Tim Farries, ERA Consulting
The reported powerful efficacy immunotherapies using genetically-modified effector cells has the potential to profoundly change the treatment of some cancers. However, the cutting-edge technologies used create a number of challenges for commercialisation, not least in terms of acceptability by regulatory authorities. This article considers some of the issues that must be addressed in converting these promising new treatments into medicinal products that can be approved for use in the European Union.
This article judges that marketing authorisation and commercialisation should both be achievable. However, after a period of painfully slow progress for many “traditional” advanced therapies (such as in vivo treatments with a viral gene therapy vector), developers and regulators alike need to adopt new ways of thinking in order to be ready for an oncoming surge of treatments with cells modified ex vivo (outside of the body), as some of these will plead for accelerated approval.
Consider the field of CAR-T cells (T cells made to express chimeric antigen receptors that target tumour antigens) that is currently grabbing the headlines. The first wave of these treatments (from Juno Therapeutics, UPenn/Novartis, Kite Pharma and others) use autologous cells (cells derived from the patient themselves) modified with retroviral vectors (such as lentivirus) to introduce expression of the chimeric protein. Here are some of the key challenges, which can overcome:
Autologous cell products.
These products are often imprecisely characterised, heterogeneous and highly variable. However, EMA have previously approved advanced therapies that use autologous cells. The most recent case of these, Holoclar, can be particularly instructive (the others are ChondroCelect, MACI and Provenge). This product has all the challenges associated with a mixed cell population combined with a high potential for patient-to-patient variability. Holoclar also exhibits the constraint of a need to administer the product within a short time window that precludes the possibility of usual levels of final product testing (even of sterility) before administration[TF1] . Without such comprehensive end-point testing (i.e. release testing), the regulators had to consider if the safety of the product is adequately assured through other measures, such as raw material and in-process testing, and validation. A notable factor is that the donor is a potentially unavoidable source of contamination of the starting material with infectious agents that would be unacceptable for the manufacture of other types of medicinal product. Furthermore, whilst not subjected to gene transfer, Holoclar manufacture does include ex vivo cell cultivation in which it is critical to keep the differentiation state of its limbal stem cells under control. The approval of Holoclar shows that the EMA can accept product controls that are limited by practicality, when in the context of other compelling evidence (such as clinical) of safety and efficacy.
Cell modification using integrating vectors.
To date EMA has only approved one gene therapy (Glybera, approved under exceptional circumstances), and which uses a non-integrating AAV vector. Vectors that integrate new genes into the genome present new regulatory challenges. Whilst ex vivo adoptive transfer can be much safer than an in vivo administration of a gene transfer vector, developers should still prepare studies to show absence of any potential for viral contamination or propagation from the final product, and absence of possibilities for transformation of any treated cells (including any minor populations of other cell types) into cancerous phenotypes. Manufacturing technologies for retroviral vectors also need maturing for scalability and purity of product. It is noteworthy that a marketing authorisation application has been submitted (and requested accelerated review) for an autologous cell product transduced with a retroviral vector (CD34+ cells transduced to express adenosine deaminase for treatment of children with Severe Combined Immune Deficiency (SCID) caused by a deficiency of this enzyme). The MAA was submitted by GSK, for a therapy originally developed by the Fondazione Telethon and Fondazione San Raffaele in Italy. If approved this case will offer considerable encouragement to the CAR-T sector.
The advantage of autologous over allogeneic cells (cells derived from another individual, frequently as a cell line) is that of immune compatibility with the patient. However, it must be considered if that compatibility could be broken by the introduction of an engineered chimeric protein or any possibility for expression of any other vector-encoded proteins. Immunogenicity of biological products can be a very complex consideration, but has nevertheless not prevented a number of other highly immunogenic biopharmaceuticals from being developed (for example adenoviral vectors).
Another revolutionary technology with related issues is gene editing. The newest technology, using the CRISPR/cas9 system, improves the ability (already developed to some extent with engineered DNA recognition proteins, such as zinc finger proteins) to introduce genetic corrections or new genes into precisely targeted genomic sites. The most immediate applications will be for ex vivo genetic modification of cells for therapeutic purposes, such as introducing genes for antigen-recognition proteins into immune cells. If genetically-modified cell products can be approved using less precise integration techniques (e.g, CAR-T cells made with standard retroviral vectors), it may be anticipated that targeted integration will necessarily be safer and more acceptable. However, a claim of specific integration must also be supported by data on aspects such as the frequency and location of any changes to any other parts of the genome (“off-target”). Introduction of new genes and regulatory elements could also convey secondary effects on the expression of other genes (notably neighbouring genes in the genome), with unpredictable safety consequences. The arrival of modern technologies for rapid analyses of genomic sequences and gene expression profiles should prove very timely for enabling analytical assurances of safety from such theoretical risks.
Gene editing may also be utilised to modify stem cell populations to be used for regenerative medicine. The major safety consideration with somatic stem cell products is usually the potential to create any transformed cancerous cells. However, the example of Holoclar is again pertinent for showing the approvability, in principle, of a regenerative medicine product containing stem cells. Any future prospect for gene editing in vivo will face challenges of avoiding any germ-line modifications. Characterisation of biodistribution after administration is important for any cell or gene therapy product, and therapies using viral vectors must additionally consider if there is potential for shedding of active virus after treatment. These issues can all be addressed by well-designed safety studies (although it is to be hoped that legislators will not inadvertently obstruct therapeutic applications of uncontroversial somatic therapies in efforts to respond to the dangers of use of gene editing on embryonic stem cells or other germ-line tissue).
Another paradigm shift of thinking that is required for this new generation of therapeutics is in the design of GMP manufacturing facilities; a model of one batch at a time is not the way to manufacture significant numbers of patient doses of an autologous product for which each patient has to be a new batch. Layouts must allow multiple batches to be processed at the same time, which therefore requires rigorous systems for isolation and traceability of each patient’s materials and highly organised management of logistics.
It is the conclusion of this article that while the next generation of cell-based therapies do present novel regulatory challenges, these can be met by application of sound scientific principles and reference to experience. Written guidance is incomplete, so developers must use Scientific Advice procedures to avoid guessing wrongly what the regulators will require. The process for obtaining such advice from the EMA is admittedly burdensome in terms effort for preparation and the total time taken before the outcome is received. The procedures can probably be made simpler, faster and more accessible but are critical for aligning a development programme to regulatory expectations.
[TF1]Foot note reference to EMA Assessment report, 18 December 2014, EMA/25273/2015