Novel Clinical Trial Designs for Gene Therapies: An Exploration of Challenges by the National Academies’ Forum on Regenerative Medicine

November 22, 2019By Sarah Wicks & James E. Valentine & Frank J. Sasinowski

On November 13, 2019, the National Academies of Sciences, Engineering, and Medicine (NASEM) Forum on Regenerative Medicine hosted a workshop on Exploring Novel Clinical Trial Designs for Gene-Based Therapies. The Forum was tasked with exploring the design complexities and ethical issues associated with gene- and gene-editing-based therapies such as optimal dosage, delivering the product effectively and successfully recruiting patients to what may be “single chance” trials. Co-chaired by FDA’s Celia Witten, Deputy Director of the Center for Biologics Evaluation and Research (CBER), with a broad array of stakeholder participants, the workshop united the voices and perspectives of academic and industry researchers, regulatory officials, clinicians, bioethicists, and patients and patient advocacy groups. This blog post summarizes some of the key challenges identified during the workshop related to designing and conducting clinical trials for gene therapies, and some of the emerging practices shared to help overcome these challenges as we work together to bring more promising therapies to patients.

Challenge 1: Collecting High-Quality Natural History Data

Recently, there has been a great deal of emphasis on the importance of natural history studies, particularly for rare genetic diseases where little is understood about the progression of disease and response to available treatment (see previous coverage of FDA’s Rare Disease Natural History Studies Guidance here). Natural history studies are not only a tool to describe the disease (e.g., to inform development of endpoints) but can also be valuable as external controls (i.e., historical control and patient-as-their-own control). However, the rare or ultra-rare nature of many genetic diseases often makes it difficult to obtain robust natural history data.

But some natural history data is better than none, as one sponsor for a gene therapy for retinal diseases described. Even with a retrospective chart review of approximately 70 individuals, the sponsor was able to describe progression of disease. Another gene therapy sponsor for spinal muscular atrophy described that robustness can come from a mix of natural history data, such as prospective data, retrospective data, and lead-in cohorts. Dr. Witten summarized that natural history datasets can be made more robust with frequent visits, standardized measures, and an effort to collect high-quality patient-level data.  She also noted the importance of identifying the genetic diagnosis in patients, which could help with understanding genotype-phenotype relationships, in order to select clinical trial populations.

Challenge 2: Approaching Patient Concerns through Community, Informed Consent, and Partnership

We know that there is a high tolerance for risk in rare, progressive diseases with no approved therapies. However, patient advocates from Parent Project Muscular Dystrophy and the Friederichs’s Ataxia Research Alliance, among others, explained that there are unique patient concerns when considering whether to participate in a gene therapy trial:

  • Because patients receive only a single dose, how do we know the first dose is therapeutic?
  • Is it worth it to participate given that participation will exclude you from clinical trials for other investigational therapies?
  • Will affected siblings be able to gain access?

Given these unique concerns, how do we set expectations and discuss this with patients, particularly with pediatric populations?

While this is no easy question, Courtney Fitzhugh from the National Heart, Lung, and Blood Institute (NHLBI) at NIH, who is investigating a gene therapy for Sickle Cell Disease (SCD), described 3 factors that she found to influence patients when considering joining a clinical trial: family, faith and other patients. By conducting a survey, Dr. Fitzhugh learned that whether a patient has family that are providing moral support, a strong spiritual belief and community, and has talked with other patients who have participated in clinical trials greatly influences a patient’s decision to enroll in a study. John Tisdale, who is also working at NHLBI with Dr. Fitzhugh on SCD, discussed how sponsors of gene therapies know that the first dose is therapeutic. He described the typical trajectory of first studying investigational therapies in cell culture, then progressing to small and large animal models before ultimately moving into first-in-human trials. Dr. Tisdale emphasized the importance of selecting large animal models that are biologically relevant so that therapeutic doses observed in the large animal model are more predictive of therapeutic dose in the first trial patients. Dr. Witten summarized that patients and families should be partners in the R&D process, and that there is an opportunity to improve the informed consent process for gene therapy trials to help answer these unique patient questions and concerns.

Challenge 3: Inclusion of Pediatric Populations in Gene Therapy Trials

For many genetic diseases, the earlier intervention is received, the better the eventual outcome, which increases the interest of including pediatric populations from the very outset of intervention development. Newborn screening is an important tool to help identify infants with these conditions, and when done at a population-level, it promotes fair access to cutting-edge technologies in clinical trials.

But to include children in interventional trials, when there is more than minimal risk, there must be a prospect of direct clinical benefit and the risk-benefit must be at least as favorable over other available therapies, which typically are nonexistent. Additionally, sponsors need to be specific across the type of pediatric population they wish to study (i.e., newborn, infant, child, adolescent) because there may be unique differences between these subgroups, and compared to adults, that leads to different trial designs and outcome measures.

Challenge 4: Developing Efficacy Endpoints for Gene Therapy Trials

For rare genetic disorders for which there are no other treatments available, a novel endpoint usually must be established. Larissa Lapteva, Associate Director of the Division of Clinical Evaluation, Pharmacology, and Toxicology (DCEPT) within CBER’s Office of Tissues and Advanced Therapies (OTAT), described three points to consider when developing efficacy endpoints for gene therapy trials:

  1. Consider the long-term or potentially irreversible effects of gene-therapy treatment; Dr. Lapteva noted that there is little room for uncertainty about endpoint performance and sponsors are therefore required to increase vigilance in validity and accuracy of endpoint measurement.
  1. Mechanistically agnostic endpoints that are reflective of common pathogenetic pathways may not be sufficiently sensitive in gene-therapy trials;

Sponsors need to consider that the increased availability of genetic screening, early diagnosis, and advanced lab testing has shifted the demand toward surrogate and clinical endpoints that are reflective of early disease manifestations. Additionally, the identification of genetic defects associated with non-well-characterized phenotypes has increased the need for novel clinical endpoints.

  1. Opportunity to identify and validate surrogate endpoints along the pathway of gene transcription, transgene protein synthesis and level, functional activity and clearance.

Following Dr. Lapteva’s remarks, an investigator studying a gene-therapy for SCD described how pain as a primary outcome is overly subjective, making it difficult to identify the unique cause of the pain and to differentiate chronic versus acute pain. She also noted that biologic endpoints in SCD, while each with limitations, are based primarily on predictive value and are only associated with disease severity, not disease modifiers. Meanwhile, the sponsor of a now-approved gene therapy for an inherited retinal dystrophy provided a case study detailing how the lack of a clinically meaningful endpoint for rod vision led the sponsor to develop a novel functional vision endpoint that was the primary basis for approval (see our previous coverage of this gene therapy’s development and endpoint here).

Challenge 5: Considerations for Long-Term Follow-Up (LTFU)

Tejashri Purohit-Sheth, Director of DCEPT, described the importance of LTFU and considerations for sponsors to manage such follow-up. LTFU is needed because gene-therapy products are designed to achieve prolonged or permanent therapeutic effects and such long-term exposure may result in undesirable or unpredictable adverse outcomes that may occur past the period of monitoring. Sponsors should take into account multiple factors when considering their risk-assessment, including product characteristics, the target cell/tissue/organ, and preclinical and clinical information. Characteristics that increase risk include integration activity of the gene-therapy product, genome editing activity, prolonged expression of the transgene, potential for latency, and establishment of persistent infections. For more information on LTFU, we direct you to FDA’s Draft Guidance on LFTU After Administration of Human Gene Therapy Products here.

Dr. Witten summarized that patient registries, mobile health applications, and other remote tools may help with the collection of patient-reported outcomes in the subsequent 10 years of post-marketing follow-up when subjects should be contacted a minimum of once a year.

In addition to those provided in-text, our readers can find some of our additional coverage of gene therapy regulation:

  • Gene Therapy and Orphan Drug “Sameness” here
  • FDA’s Comprehensive Policy Framework for Regenerative Medicines here
  • FDA “State of Cell and Gene Therapy Statement” here
  • Historic Approval of First Systemically-Administered Gene Therapy here