Introduction
Cell-based therapies are at the forefront of drug innovation. They are highly science-driven, involve innovative drug manufacturing processes and offer potential curative treatment options for diseases, which currently have no or limited treatment options [1]. Hematopoietic stem cell transplantations, as well as adoptive T-cell-therapies and dendritic cell-based therapies, have been used for decades. However, only recently cell-based therapies have evolved from promising treatment options to products for which the therapeutic effect has been established [2,3,4,5]. Since the introduction of the advanced therapy regulation in the EU in 2007 [6] cell-based therapies are regulated as advanced therapy medicinal products (ATMPs) Given the innovative nature of cell-based products and their high variability, development often involves a case-by-case approach [7]. This particularly relates to the non-clinical development program, which is expected to provide a rationale for entering clinical development of a medicinal product and to gain insight into its safety profile. To this end, in vitro and in vivo animal studies are performed, in which the characteristics of the product and the targeted disease determine the developmental approach to a large extent.
Designing a non-clinical development program for human cell-based products is complex, because species-specificity limits the value ofin vivo animal studies. Consequently, in vivo animal studies with the intended human cell-based product (heterologous model) may not always be relevant. While an animal study with the animal equivalent cells (homologous model) may be informative, some insights into the safety and efficacy of a cell-based product can only be obtained when the product is tested clinically [8,9,10].
While some guidance on the non-clinical development program for cell-based therapies is available in regulatory guidelines [11,12] these are not very prescriptive, as flexibility is needed given the large variability in the types of products. For example, these guidelines state that conventional non-clinical pharmacokinetic studies are not relevant, while animal studies on cell migration (biodistribution) and persistence of the therapeutic cells are generally expected. Presumably, because these data are considered to be of value for the characterisation of the risk profile of the cells and the understanding of their potential therapeutic effect. Still, following the fate of cells in vivo is challenging. There are many techniques that can be used to detect specific cells in the body, from basic tissue sampling and subsequent detection of cells by immunohistochemistry (IHC) or specific deoxy nucleic acid (DNA) fragments by polymerase chain reaction (PCR) [13] to highly sophisticated techniques of in vivo cell tracking [14]. Nevertheless, they all have their own pro’s and con’s, and no guidance is provided on which methods are considered most informative.
Regarding the oncogenic potential of cell-based products, the guidelines state that conventional carcinogenicity studies, in which the ability of a medicinal product to transform host cells into tumourigenic cells is evaluated in non-clinical models [15] are not considered appropriate for cell-based therapies [11]. Instead, the potential for tumourigenicity, i.e. tumour formation of the administered (grafted) cells [16] should be considered and addressed as appropriate. Several approaches to study the tumourigenic potential of a cell-based product can be envisioned. These include an in vivo assessment of the human cell-based product in animals but also in vitroassessment of, for example, genetic stability (e.g. by karyotyping and FISH [17], the presence of transformed or undifferentiated cells in the product (e.g. flow cytometry and qRT-PCR [18], or the ability for anchorage or growth factor-independent growth [19].
To increase the insight in the chosen approach for the non-clinical development for cell-based products we analysed the non-clinical data package of cell-based medicinal products in scientific advice reports from 2013 until 2018 provided by the European Medicines Agency (EMA). In particular, the number and purpose of in vivo animal studies was analysed to gain better insight in the prevailing view of sponsors and/or on the relevance of in vivo animal data. Furthermore, as biodistribution and the potential for tumourigenicity are most difficult to study in humans, because of the complexity of detection of the administered cell in a patient [20]) and the need for long-term follow up for the assessment of tumourigenic risks [21] we particularly focussed on the need for and type of studies addressing these aspects in the non-clinical development program.