1. Introduction
Chimeric antigen receptor T cells, known as CAR-T cells, have generated extraordinary results in phase I/II clinical trials in the treatment of CD19+ B-cell hematological malignancies. The principle is based on the genetic modification of the patient’s immune T cells by transferring a transgene coding for a chimeric receptor. This receptor recognizes antigens present on the surface of targeted tumor cells, regardless of any major histocompatibility complex (MHC) restriction, leading to their destruction. CARs are usually composed of an extracellular domain—a single-chain variable fragment (scFv) of a monoclonal antibody implicated in the recognition of the target cell antigen—and an intracellular domain responsible for the activation and the function of T cells [
1]. Different generations of CAR-T cells have been raised, depending on the composition of the intra-cellular domain: (i) first generation, CD3z chain [
2]; (ii) second generation, CD3z chain and a co-stimulation domain such as CD28, 4-1BB or OX40 [
3,
4]; and (iii) third generation, CD3z chain and two co-stimulation domains. A fourth generation called TRUCKS (T cells Redirected for antigen-Unrestricted Cytokine-initiated Killing) has been recently developed, where the transgene coding for a second-generation CAR-T is completed with a gene coding for a cytokine such as IL-12 or IL-15, for example [
5].
Clinical trials for CD19 CAR-T cells have shown a high level of complete or partial remissions in patients with poor prognoses. The results from the ELIANA and ENSIGN studies [
6] showed a complete remission of 67% at 3 months in patients with acute lymphoblastic leukemia (ALL), which was maintained in almost 40% of patients after a median follow-up of 9 months. For non-Hodgkin lymphoma (NHL) patients, with a minimum of 6 months of follow-up, the objective response rate was 82%, with a 54% complete remission rate in 101 patients (ZUMA-1 study) [
7]. The median overall survival rates were 78% at 6 months and 52% at 18 months. These significant results concern patients who had already received several lines of treatment; however, they were associated with severe adverse effects, especially cytokine release syndrome (CRS), an over-activation of the immune system, and neurotoxicity. Some of these effects are responsible for significant morbidity and mortality. The ELIANA and ENSIGN studies reported a major CRS and severe neurological toxicity in 77.2% of patients (ELIANA). The ZUMA-1 study observed a grade >3 CRS in 13% of patients, responsible for the death of 2 patients, and neurotoxicity in 28% of patients. Adverse effects have sometimes led to the discontinuation of a clinical trial. For example, Juno Therapeutics stopped its phase II ROCKET trial in ALL B in 2017 following five deaths caused by cerebral oedema. The experience gained from the clinical data has led to recommendations for the management of these events (tocilizumab availability and intensive care unit), to reduce their severity and increase patient survival. The use of a suicide gene, currently being tested in clinical trials (ClinicalTrials.gov Identifier: NCT02761915, NCT03373071 and NCT03373097), allows control of the CRS in cases of excessive over-activation. Nevertheless, despite these results and an effective management of adverse effects, more than half of patients will experience a relapse. In fact, according to data from follow-up studies, between 30-50% of patients who have been in remission are found to relapse within one year of the infusion of anti-CD19 CAR-T cells [
8]. Cases of tumor escape have also been observed due to the absence or loss of CD19 expression by the tumor cells. To overcome this escape, other models with multispecific CAR-T cell are studied to decrease the risk of relapse most commonly by targeting CD19 and/or CD20 and/or CD22 in B cell malignancies. Moreover, it is still difficult to transpose this therapy to solid tumors. These involve numerous barriers of physical (access to the tumor), immunological (immunosuppressive environment induced by the tumor) and tumor targeting specificities, with an “off target” effect observed on healthy tissues. The development of this therapy in solid cancers is a major challenge for academic research teams and the pharmaceutical industry.
The first CAR-T cells were commercialized in 2017 as Yescarta
® (axicabtagene ciloleucel) from Kite/Gilead (Foster City, California) [
7,
9] and Kymriah
® (tisagenlecleucel) from Novartis (Basel, Switzerland) [
6] following U.S. Food and Drug Administration (FDA) authorization under the Advanced Therapy Medicinal Product (ATMP) status (
Figure 1A). Marketing authorizations were then obtained for (i) the anti-CD19 Tecartus
® (brexucabtagene autoleucel) from Kite/Gilead (Foster City, California) in 2020 for relapsed or refractory (r/r) mantle cell lymphoma and for adult patients with r/r B-cell precursor ALL, and (ii) for the anti-B-cell maturation antigen (BCMA) Abecma
® (idecabtagene vicleucel) from Celgene (Summit, New Jersey) in 2021 for the treatment of adult patients with r/r multiple myeloma after four or more prior lines of therapy, including an immunomodulatory agent, a proteasome inhibitor and an anti-CD38 monoclonal antibody. At the end of 2021, another anti-CD19 CAR-T cell, Breyanzi
® (lisocabtagene maraleucel) from Juno Therapeutics, Inc., a Bristol-Myers Squibb Company (Seattle, Washington), was granted authorization for the treatment of adult patients with r/r large B-cell lymphoma after two or more lines of systemic therapy, including diffuse large B-cell lymphoma (DLBCL), high-grade B-cell lymphoma, primary mediastinal large B-cell lymphoma, and follicular lymphoma grade 3B. At the beginning of 2022, Breyanzi
® was under review in Europe, Switzerland and Canada. Soon after, an anti-CD38 CAR-T cell, Carvykti™ (ciltacabtagene autoleucel) was authorized by the FDA in r/r multiple myeloma. As can be seen in
Figure 1A, the development of CAR-T cells first began in academic centers before licensing to a pharmaceutical company, either directly (Novartis for Kymriah
®) or through the creation of a start-up (Kite Pharma/Gilead for Yescarta
®) (Foster City, California).
The production of CAR-T cells in an autologous context was not easy to implement for pharmaceutical companies used to producing batches of medicinal products from one batch of raw material. For autologous CAR-T cell generation, each raw material is for one batch of ATMP dedicated to the same patient. This led to a very intricate organization, requiring the participation of the academic leukapheresis center, the academic cell therapy unit, transport companies, and the academic hospital pharmacy (
Figure 1B) [
11,
12]. As these key steps, at least, are not under industrial control, pharmaceutical companies insisted that each academic center observed specific rules and procedures to validate the opening of the center [
13]. Such an intricate organization presents different drawbacks, such as (i) the complete production time (between 17 to 22 days according to the manufacturers), requiring production platforms as close as possible to the collection centers, to reduce this delay; (ii) the high production capacity of the pharmaceutical company platforms, to be able to answer the increasing medical need; (iii) an impact on the final cost of the CAR-T cells, which is very high, limiting the dissemination of this treatment all over the world. This final cost is partly related to this organization, for a pharmaceutical company that needs to ensure complete control of the ATMP production process. On the other side, academic centers are used to producing cells in an autologous context with a dedicated organization for each patient, as this is already performed around the world for hematopoietic stem cell transplantation (HSCT).
In this systematic review, we present the state-of-the-art clinical trials involving CAR-T cells in the world in 2022, before performing a comparative analysis between European trials and the world leaders (China and USA). Finally, we have attempted to analyze the modes of CAR-T manufacturing carried out in European trials led exclusively by academic centers. The results allow us to discuss the reasons for Europe’s delay and the role of academic centers in the production of CAR-T cells, particularly autologous CAR-T cells.
4. Discussion
The overview of international trials up to 2022 showed that most of the trials are in recruitment or active status with a majority of phases I and II trials, even 20 years after the first patient infusion. As seen in the analysis, clinical trials for CAR-T cells are mainly carried out by USA and China, which remain the two leaders in terms of the number of trials, variety of conditions and targets studied. On the other hand, European trials on CAR-T cells are still very limited. We were able to see that Europe also has a large majority of trials in phase I and phase II. One advanced phase trial (Phase III) was led by Novartis Pharmaceutical (Basel, Switzerland) was suspended. Moreover, the European trials are sponsored by either an academic center or by a pharmaceutical company, and the collaboration between the two is still very weak (three European trials), whereas the latter represents most cases for international trials. The indications studied relate mainly to hematological malignancies, especially lymphoblastic leukemia and lymphoma, which were the first indications targeted by the treatment of CAR-T cells. For these hematological malignancies, CD19 remains the main target studied by European trials, although it has been widely studied for several decades. Moreover, even though there are struggles with solid tumor treatment in general, Europe gathers only 14 studies compared to 400 international studies, with just two innovative targets compared to international trials. From the data collected, we can see that Europe remains far behind the USA and China in this field in terms of the number of CAR-T cell trials. This observation could be explained at least partly by the difference in financial and regulatory policies dedicated to the development of these ATMPs. Indeed, the USA and China implement programs that facilitate the funding of research for innovative drugs.
4.1. Regulatory Differences
The American, European and Chinese regulatory agencies have their own specialized committees to evaluate advanced therapies [
27]. With the development of ATMPs over the last 20 years, regulatory authorities have had to adapt to allow their marketing. collaboration between the FDA and the European Medicines Agency (EMA) has improved this regulation, despite differences still existing. These different regulatory authorities have different classifications for the qualification of ATMPs. In terms of marketing approval, there is specific legislation depending on the legal categorization of the product that can explain why some ATMPs are marketed in some regions but are not authorized in others, especially in Europe. China has also launched a policy of developing ATMPs with clear guidelines for rapid evolution [
28]. The National Medical Products Administration (NMPA) of China wants to align with the ICH (International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use) guidelines. For clinical trials, a dual track regulation mode exists for the development of ATMPs. Clinical trials are typically initiated in investigator-initiated trial (IIT) mode in individual hospitals, before being transferred to the investigational new drugs (IND) file for marketing submission. Product registration from foreign data can be directly transposed in China for rare diseases and high-unmet medical needs, without running a trial. For high-unmet needs with a low prevalence in the population, a bridging study needs to be run with a limited patient number. The current regulatory requirement in China is less stringent than in the USA, requiring permission only from the internal hospital ethics committees to run a clinical trial on a cell and gene therapy product. This makes it easier to navigate the complex regulatory processes and get ATMPs into clinical trials, but it can also lead to the introduction and study of questionable products if there is not appropriate oversight and rigorous product validation criteria.
In recent years, regulatory agencies have launched various accelerated programs to reduce the processing time and enable products to quickly reach the market. ATMPs are generally eligible for these programs, which mainly concern products that fulfil an unmet medical need or have a potential major therapeutic benefit. The FDA has launched the “Fast Track Designation, Breakthrough therapy”, “Accelerated approval”, and “Priority review and Regenerative Medicine Advanced Therapy” programs. The NMPA encourage biotechnology innovation with reinforcement in the same vein, with “Breakthrough Therapy”, “Priority review”, and “Conditional approval” designations. For the EMA, we can mention the following programs: “Conditional Marketing Application”, “Authorisation under exceptional circumstances”, or “Accelerated assessment” [
29]. In 2016, the Priority Medicines (PRIME) program allowed for the initiation of an early dialogue with the product developer in order to accelerate the process. The Breakthrough Therapy, Fast Track, and PRIME designation systems share the same objective of faster access to innovative medicines, but these programs have a different legal basis, making comparison and harmonization difficult [
27]. In 2022, the EMA set up a pilot study to support non-profit academic organizations in the development of ATMPs. This pilot will provide enhanced regulatory support for up to five selected ATMPs that address unmet clinical needs and are only developed by academic and non-profit developers in Europe. These regulatory processes concern the best practice principles for manufacturing and clinical development planning that meets regulatory standards (regulatory flexibilities, development support measures, fee reductions). The first ATMP that will benefit from this pilot study is ARI-0001, developed by the Hospital Clínic de Barcelona [
22], one of the most advanced centers for the academic production of CAR-T cells. This CAR-T cell product also previously received PRIME designation in 2021.
The question of intellectual property must also be clear for both parts, with the pharmaceutical company remaining the one to carry the patent until marketing authorization. In 2019, the Novartis group withdrew its patent on anti-CD19 CAR under pressure from the associations Médecins du Monde and Public Eye, thus allowing academic centers to study and manufacture anti-CD19 CAR-T cells. For several years, the Chinese government has been moving towards harmonization for the protection of intellectual property, which interests pharmaceutical companies looking to enter the Chinese ATMP market [
30].
4.2. Financial Support
The three regions benefit from financial support to promote ATMP development, but not to the same extent. In the USA, the NIH, part of the U.S. Department of Health and Human Services government institutions, conducts medical and biomedical research. They generate numerous collaborations with academic and/or industrial centers for the development of CAR-T cells. Moreover, the U.S. Department of Health and Human Services is working with the Biomedical Advanced Research and Development Authority (BARDA), which benefits from a panel of industrial collaborators, allowing the support of ATMP development. This system was particularly highlighted during the COVID-19 pandemic, with a fast and efficient reaction in the production of vaccines, for example. In the same vein, China emphasizes the collaboration of multinational companies and the investment of capital in cell and gene therapies. This is evidenced by the number of clinical trials, although not all the information on these trials can be found on the clinical trials website. Collaborations between Chinese companies and international companies, such as Johnson & Johnson (New Brunswick, New Jersey), Roche (Basel, Switzerland), or Merck (Rahway, New Jersey), have allowed the manufacturing services in China to grow. The discussion of intellectual property is also progressing, making it possible to interest and promote companies seeking to engage in gene and cell therapy products. Within Europe, we have seen that collaborations with industrial groups are not yet on the agenda. The majority of trials are either carried out by an academic center or by a pharmaceutical company. The close collaboration of government organizations and industrial groups is poorly developed, which can notably explain this delay compared to the USA and China. The support of an industrial partner seems to be a necessity for academic centers, in order to pursue development towards advanced phases of clinical trials, as we have seen with the latest medicines approved on the market [
31]. The implementation of policies supporting ATMPs is beginning to take shape in Europe. For example, in France, a bioproduction/biotherapy policy in the program France 2030 is making it possible to release investment funds in this area, encouraging collaborations between pharmaceutical companies and academic centers. Another example in Europe is the CARAMBA trial, which is led by a consortium of 11 partners from Germany, Italy, France and Spain, and pharmaceutical companies, notably DRK-Blutspendediens (Hessen, Germany), who possess 18 manufacturing sites in Germany. This project focuses on an anti-SLAMF7 CAR-T cell in multiple myeloma, which is manufactured virus-free [
32]. This trial is part of the Horizon2020 research and innovation program on new therapies and rare diseases, and has been supported for over four years with funding of 6.1 million euros [
33].
4.3. Academic Center for Autologous Production
As previously mentioned, the production of autologous CAR-T cells generates an intricate circuit, with several healthcare and industrial partners having to work together to ensure patient treatment and the management of side effects. This complexity leads to an increase in the raw cost of this therapy. Academic centers, which already manage both autologous and allogeneic HSCT for patients, could legitimately pretend to be CAR-T cell producers in an autologous setting. Indeed, through their proximity to hospitals, academic production centers allow translational research to be carried out, working directly on patient samples. It is possible to screen patients’ tumors to study potential new tumor targets, especially in solid tumors. The study and optimisation of several types of cell sources, different gene sequences and expansion conditions are generated primarily in these centers. The development of virus-free CAR transduction techniques (e.g., Sleeping beauty, CRISPR/Cas9) is also one of the current challenges in the manufacturing process [
34,
35]. These centers may also benefit from nearby GMP facilities approved by regulatory authorities, which allow the production of ATMPs for patients [
36]. In France, the French-speaking society of bone marrow transplantation and cellular therapy (SFGM-TC) provides recommendations for the research and development of CAR-T cells by academic centers [
37]. To date, clinical trials are mainly focused on an autologous cell source with a hospital and industrial circuit that is now well described, with preventive management of side effects [
38]. The final product derived from a patient is dedicated to the same patient. Due to the complexity of CAR-T cell production, it would be easier to consider this type of production with the use of automated systems. These systems have been on the market for several years, the best known being the Miltenyi Prodigy
® (Bergisch Gladbach, Germany, and more recently the ADVA X3 from AdvaBio (Haïfa, Israel) [
39]. In addition to CAR-T cell production, this last platform provides interesting flexibility in the choice of buffers and on-line monitoring throughout the process. It includes software that enables the programming and design of any predefined protocols, so it can provide a customized product for each individual patient. These new features, which were not available on previous devices, could facilitate point-of-care manufacturing for several patients. This device will be tested in a clinical trial for the first time in 2023. The development of fully automated production platforms with in line controls and advanced data analysis are currently being implemented to overcome these challenges [
17,
21].
4.4. Allogeneic CAR-T Cells
In the near future, the development of allogeneic CAR-T cells seems the best strategy to allow wide use of this therapy, provided immune tolerance is achieved. Nevertheless, studies are still ongoing to generate a product from a controlled cell source, off-the-shelf, reducing infusion delay and cost. This configuration seems to match better with pharmaceutical companies, with standardized procedures for the production and qualification of a batch. This would require scale-up production systems, with the possibility of on-line product controls and real-time feedback. It could allow production to be adapted to each raw material from different cell types, and quality control throughout the process. These controls remain heterogeneous tests, with different parameters followed due to the large number of clinical trials on CAR-T cells, and there is an absence of guidelines for the implementation of these controls [
40]. Harmonization of these phenotype, expansion, functional, and potency tests would allow CAR-T cell controls to be aligned on the same basis, improving the final comparability of CAR-T cell characteristics before release. It would also provide more important manufacturing, storage and logistic capacities than can be achieved by an academic center.
The scale-up of allogeneic CAR-T cell production in bioreactors will allow the generation of batches containing a high number of specific cells [
41,
42]. The product could be followed throughout the process with automated and on-line monitoring systems, with the possibility of feedback, while remaining in a closed system. Indeed, CAR-T cells remain a living cellular product with inter-individual variability. This has an influence on the quality of the final product; being able to adapt the state of the intermediate product and modify the parameters would be ideal to improve the final yield, i.e., the number of cells and their viability. The development of these programs is in the near future, with the advent of “machine learning”. Moreover, manufacturers would benefit from a digital twin in silico, which would also allow them to perform upstream simulations on the variation of a parameter during production. This in-line control and feedback would save time and money, with a better knowledge of intermediate and final products. The implementation of this new kind of software, based on artificial intelligence, would allow the production of ATMPs to evolve to a new stage [
21]. This therapy is undergoing changes in its conception and in the management of its academic and industrial production. Regulatory evolution concerning ATMPs and financial support from each country would also allow solutions to be found to accelerate their development and market authorization, especially in Europe.
5. Conclusions
The development of the first CAR-T cells started in an academic center, before the establishment of industrial collaborations, leading the product towards marketing authorization. Four new CAR-T cell ATMPs have recently been authorized, all the result of collaborations, showing that this alliance is still necessary to bring this therapy to this crucial stage. Europe, which lags behind the USA and China, has not put forward this kind of collaboration.
However, industrial production implies a high cost, with restrictive logistics, especially for the patient and the agents involved in the hospital circuit. It is conceivable that academic centers could take over the production of autologous CAR-T cells. Indeed, academic centers remain the primary players in the development of this therapy. Efforts by regulatory authorities are beginning to emerge to fill the unmet gaps that remain in anti-cancer therapy, notably in Europe. These efforts should support academic centers efforts to bring their ATMPs to the market, with regulatory requirements that match the capacity of an academic center. We have seen that Europe is lagging behind, probably due to the low level of academic-industry collaboration, which has not been encouraged by financial support and is hindered by unfavorable regulations. The development of CAR-T cells should certainly evolve towards collaborations based on the USA model, while providing support to academic centers who could use their experience to participate in the production of autologous CAR-T cells. In France, considering public institutions are not legally authorized to produce and sell pharmaceutical products, the participation of academic centers is still limited to clinical trials. This does not apply to other European countries, who could rely on their academic centers with an organization model that could reduce production costs and thus the cost of a CAR-T cell product.
The delegation of production to academic centers, under the cover of an industrial partnership or the creation of a spin-off from the academic center, would make it possible to reduce this cost, while industry could focus on the production of allogeneic CAR-T cells, known as off-the-shelf.