When Doctor and Scientist Collide: The Ethical Fault Lines of Pediatric Cell and Gene Therapy Trials

When a child with a devastating genetic disease becomes the first human ever to receive an experimental cell or gene therapy, the physician at the bedside and the scientist running the trial are usually the same person. That dual identity sits at the heart of a new review published in Frontiers in Genetics, in which Stanislav Birko and Ma’n H. Zawati of McGill University’s Centre of Genomics and Policy map the ethical terrain that researcher-clinicians must navigate when first-in-human trials move from laboratory animals into the bodies of minors. Their conclusion is uncomfortable but clarifying: the two central concepts bioethics uses to police that boundary, the therapeutic misconception and clinical equipoise, rest on logically incompatible pictures of what clinical research actually is, and no amount of careful consent paperwork can make the contradiction disappear.

The review, a hermeneutic narrative analysis rather than a systematic census of the field, set out to find every English-language paper addressing the professional duties of researcher-clinicians in trials that were simultaneously pediatric, early-phase, and cell or gene therapy focused. The search was deliberately iterative: rather than relying on keyword database queries, which returned vast numbers of irrelevant hits, the authors tracked references and citations outward from a small seed set of papers they already knew, supplementing the process with alerts from researchers active in the space. Sixteen papers ultimately satisfied all three inclusion criteria, and the scarcity itself is a finding. Only one paper, a 2023 analysis by Bateman-House and colleagues of somatic gene therapy research in pediatric populations, addressed the exact intersection of all three contexts. The remaining fifteen circled the target, covering pediatric phase I oncology trials, first-in-human cell therapy trials in adults with Parkinson’s disease, and the general ethics of pediatric research, without ever quite landing on the combination that defines modern first-in-human pediatric gene therapy.

What the literature does say, however, converges on a set of concrete responsibilities. In the pediatric context, informed consent transforms into parental permission plus, where developmentally appropriate, the child’s assent. Roth-Cline and Nelson, whose work anchors much of the review, insist that parental permission must be held to the same standard as adult informed consent, covering potential risks, benefits, and alternatives, and that a child’s assent must be an affirmative agreement rather than a mere failure to object. They describe assent as evolving along a developmental arc, from a choice largely dependent on the parent’s decision in young children, through joint decision-making as children mature, to a largely independent decision by older adolescents with parental affirmation, and finally to fully independent consent when the young person is deemed capable of providing it.

The review also catalogues who must be involved and what must be communicated. Benedetti and Marron describe a triadic relationship among the child-patient, the parents, and the physician, who acts as fiduciary for the child. Roth-Cline and Nelson push back on framing child and parents as fully independent parties, arguing that assent only makes sense within the protection afforded by parental permission. On the question of how many parents must consent, the answer scales with risk: one guardian suffices for minimal-risk research and for greater-than-minimal-risk research offering the prospect of direct benefit, but both parents must consent when greater-than-minimal-risk research offers no prospect of direct benefit. As for the age at which assent becomes required, the review notes Diekema’s suggestion of seven years as the threshold after which most children are developmentally capable of assenting, while other scholars caution that older age does not reliably correlate with greater decision-making capacity.

The content of disclosure receives equally granular attention. De Vries and colleagues enumerate the concepts parents most often fail to grasp: the distinction between research and therapeutic intent, randomization, voluntariness, the risk-benefit ratio, the procedures themselves, the duration of participation, and the right to withdraw. Their recommendations for communication read like a checklist for anyone designing a pediatric trial consent process: clarify how the physician-investigator and patient-subject relationship differs from the traditional physician-patient relationship, mention alternatives to research participation explicitly, discuss clinical and parental equipoise, indicate which interventions are performed solely to measure trial outcomes, assure freedom to withdraw, and obtain research consent on a separate form from treatment consent, ideally administered by someone other than the treating physician, such as a research nurse. Boydell and colleagues add that having an arms-length research coordinator obtain consent is an effective means of minimizing coercion or undue influence.

Vulnerability is a recurring theme. Woods and colleagues warn that consent obtained under stress may be too fragile to withstand the ethical work it is required to do, and that the vulnerability deepens when the request to consent arrives close to the diagnosis of a serious condition. Some parents report feeling pressure to enroll their children in phase I studies so as not to be seen as giving up on them. Empirical studies cited by Roth-Cline and Nelson found that parents who reported lower voluntariness also perceived more external influence and time pressure, worried that declining would harm their child’s care, and felt they had received either too much or too little information. De Vries and colleagues emphasize parents’ fear of jeopardizing the relationship with the physician by refusing. The researcher’s job, on this account, is not merely to hand over a document but to assess understanding, explore possible misconceptions, allow time for deliberation, and actively palliate the power imbalances that saturate the encounter.

The phase I context sharpens these concerns into a philosophical dispute. Under United States regulations, whether a pediatric trial offers the prospect of direct benefit determines the level of risk children may be exposed to, the consent mechanisms required, and whether local or national review is needed. Ross, drawing on Nancy King’s tripartite distinction between direct benefit arising from receiving the intervention, collateral benefit arising merely from being a subject, and aspirational benefit accruing to society, argues that phase I trials do not offer the prospect of direct benefit, and that classifying them as if they do ignores the moral relevance of the researchers’ intent. Hester and Ross reinforce the point: phase I trials are designed not to detect therapeutic benefit but to evaluate safety and toxicity, and if one could say with confidence what the chances of medical benefit were, the trial would not be needed. Yet dissent exists. The American Society of Clinical Oncology has characterized phase I cancer trials as a real therapeutic option for some patients, and Kodish has defended a relativistic understanding of the prospect of benefit. Ross counters that most subjects enroll hoping for personal benefit despite knowing the study is designed to determine toxicity, and cites Horng and colleagues’ finding that fifty-one percent of phase I consent forms alluded to the possibility of benefit outside the designated benefit section.

Cell and gene therapies add wrinkles that conventional drug trials do not present. Hug and Johansson note that participants in first-in-human cell therapy trials cannot withdraw in the sense of reversing the treatment, since injecting cells or creating lesions in the brain is irreversible. Bateman-House and colleagues make the same point for gene therapy: once edited genes are introduced, the effects cannot be undone or turned off, unlike a drug whose concentration falls once dosing stops. Heyder and colleagues stress the need to disclose the components and origins of the cell product, the preclinical data, the relative unpredictability of consequences, the irreversibility of cell proliferation in the body, and the risk of tumour formation and possibly death. They also flag a responsibility no one else in the sample addressed: respecting the stem cell donor’s right to decide how their cells may be used, even though the time, manner, and purpose of future use may be unknowable at the moment of donation. Eaton and colleagues, meanwhile, examine the ethics of stopping trials purely for financial reasons, arguing that researchers must ensure trials are responsibly started in the first place to minimize that risk.

The theoretical core of the review is its diagnosis of an incoherence that Miller and Brody identified two decades ago. The therapeutic misconception, defined by Henderson and colleagues as the failure to appreciate the difference between research and treatment, only makes sense if research and treatment really do differ, which is the difference position in the long-running similarity-versus-difference debate. Clinical equipoise, the requirement that it remain unknown whether the experimental or control treatment is therapeutically superior, only makes sense if investigators have a therapeutic obligation to participants, which is the similarity position. Endorsing both simultaneously, as the federal regulatory system and the Belmont Report arguably do, leaves the ethics of clinical trials in a state of incoherence. Empirical work complicates the picture further: Czoli and colleagues interviewed thirty pediatric researcher-clinicians across three major Canadian hospitals and found that none described their dual roles in a way that tightly reflected any existing theoretical framework, though many leaned toward a weaker version of the similarity position, seeing themselves primarily as physicians.

Birko and Zawati refuse to jettison either concept. Following Elizabeth Grosz and Rosi Braidotti, they treat the irreconcilability as a productive paradox, one worth lingering on rather than rushing to resolve. Researcher-clinicians who take their ethical duties seriously need not solve the similarity-versus-difference debate, but neither can they use its unresolved status as an excuse to disengage from it. The review closes with a warning aimed at the shifting legal landscape in North America, where Utah in 2024 and Nevada in 2023 passed laws permitting healthcare providers to perform stem cell therapies not approved by the United States Food and Drug Administration, provided patients receive written notice beforehand. Nothing in the ethical literature surveyed, the authors conclude, suggests that the greater uncertainties inherent in early-phase cell or gene therapy trials justify any relaxation of informed consent standards. On the contrary, the scholars engaged in these debates agree that increased uncertainty demands stricter adherence to procedures ensuring that consent, especially when it is a parent consenting on behalf of a child, is truly informed.

Subject of Research: Ethical responsibilities of researcher-clinicians in pediatric first-in-human cell and gene therapy clinical trials

Article Title: More than one and less than many: research and care in pediatric first-in-human cell and gene therapy clinical trials

Article References: Birko, S., & Zawati, M. H. (2026). More than one and less than many: research and care in pediatric first-in-human cell and gene therapy clinical trials. Frontiers in Genetics, 17, Article 1952672. https://doi.org/10.3389/fgene.2026.1952672

Image Credits: AI Generated

DOI: 10.3389/fgene.2026.1952672

Keywords: pediatric ethics, gene therapy, cell therapy, first-in-human trials, therapeutic misconception, clinical equipoise, informed consent, research ethics, phase I trials, dual-role clinicians, stem cell regulation, right-to-try laws