvested sperm for in vitro fertilization [16, 17]. Tremellen and Savulescu [16] used a predominantly theoretical approach to the argumentation in favour of presumed consent, while Hans [17] based his conclusion on a sociological study that supported this option. The translation presented by both articles (from clinical practice to social health policy) is associated with ethical issues. In order for a theoretical approach to be allowed for developing social policies, it should be based on a set of accepted general moral guidelines and an exhaustive analysis of the issue (including legal, social, economical and anthropological issues). An analysis not based on the above-mentioned issues could lead to wrong conclusions, which implemented in clinical practice might lead to clearly unethical actions. For example, Tremellen and Savulescu, arguing for a need of a presumed consent on the issue, stated that:

In our view, it can be argued that the welfare of the living is a far more important consideration than splitting hairs over degrees of consent and infringement of alleged autonomous rights of a deceased person. It is important to remember that the dead person no longer exists, so at that time deceased cannot have interests or be autonomous. Any ‘respect’ is related to creating policies that ensure that the living now are not harmed or fail to have their autonomous wishes respected—that is satisfied by an opt out system for posthumous conception. Put succinctly, if you don’t want it, sign out now. (page 9) [16].

Most Civil Codes, at least from Europe, consider that a person, even if it has died, can exert posthumously a series of rights, through testamentary dispositions. This means that the person retains some “autonomous rights”. The Declaration of Geneva (2006) and the World Medical Association (WMA) Code of Ethics (2006) both state that medical confidentiality must be maintained after the patient dies. Medical confidentiality is based, at least partially, on the right to autonomy (to autonomously choose how to share medical information about one self). Most countries consider that organ harvesting from deceased should be based on a previous consent of the donor, or at least on the consent of the family (the opt in variant for organ donation, present in most countries including Germany, Japan, Romania). There are however some counties, especially in Europe, which have a form of presumed consent (opt out) procedure for organ donation (including Austria, Croatia, Spain, Belgium). The type of procedure (opt-in or opt-out) seems to be a very important reason for organ donation, as revealed by the rates from Germany to Austria, two countries with highly similar cultures—in Germany, a country that preferred the opt-in method, the consent rate is around 12 % while in Austria, a country that preferred the opt-out method, the consent rate is almost 100 % [18‐21].

Therefore, we cannot really talk, at least in current medical practice, about the absence of autonomy after the death of the patient. In our opinion, using subjective argumentation, not based on current legislative and generally accepted practices in order to support the development of a health-care policy is unethical, as it may be considered a form of manipulation, exercised over stakeholders. This manipulation could lead to the creation of health-care policies that would have to be applied in clinical practice by physicians, who will be put in a position to respect either a fundamental code of ethics or deontology (again, at least in Europe, many countries have Physicians’ Codes of Ethics that rely heavily on the WMA Code of Ethics), or the national health care policies. By not respecting the healthcare policies, they could do something illegal, while by not respecting the deontological/ethical norms of the medical practice they would do something immoral. Hans used a sociological survey that showed an increased support for posthumous sperm procurement [17]. This approach again contradicts fundamental ethical principles in medical practice. The WMA Code of Ethics states: “A physician shall respect the right of the patients.” [22]. Imposing a certain procedure based on statistical analysis (what most of the population might want) contradicts the principle of autonomy in medicine, which basically states that a person should be able to make informed decisions about one-self. For example, most patients would agree to suffer a surgical intervention for peritonitis (so statistically there is an agreement about accepting the procedure if they were the patients needing that intervention); however, if they are conscious, they are still required to give their consent before entering into the surgery, and are allowed to refuse it. Again, the patient status of the deceased is debatable; however, as the medical codes of ethics recognize some autonomy related rights for deceased patients, it is assumed that they have at least, a form of autonomy, that they could very well exercise before dying. So, in order to respect the autonomy of the patient an opt-in alternative to medical procedures after death should be imposed, not an opt-out variant. The patient should be able to make informed decisions about what will happen to him/her after death; if this does not happen, unsubstantiated assumptions about his/hers agreements for various medical procedures should not be allowed, as they are legally and morally problematic (affecting his/hers self-determination).

A very hot topic in translational research, which presents perfectly the intricacies and limits of bioethics in translational research is represented by stem cell research. Tremendous progress is made every year, requiring a permanent redefinition of what is ethically permissible in stem cell research, and what topics remain actual in this field. For example in the 1990s, major issues in this area were focused on obtaining embryonic stem cell research from aborted fetuses or cloning for the creation of embryonic stem cells for transplantation [23‐28]. In the next decade the number of articles in this area exploded, a more than tenfold increase being identified on Web of Knowledge; the articles began to deal with more specific aspects of stem cells ethics, including patents, commercialization, creation of adult cell lines, creation of part animal stem cells, etc. [29‐37]. From 2011 to 2015 a shift is again seen, this time from embryonic stem cells ethics to adult or induced pluripotent stem cells (IPSC) ethics [38‐42]. When working with IPSCs researches do not have to analyse the morality of using embryos, and therefore debates around abortion, moral status of the embryo, commercialization of human embryos, teleology of the human embryo are not relevant. They do have however important ethical issues to deal with. For example IPSCs are known to form teratomas when injected in immunodeficient mice [43, 44]. Therefore, IPCSs may have the potential to generate human cancers when injected to treat various diseases. This could lead to ethical debates regarding the primordiality of beneficence (physicians have a moral duty to help their patients) versus non-maleficence (physicians have a moral obligation not to do harm to their patients). Uncertainty related to the long term benefits or risks is another ethical issue related to IPCS—should physicians recommend such a therapy taking into account that the long term benefits and risks are not fully quantifiable? [42]. Should IPCSs used as an alternate method to gamete donation in assisted reproductive technologies? [45].

As shown in Table 1, translational research has ethical issues related to any of its four main elements (phases, gaps, transfer of knowledge, the process itself). Even if all are potentially equally important, most researchers in this area focused solely on the phase related ethical issues [1, 2, 9‐12, 46‐54], which, in our opinion, is not enough for a comprehensive ethical analysis of translational research. However, by analysing all ethical related issues associated with all its elements, and taking into account the high variability of translational research, we could easily find ourselves in the position of considering that translational bioethics should deal with most, if not all, imaginable bioethical issues. This approach, even if exhaustive, does not grasp the particularities of the translational research bioethics. Therefore, our opinion is that we should try to analyse more specifically what makes translational research unique from other forms of biomedical research, and to draw a series of general ethical norms with a wide applicability in this field.

Most codes of research ethics require all biomedical research to be carried out under the supervision of a certified physician or other properly qualified health care professional. For example, the Helsinki Declaration, Art 12 states that:

Medical research involving human subjects must be conducted only by individuals with the appropriate ethics and scientific education, training and qualifications. Research on patients or healthy volunteers requires the supervision of a competent and appropriately qualified physician or other health care professional. (Article 12) [55].

However, there are instances of translational research in which most of the phases contain non-medical studies/activities. See Table 2 for such an example, in which the researchers are trying to develop an assistive device for visually impaired people by using haptic and audio signals; the translational process include 12 steps, of which only four can be considered as biomedical research. Most of them are either IT development or public health related (social policies, interactions with local administration, and so on). If we consider them to be biomedical research, they should comply to all legal and ethical requirements of biomedical research, which means, among others, obtaining consent from the subjects, obtaining an Ethical Committee approval for the study, supervision by medical personnel, and so on. Let us take an example from Table 2. Suppose that one step of this research is to develop a medical device constructed as a wearable head unit that enables the user to precisely determine the spatial characteristics of sounds. For this, some testers must walk around the lab and be subjected to various sound patterns. The responses of the testers are recorded. By relating this research to the principles from the Helsinki Declaration this activity fits within the purposes of biomedical research: “The primary purpose of medical research involving human subjects is to understand the causes, development and effects of diseases and improve preventive, diagnostic and therapeutic interventions (methods, procedures and treatments)” [55] because its purpose is to develop a therapeutic intervention for visually impaired people. However, the same activity can be considered a development phase for an IT product. How should the researchers consider this activity? If is it considered biomedical research, it must comply with all medical research regulations that would maximize the protection of the testers (physically, ethically and legally), but also increase the costs maybe tenfold. The precautions may be excessive and unnecessary as the risks are minimal and can be safely handled by other means; e.g., work protection, and the testers are competent to make decisions about the precautions needed for the study. If it is considered an IT development phase—instead of a biomedical research—the importance of the risks will be lowered because a risk assessment might not be performed before starting the tests. The issues related to protection and sharing of personal data (e.g., data obtained from the records of the responses of the subjects) might also be minimized. Our opinion is that the problem here is determined by the phase analysis of translational research. By using it we tend to reffer to biomedical/research ethics guidelines even in instances in which they are too strict, in order to apparently maximize the protection for subjects/patients/citizens. However, this apparent strictness affects the efficiency of the phases that are not actually biomedical research, by imposing time consuming and highly expensive tasks. A more holistic approach to bioethics in translational research might be of use for the whole process aimed toward minimizing the risks for subjects, other involved persons, and the society as a whole. This should be associated with a clear definition of what biomedical research is, and of course with imposing additional safeguards for activities that will comply with the definition of biomedical research. For the above example, we could do a risk assessment before the practical phases and develop methods for minimizing/removing the risks by a committee consisting of both investigators with healthcare specialization and non-medical personnel. Afterwards, the non-medical tasks will be handled according to the IT development principles, in full compliance with the risk mitigation methods established in the initial phase.

The ethics of translational research should go beyond the classical topic of research ethics, or medical ethics. It should not only analyse the ethical issues that can be directly derived from the translational phases but also those derived from the gaps between translational phases, transfer of knowledge and the particularities of translational research per se. There are many potential issues appertaining to the bioethics of translational research, of which only a few have been summarized in this article, and that should be reanalyzed within the translational research framework.