Full Summary of ASGCT's FDA Liaison Meeting

Six field-leading experts presented as part of the first-ever ASGCT liaison meeting with FDA on September 13, 2018. The meeting covered Society recommendations to FDA CBER on manufacturing, safety, and testing requirements for gene therapies, as well as an FDA update on the new RMAT designation.

Read the Full Meeting Agenda

Replication competent lentivirus (RCL) and replication competent retrovirus (RCR) testing of drug product

Kenneth Cornetta, M.D., Indiana University

Presentation Slides

In its recent draft guidance document, Testing of Retroviral Vector-Based Human Gene Therapy Products for Replication Competent Retrovirus During Product Manufacture and Patient Follow-up, FDA CBER noted the potential pathogenicity of replication competent retrovirus. Dr. Cornetta indicated that many of the early packaging cell lines developed replication competent retrovirus (RCR) through recombination between either homologous regions of vector and packaging sequences or endogenous retrovirus sequences in the packaging cells. Since that time, redesign to decrease homology has decreased the chance of recombination.

Dr. Cornetta provided substantial supporting data in his presentation, linked to above, for the assertion that in the commonly used production systems, RCR and replication competent lentivirus (RCL) has not been reported in the vector product, in the cell product, or in patients. ASGCT appreciates that the recent guidance provides for limiting RCR and RCL testing and agrees with different levels of testing for novel versus established vector systems. However, the Society recommends that for well-characterized systems, testing of transduced cell products does not add value and should no longer be a requirement.

There have been no adverse events from lentiviral vectors in clinical trials to date. ASGCT therefore recommends RCL testing post-infusion only in the case of an adverse event.

FDA representatives indicated that the agency appreciates data support for comments submitted to the docket in response to the draft guidances. Generally, draft guidance documents are intentionally vague to avoid being overly prescriptive, so identifying specific concerns and questions is helpful to FDA staff.

Testing Methods and Recommendations for Off-Target Analysis of Gene Editing  Technologies

J. Keith Joung, M.D., Ph.D., Massachusetts General Hospital

Presentation Slides

To address concerns about off-target effects of gene editing technologies, this presentation reviewed the challenges and methods for defining genome-editing nuclease off-target effects and provided ASGCT recommendations for off-target testing. Recommendations related to applications included assessing multiple nucleases early on for specificity and using these results to select nucleases for further work; using strategies to reduce off-targets (e.g., high-fidelity nuclease variants); considering both absolute on-target activity and on- vs. overall off-target activity for evaluating efficacy and safety; and performing well controlled, well characterized, state-of-the-art assays for IND and other FDA submissions.

A three-step strategy is recommended for identifying off-target mutations. At the discovery step, identifying potential sites of cleavage/mutation in a surrogate setting using multiple orthogonal approaches is recommended (because no gold standard exists); and cell-based assays are most appropriate for ex vivo assays if performed in the cells to be used therapeutically. Dr. Joung indicated that his personal opinion is that current computational approaches to date have utility as adjuncts to experimental approaches, rather than as a substitute for experimental approaches. An attendee expressed that bioinformatics play a role in in the design of site-specific nucleases.

For the risk assessment step, a recommendation is to assess all sites identified in discovery and discard nucleases that raise significant concern (especially those affecting tumor suppressor genes and oncogenes). In the final, confirmation step, the recommendation is to examine any sites found in discovery in nuclease-treated intended target cells or tissues and look for functional consequences in animal or cell-based assays at appropriate nuclease expression level/dose. It is important to recognize that the sensitivity of such assays will be influenced by the detection limits imposed by NGS.

Because ensuring that no off-target effects are occurring is not possible, the goal should be to minimize off-targets and to acknowledge that not all off-target mutations will necessarily be problematic. Future priorities include continued improvement and extension of experimental and computational off-target assays to improve sensitivity and predict functional consequences. Support for continued development of next-generation sequencing technologies will be critical. Development and characterization of off-target assays for cytidine and adenine base editors are priorities, as is accounting for human genetic variation and development of feasible, well-characterized, and sensitive functional assays for assessing tumorigenic risks in human cells.

Regenerative Medicine Advanced Therapy (RMAT) Designation

Wilson Bryan, M.D., Director of the Office of Tissues and Advanced Therapies, FDA CBER

Presentation Slides

The 21st Century Cures Act, Section 3033 created the Regenerative Medicine Advanced Therapy (RMAT) Designation to expedite the development and review of regenerative medicine advanced therapies. Dr. Bryan indicated that the overwhelming majority of gene therapies are eligible, since regenerative medicine advanced therapies include genetically modified cell therapies and gene therapies producing durable (vs. transient) effects. A drug is eligible for designation if it is a regenerative medicine therapy; the drug is intended to treat, modify, reverse, or cure a serious or life-threatening disease or condition; and preliminary clinical evidence indicates that the drug has the potential to address unmet medical needs for such disease or condition.

The benefits of RMAT designation, as is the case for breakthrough designation, are interactions with FDA to expedite development and review, including early discussions of potential surrogate or intermediate endpoints. Products that receive RMAT designation may be eligible for priority review and for accelerated approval, as agreed upon during product development, based on surrogate or intermediate clinical endpoints reasonably likely to predict long-term clinical benefit, or reliance upon data obtained from a meaningful number of sites, including through expansion to additional sites, as appropriate. RMAT designation differs from breakthrough designation in that breakthrough designation requires preliminary efficacy over currently existing treatments. Applying for both designations does not assist a sponsor.

After a request for the designation, FDA has 60 calendar days to determine if designation criteria are met. FDA will provide written response, and if not granted, FDA will provide a written description of the rationale. As of 9/12/18, 26 requests have been approved, 39 have been denied, and 8 are pending. The number of gene therapy requests was 14. Reasons for denial include administrative issues (i.e., an inactive IND), CMC reasons (i.e., for a different product); and insufficient preliminary clinical evidence (i.e., study design issues, such as a single arm study; inconsistent results among various important endpoints). Additional information on the RMAT designation is available on FDA.gov.

Dr. Bryan indicated that the number of gene therapy INDs increased by 30 percent from 2016 to 2017, with the trend continuing, so the number is anticipated to double in three years. This substantial increase is creating workload issues for FDA CBER.

Manufacturing Considerations

John T. Gray, Ph.D., Audentes Therapeutics

• Appropriate manufacturing controls for genetic/biological starting materials and process intermediates (i.e., plasmids, producer viruses)
• Suitability of certain cell lines for manufacturing specific vector classes and recommendations to address risk

Presentation Slides

Dr. Gray discussed concerns and recommendations related to the recently updated draft guidance, Chemistry, Manufacturing, and Control (CMC) Information for Human Gene Therapy Investigational New Drug Applications (INDs). Terminology used to describe materials used during production of the final product has a great impact on the regulatory burden associated with generation and use of those materials. Specifically, referring to plasmids used for transfection to make vectors as process intermediates in the guidance could encumber plasmid manufacturers with process validation requirements that would be burdensome. Improving specifications and traceability for raw materials are minimizing or eliminating the risks of adventitious infectious agents, and testing the activity of plasmids produced from bacterial master cell banks in the vector production process should capture deviations of additional unknown quality attributes. Process segregation measures should control the problem of generating material that is cross- contaminated with other plasmids. In the guidance (line 1042-1044), ASGCT recommends stipulating that if plasmids do not directly become part of the drug substance or drug product, they may be defined as raw materials or reagents, and their quality ensured by supply chain control.

Similarly, viral vectors used for ex vivo modification of cells are referred to as drug substance in the guidance, which encumbers production facilities with additional regulatory requirements and differs from EMA classification of such vectors as starting material. ASGCT recommends stipulating that viral vectors manufactured for ex vivo modification of cells may, with proper supply chain control, be defined as raw materials.

Finally, the World Health Organization’s standard of 10 ng host cell DNA/dose is not supported by experimental data quantitating the oncogenic risk associated with contaminating host cell DNA in rAAV preparations, and may be a gross overestimation of the true risk. Dr. Gray provided data to support that even the administration of a large number of potentially oncogenic materials did not induce tumors. While oncogenes are being transmitted, they do not have an effect in patients. Promoters in vectors may drive oncogenesis, since in mouse studies, oncogenes in the liver were inconsequential in the absence of strong promoter but induced tumors in the presence of a strong promoter.

Transfer of cellular proto-oncogenes via rAAV preparations are not likely to be a significant oncogenic risk in immune-competent individuals, due to the rarity of the events, the need for high copy transmission or activating mutations, and the likelihood that tumor suppressive processes will counteract progression (relevant to all production cell types). In the context of HEK293 cell production, the rare transfer of single copies of adenoviral serotype 5 E1 genes are also likely to be inconsequential for the above reasons.

The Society recommends that regulators require sponsors to document levels of contaminating host cell DNA and strongly transforming oncogene DNA in products, as opposed to complying with an arbitrary set limit.

Data From Long-Term Follow-Up For Persistent vs. Transient Gene Therapy Product Classes on Critical Safety Parameters

Helen Heslop, M.D., Baylor College of Medicine

Presentation Slides

Dr. Heslop reported on the safety of gene therapy clinical trials using lentiviral vectors into HSCs for a number of diseases, with no observed vector-related SAEs and no RCL reported in product or patients. Integration site analyses do not show preferential integration near oncogenes, nor significant clonal expansion. Over 1,000 patients in the literature received immune effector cells modified with retroviral or lentiviral vectors with no reported insertional mutagenesis and no RCR. She reported that the challenges of LTFU monitoring include requirements for a detailed record of  exposures to mutagenic agents and other medicinal products, which may be multiple, and reporting of delayed adverse events, including unexpected illness and hospitalization, although many adverse events are potentially related to subsequent therapies. She indicated that clarity is needed on the degree of reporting for expected side effects. Dr. Heslop recommended that inclusion in a registry should be an acceptable alternative for gathering and reporting LTFU data.

DNA Vectors

Ronald Crystal, M.D., Weill Cornell Medical College

Presentation Slides

Dr. Crystal indicated the existence of extensive experimental animal and human safety data relating to in vivo administration of DNA vectors, and that toxicology-related studies are often repetitive, providing little additional information. He recommended that FDA publish periodic guidance documents that state guidelines that could eliminate some of the repetitive toxicology requirements. Because of a 25-year experience with in vivo administration of DNA vectors, the 15-year reporting requirement for many gene therapy applications is no longer relevant to many in vivo DNA vector applications. ASGCT therefore supports requiring LTFU reporting that is vector- and disease-specific, as is proposed in the July 2018 draft guidance document, Long-Term Follow-up After Administration of Human Gene Therapy Products.

Dr. Bryan indicated that FDA enforces the circumstances under which data must be gathered, not how it is gathered, so use of a registry is allowed for LTFU data collection and reporting. An attendee recommended that FDA state in the guidance that it allows use of a registry for LTFU, which may not be well known in the field. FDA staff indicated that ASGCT may request a joint symposium with FDA at its Annual Meeting, as Dr. Crystal recommended, related to issues such as communication of safety/toxicology information and use of natural history studies in rare disease trials.

Related Articles