Establishing a Safe and Efficacious Human Starting Dose: A Hurdle for Cell and Gene Therapies
By Anika Schröter and Marko Repic
“How to define a human starting dose for cell and gene therapies: a key question for reaching clinical stage of development”
Definition of a human starting dose is a challenging step in every development process, representing a particular hurdle for gene and cell therapy products. In contrast to small molecules or biologics, there is no specific guidance, providing clear recommendations on how to determine a human starting dose. Individual, product-tailored and innovative approaches are needed, which require a thorough understanding of the indication, mode of action and potential risks related to the product. Read further to learn more about specific approaches and our experience with different cell and gene therapy products.
How to determine the dose of a locally applied gene therapy vector? Would the dose of CRISPR/Cas9 edited CD34+ cells tested in irradiated NSG mice be efficacious in treating a particular inborn hematological disorder?
Determination of a safe human starting dose is essential to the successful transition into the clinical phase of the pharmaceutical development. While safety is always paramount, it is important that a starting cell or gene therapy
dose should provide a reasonable chance to reach a therapeutic effect. These types of products are aiming to address a serious unmet medical need, and thus may be the patient’s only hope for improvement of the quality of life or even cure.
This is particularly true for products that can be applied only once, such as AAV gene therapy vectors, inducing a potent neutralizing antibody response, which usually prevents a second administration. Other products may require administration
using highly invasive interventions, e.g.
, CD34+ cell engraftment to preconditioned patients or surgical implantation of a regenerative medicine cellular product;therefore,administration-associated risks need to
be balanced with the likelihood of disease improvement. Further, therapies for ultra-rare conditions will only have a handful of eligible patients available. Any misestimation of the therapeutically relevant dose level for the first-in-human
study may diminish the chance of demonstrating efficacy and result in termination of the development program.
For many small molecules and biopharmaceuticals, the starting dose can be extrapolated from dose levels administered in animal studies by applying a scaling factor based on body surface area or body weight according to recommendations
given by FDA (2005). To arrive at a scientifically sound dose extrapolation strategy for a cell or gene therapeutic, converting the animal-to-human dose by body sizes generally not a viable approach. Peculiarities of the product class and
potential limitations of non-clinical testing need to be recognized and taken into account. This thinking is also reflected by the current cell and gene therapy guidelines, which do not provide an universal approach but encourage developers
to seek for individual solutions based on the specific nature of the product (FDA, 2015; EMA, 2018; EMA, 2019).
The relevance of animal disease and toxicology models to predict the behavior of a candidate cell or gene therapeutic in patients is often much more limited compared to small molecules or biologics. Also, suitable animal models might
not be available at all. For example, in case of engineered T cells cancer therapies, non-clinical work can demonstrate the proof-of-mechanism and provide reasonable confidence in the product specificity, but does not inform on a reasonable
human starting dose.
Frontrunners of the successful in vivo
gene therapy have been AAV vectors, achieving disease-modifying effects in patients with hemophilia, retinal dystrophy or spinal muscular atrophy (Verdera et al.,
2020). Transition from proof-of-concept animal studies to safe and efficacious human starting dose for a multitude of other diseases, amendable to systemic or localized gene therapy, invariantly requires understanding of the animal vs. human
differences in tissue tropism, transduction efficiency, transgene expression levels, turnover of the targeted cell population and vector and transgene immunogenicity, to name a few.
For cell therapies, the in vivo
proof-of-concept or safety testing is generally limited to immunocompromised rodents. While these models often support cell engraftment and sometimes even partial integration
of human cells into the mouse physiological processes, these approaches suffer from the risk of creating artificial data with a potentially limited correlate to the human situation. Clinical experience with a particular cell type, e.g.
CD34+ cell transplantations, can therefore provide a more relevant guidance on the human dose selection and the likelihood of engraftment. In principle, engraftment is a function of proper conditioning in combination with a theoretically as
high number of cells as possible. Consequently, the upper limit for the cell dose might be defined by technical limitations and / or toxicities potentially induced by a cryo-preservative. Parameters such as the percentage of long-term repopulating
(genetically modified) cells, expression of a given target gene and, with that, determination of target protein levels can be used in support of a potential clinical effect. This information provides guidance for further product development
and studies. For cell-based regenerative medicine products applied locally, scaling based on size and anatomy of the tissue being treated, e.g.
, bone, cornea, central nervous system, may guide the human starting
The following examples illustrate the variety of potential approaches which were used for previous programs and successfully accepted by regulatory authorities for different cell and gene therapies.
Strategy for Determination of Starting Dose
Genetically modified CD34+ hematopoietic and progenitor cells
Follow clinical practice for hematopoietic stem cell transplantations; “the-more-the-better”
Autologous human cells seeded in a biopolymer matrix
Cell density (cells per biopolymer volume unit) equal in constructs for human use and animal study
Local administration of a viral vector into a defined anatomical compartment (e.g.,brain region, joint)
Dose extrapolation to include a comparative structural and volumetric analysis of the anatomical compartment to be treated
Adeno-associated vector with novel capsid for in vivo gene therapy
Dose guided by transduction efficiencies in multiple preclinical models to characterize species-specific differences in tropism and transduction rates
Allogeneic stem cells for administration in the eye
Scaling based on eye corneal surface area and by comparison to dose of approved cellular product using the same route
Autologous muscle derived cells for injection into rhabdosphincter
Dose extrapolation based on the comparison of target muscle size in the animal proof-of-concept model and human. Clinical experience with other muscle cell-based therapies further taken into consideration.
Allogeneic human heterologous liver cells
Adjusted to the relative liver cell count in animal model vs. human
Due to the complexity of the process of dose translation and with that, determination of the human starting dose for cell and gene therapeutics, it is important to work on a scientifically soundfirst-in-human dose finding strategy early
on, including the selection and definition of appropriate efficacy and safety models, if at all available. This effort should comprise considerations regarding the design of safety studies, which again, usually do not follow the paradigm the
industry has been used to for small molecules and biologics. While it is also important for the development of latter drug product classes to not just adhere blindly to the regulatory guidance available, but to make use of a coherent scientific
rationale, this becomes the only workable approach for cell and gene therapies.
Whatever rationale is chosen, it is almost mandatory to discuss it with regulators early on to make sure that buy-in for the overall non-clinical plan as well as for the dose rationale is obtained, and also to implement the corresponding
quality criteria, e.g.
, release criteria with a special emphasis on potency assays.
MC Toxicology Consulting has been extensively engaged in selecting the product-tailored dose translation / starting dose extrapolation strategies for various cell and gene therapies and is continuously expanding its expertise, ready
to share with clients. This includes identifying relevant non-clinical and clinical data, interdisciplinary discussions with quality and clinical experts, as well as preparation of a written rationale for dose selection, ready to be included
in regulatory documents (e.g.
, Briefing Documents or Investigator’s Brochures).
Take Home Message
- Determination of human starting dose that is safe and potentially efficacious is indispensable to the successful development of cell and gene therapies
- Highly customized approaches are needed, taking into account novelty of the product type, clinical experience with a similar product class and the patient population, while objectively recognizing any limitations of the non-clinical test
- Start discussion on strategy early on including an interdisciplinary team of experts as well as regulators
- EMA, 2017. Guideline on Straegies to Identify and Mitigate Risks for Fírst-in-Human and Early Clinical Trials with Investigational Medicinal Products. EMEA/CHMP/SWP/28367/07 Rev. 1.
- EMA, 2018. Guideline on the Quality, Non-Clinical and Clinical Aspects of Gene Therapy Medicinal Products. EMA/CAT/80183/2014.
- EMA, 2019. Guideline on Quality, Non-Clinical and Clinical Requirements for Investigational Advanced Therapy Medicinal Products in Clinical Trials (Draft). EMA/CAT/852602/2018.
- FDA, 2005. Guidance for Industry - Estimating the Maximum safe Starting Dose in Initial Clinical Trials for Therapeutics in Adult Healthy Volunteers.
- FDA, 2015. Guidance for Industry - Considerations for the Design of Early-Phase Clinical Trials of Cellular and Gene Therapy Products.
- Verdera, H.C., Kuranda, K., Mingozzi, F., 2020. AAV vector immunogenicity in humans: A long journey to successful gene transfer. Molecular Therapy 28, 723-746.