CAR T cell therapy
Chimeric antigen receptor (CAR) T cell therapy genetically engineers a patient's own T cells to target a cancer and returns them to the patient, where they go on multiplying. The same expansion of the cells brings both remission and the therapy's main toxicity, and making each treatment for one patient at a time, at a high cost, limits who receives it.
How CAR T cells act
CAR T cells are genetically engineered T cells that kill cancer cells directly by targeting specific surface antigens. They harness the intrinsic "seek and destroy" capabilities of T cells, and they can be customised with genes, inserted by viral vectors or nonviral methods, that redirect the T cell for a more targeted effect. CAR T cell therapy is a new type of cancer immunotherapy and a type of cell and gene therapy, and it offers hope to patients who have gone through multiple lines of treatment to no avail. When a CAR T cell meets a cancer cell displaying the target antigen, the extracellular domain of its receptor binds the antigen. The binding triggers a signalling cascade inside the cell that leads to T-cell activation and proliferation, and the activated cells mount a robust immune response, consisting of cytokine release, that helps eliminate the tumour cells. Chimeric receptors do not only target specific antigens; they can also activate other CAR T cells to help fight the cancer. Many major developers of CAR T cell therapies use the gene-editing tool CRISPR-Cas9 to improve the potency and safety of their therapies, and the tool has been adopted widely because genome editing with it is easy to customise. Each of the six therapies approved by 2023 needs only one dose, after which the cells go on proliferating in the patient's body and, in theory, confer long-term protection against the disease.
In the account of Stephan Grupp, of the Children's Hospital of Philadelphia, remission comes from an initial, very significant proliferative burst of T cells, which produces complete remission within about a month in 85% to 95% of patients. The small number of CAR T cells that remain in the body, still hunting leukaemia cells, then maintain the remission. Every CAR T cell has an activation domain, and the co-stimulatory domain that supplies a second activation signal differs from product to product. Tisagenlecleucel (CTL019) uses 4-1BB, which is associated with a good initial proliferative burst, does a good job of getting patients into remission and can keep the T cells in the patient for years.
Remission and relapse
Clinical trials showed high response rates, with many patients seeing their cancer disappear completely, but some patients still relapse and the majority experience side effects. Among the factors that explain why the effects vary are prior medication use, the timing of treatment, how the CAR T cells were grown and the cellular structure of the CAR T cell.
Standard treatment works well for most children newly diagnosed with acute lymphoblastic leukaemia, but as of 2017 children with refractory or relapsed disease were in really bad shape. They were given very intensive chemotherapy to reach a remission that would make them eligible for a bone marrow transplant, which itself works only half the time, and the chemotherapy, the time in hospital and the risk of that treatment are significant. In recent registration studies of patients in second or later relapse, 20% to 40% reached initial remission and fewer than 10% remained in remission a year or two later. In a single-centre phase 1/2a trial at the Children's Hospital of Philadelphia, whose first paediatric patient was treated a little over five years before 2017, the initial complete remission rate held at about 90% to 95%, and about 80% of patients were alive a year or more out. A number of these patients not only went into remission but kept the engineered T cells in their bodies and stayed in remission without further treatment, transplant in particular. About 10% of the patients treated at the Children's Hospital of Philadelphia went on to transplant after CAR T cell therapy, and most did not. The first two adults treated at the University of Pennsylvania more than six years before 2017 remained in remission and still had engineered T cells when last checked, and the first paediatric patient treated at the Children's Hospital of Philadelphia still had engineered T cells too. That longevity of the engineered T cells first suggested that the therapy could be used without a transplant. Because the children receive no further treatment once they reach remission, they recover quickly and return to school and play.
Cure is a real possibility with CAR T cell therapy, according to a 2023 review of the field. By 2023 two of the first patients to receive CAR T cell therapies had been cancer-free for over a decade, and the infused cells still lived on in them, a positive sign that the cells may go on protecting them against relapse.
Cytokine release syndrome and neurologic toxicity
The main safety concerns of CAR T cells are cytokine release syndrome and neurologic toxicities, and both can be life-threatening and in some cases fatal. Cytokine release syndrome is a systemic response to the activation and proliferation of the cells that causes high fever and flu-like symptoms.
Cytokine release syndrome is the therapy's most significant toxicity in Grupp's account, and his team met it in their very first patient, a girl with completely refractory disease who began CAR T therapy with two thirds of her bone marrow replaced by leukaemia and developed life-threatening cytokine release syndrome. Through intensive research and analysis and quite a bit of luck, the team arrived at the approach it has used for every CAR T patient since: blocking the inflammatory cytokine interleukin-6 with a drug originally developed for rheumatoid arthritis, sometimes within hours. Without that blockade, in Grupp's view, the toxicities would be too great for there to be a CAR T therapy at all. The toxicity control works most of the time, though not every time, and makes these patients much safer to treat. A high tumour burden is a real risk factor for more significant cytokine release syndrome. The cells grow in response to the leukaemia cells they meet, and the more leukaemia cells they see, the more they grow, so the same growth that can take patients with even a very high disease burden into remission also means more T cells, which make more inflammatory proteins and can make the patients more clinically unstable, especially through cytokine release syndrome. Patients treated earlier, before they have refractory disease or a high disease burden, have no dramatic toxicity: they get fevers and some muscle aches and may spend up to a week or so in hospital. In Grupp's view, that toxicity bears no comparison with the toxicity of transplant or intensive chemotherapy.
Development and approvals
The first chimeric T cell receptors were engineered in the late 1980s and showed for the first time that T cells can be supercharged with chimeric genes to target specific cells. In 1993 Zelig Eshhar of the Weizmann Institute of Science in Israel, one of the researchers behind that work, proposed a method of using CAR T cells to target tumours, creating the first generation of CARs. In the mid-1990s Carl June, at Penn Medicine, treated patients with human immunodeficiency virus with genetically engineered T cells. The treatment was not effective against the virus, but months later the activated T cells were found in the blood circulation of almost all the patients, indicating the potential for long-term protection. This persistence is an important element of how CAR-based therapies work. CAR T cell therapies entered their first clinical trials in the early 2000s, against kidney and ovarian cancers, two solid tumours. There the cells were not only ineffective but had "on target, off cancer" effects, destroying healthy tissue, findings that spoke to the importance of target specificity and the need to engineer more precise T cells. A few years later, preclinical research by three independent teams, led by Michel Sadelain, Carl June and Steven Rosenberg, showed that CAR T cells targeting CD19, a common marker on B cells, had significant anti-leukaemic activity. Clinical trials of anti-CD19 CAR T cells were under way in the early 2010s in acute lymphoblastic leukaemia and chronic lymphocytic leukaemia, two malignancies that mostly involve B cells, and for the first time patients achieved partial or complete remission with the treatment. One of those patients, enrolled as a child given only a few weeks to live after no other treatment had worked, had been cancer-free for ten years by 2023.
In 2017 the FDA's Oncologic Drugs Advisory Committee voted in favour of approving tisagenlecleucel (CTL019), the first therapy of its kind in the United States, for paediatric and young adult patients with relapsed or refractory B-cell acute lymphoblastic leukaemia. On 30 August 2017 the FDA approved it, as Kymriah, the first approved CAR T therapy, for patients up to 25 years of age with B-cell precursor acute lymphoblastic leukaemia that is refractory or in second or later relapse. Kymriah is a one-time treatment, unlike chemotherapy or radiation therapy, which need multiple courses over weeks or months. The patient's own T cells are removed, engineered in cell culture outside the body and reinfused, on the theory that the engineered cells will bolster the immune system and attack the cancer. The Clinical Manufacturing Facility of the Raymond G. Perelman Center for Cellular and Molecular Therapeutics at the Children's Hospital of Philadelphia was instrumental in the early development of techniques to manufacture the viral vector that alters the patients' T cells, research that, in collaboration with June, Grupp and others, led to Kymriah. As of October 2017, 154 clinical trials of CAR T therapies were ongoing or recruiting, 44 of them in the United States, so most were being run outside the United States.
On 29 June 2018 the European Medicines Agency's human medicines committee (CHMP) recommended granting marketing authorisations for Kymriah (tisagenlecleucel) and Yescarta (axicabtagene ciloleucel), the first two CAR T cell medicines in the European Union. They are advanced therapies for blood cancer, of a new generation of personalised cancer immunotherapies based on collecting and modifying patients' own immune cells. Kymriah was recommended for paediatric and young adult patients up to 25 years of age with B-cell acute lymphoblastic leukaemia that is refractory or in second or later relapse, and for adults with relapsed or refractory diffuse large B-cell lymphoma after two or more lines of systemic therapy; Yescarta for adults with relapsed or refractory diffuse large B-cell lymphoma and primary mediastinal large B-cell lymphoma after two or more lines of systemic therapy. A risk management plan is an integral part of the authorisation, and the product information describes monitoring and mitigation strategies for cytokine release syndrome and neurologic toxicities. The committee also recommended adding the treatment of CAR T cell induced cytokine release syndrome as an indication for the medicine used against it. Another important risk management measure, and a condition of the authorisation, is the use of a patient registry to monitor the long-term safety and efficacy of the two therapies, and the agency has qualified a registry for collecting post-authorisation safety and efficacy data. Kymriah and Yescarta were the first medicines supported through the agency's PRIME scheme to receive positive opinions from the committee. As advanced therapy medicinal products, they were also assessed by the Committee for Advanced Therapies, the agency's expert committee for cell-, gene- or tissue-based medicines.
Since the first regulatory approvals of CAR T cell therapies in 2017 there has been an explosion of interest in the field, and by 2023 every aspect of the treatment was evolving rapidly: regulation, manufacturing, supply chain and logistics, pricing and access, and the possible indications. By December 2023 six CAR T cell therapies had been approved by both US and EU regulators, and many more were expected over the next decade.
| Brand name | Scientific name | Disease treated | Target antigen |
|---|---|---|---|
| Abecma | idecabtagene vicleucel | Multiple myeloma | B-cell maturation antigen |
| Carvykti | ciltacabtagene autoleucel | Multiple myeloma | B-cell maturation antigen |
| Breyanzi | lisocabtagene maraleucel | Diffuse large B-cell lymphoma, high-grade B-cell lymphoma, primary mediastinal large B-cell lymphoma, follicular lymphoma grade 3B | CD19 |
| Kymriah | tisagenlecleucel | Acute lymphoblastic leukaemia, diffuse large B-cell lymphoma, follicular lymphoma | CD19 |
| Tecartus | brexucabtagene autoleucel | Acute lymphoblastic leukaemia, mantle cell lymphoma | CD19 |
| Yescarta | axicabtagene ciloleucel | Diffuse large B-cell lymphoma, follicular lymphoma | CD19 |
The indications combine marketing authorisations from the EU and the US and differ by region. B-cell maturation antigen and CD19, a protein expressed on the surface of B cells, are both common targets for CAR T cell therapies.
Manufacturing, waiting times and cost
Tisagenlecleucel was, in Grupp's account, the first treatment of its kind ever presented for FDA approval, being gene therapy and engineered cell therapy at once. It is a personalised product manufactured for each patient: the patient's own T cells are taken out, genetically reprogrammed to go after tumour cells and reintroduced. Other immuno-oncology drugs instead reactivate natural T cells in the patient's body that might be able to recognise tumours, which works well for certain patients across a variety of tumour types. The genetic modification uses a lentivirus, which has regulatory and safety implications, and the FDA advisory committee was very interested in the long-term safety of that viral technology. Showing that the therapy was safe over the long term was very important to the agency. The University of Pennsylvania licensed the therapy to Novartis, which was willing to develop it first for paediatric and young adult patients with acute lymphoblastic leukaemia. Novartis had transferred manufacturing from the academic setting to the pharmaceutical company so that hospitals across the country could prescribe the therapy. The Novartis registration trial had a global reach, 25 centres in 11 countries including Canada, Japan, the United States and several EU countries, with cells made from patients around the world and shipped across the globe, which had never been done before. A paediatric indication was also offered first for FDA consideration, whereas most indications are approved in adults first, with paediatric indications following or children treated off label.
One of the greatest challenges facing CAR T cell therapy, and cell therapy more broadly, is its complicated manufacturing. The six therapies approved by 2023 are made from autologous T cells, which are collected from and delivered to the same patient. The T cells are isolated from peripheral blood mononuclear cells, enriched, grown to the dose and preserved until administration, in a vein-to-vein workflow that typically takes two to three weeks, and they are often transported across continents several times before they reach the patient. Allogeneic, "off the shelf" CAR T cells would come from any healthy donor and otherwise be processed the same way. The race is on to develop them because they would make the therapy more accessible, but their major disadvantages are a higher risk of graft-versus-host disease and of the innate immune system attacking the CAR T cells. Cell and gene therapies are individualised rather than one-size-fits-all, and each treatment is made one at a time, which results in long waits. In one physician survey, about 25% of patients eligible for CAR T cell therapy died while waiting for treatment. About a third of patients receive the treatment too late, which lowers its effectiveness and increases mortality risk. Most patients who hope to receive the treatment face unacceptable wait times, and bringing cell and gene therapies to patients at scale is not yet possible with current methods. Bringing CAR T therapies to more patients will involve further testing of their effects as first- and second-line treatments and building a faster manufacturing process.
Bringing down the cost of CAR T cell therapies is a top priority. In 2023 the treatments on the market cost £450,000 for the treatment alone, and the total cost of care in the US healthcare system had been found to exceed US$1 million for some patients. Discussions of lowering the cost of goods cite Pharma 4.0, which involves digitisation, automation and continuous manufacturing, as a necessary step. Future CAR-based treatments will include automated and continuous manufacturing protocols that will help reduce the cost of goods. Another way to reduce costs is to enable allogeneic therapies manufactured closer to the patient, eliminating overseas shipping and vastly improving patient reach. Industry was working rapidly to train scientists to expand manufacturing capacity in the rush to get products to the clinic.
Among the aspects of CAR T that matter to a writer, in Grupp's view in 2017, are the challenges of manufacturing in which one lot goes into one patient and all the release testing is repeated for every product made, the use of a lentiviral system to modify the T-cell genome, and its cost, which would be known only once the therapy was approved. Its future, in his view, turns on two questions: whether manufacturing can be made less manual and more automated, faster, cheaper and with fewer people, and whether the therapy can be extended beyond haematological malignancies.
Solid tumours and other uses
The immunosuppressive tumour microenvironment and the varied nature of solid tumour cells make it harder for current-generation CAR T cells to target solid cancers selectively. All the therapies approved by 2023 are indicated exclusively for blood cancers, which tend to express antigens more uniformly. In Grupp's account, response rates were good across leukaemia, lymphoma and multiple myeloma, with acute lymphoblastic leukaemia by far the best. Solid tumours such as lung, breast and pancreatic cancer are very good at excluding T cells, so there the challenge is to get CAR T cells not only into the body, where they can proliferate, but into the tumour tissue itself. A writer should know, in his view, that the therapy is great for blood cancers but that there is still a way to go before it has the same efficacy in solid tumours, a hard problem to solve. One possibility he sees, as sometimes happens in responses to checkpoint inhibitors, is that inflaming a tumour starts a beneficial immune response that pulls in T cells and can destroy the tumour. Grupp would not over-promise, but with so many potential approaches to opening solid tumours to T cells he was very hopeful there would be a way.
Future CAR-based treatments will include additional cell types and indications. Autoimmune and cardiometabolic diseases may one day be treated with them, a variety of cell types could serve as starting material, and in 2023 clinical trials were under way of CAR natural killer cell therapy, CAR macrophage therapy, tumour-infiltrating lymphocyte therapies, T-cell receptor-based therapy and extracellular vesicle therapy. The 2023 review of the field holds that what is known about cell and gene therapies in 2023 will undoubtedly change within five to ten years, and holds that the hurdles CAR T has overcome, and the hope that one dose could ward off cancer for good, guarantee that its study will continue.