top of page

Switching Up T Cell Therapies

Author: Katie Dale

With the huge effort that’s currently being spent developing more resistant T cell therapies with new targets and increased efficacy against solid tumours, it’s easy to overlook the toxicity concerns and the innovative ways scientists are tackling them. Both chimeric antigen receptor (CAR) T cells which recognise external antigens such as CD19, and T cell receptor (TCR) engineered T cells that recognise internal antigens presented by MHC complexes, show toxicity due to cross reactivity and cytokine release syndrome (CRS). Despite the numerous success stories of the CAR-T cell therapy Kymriah (one of two FDA approved therapies) in the treatment of blood cancers, 21% of patients died from adverse effects; of these, 47% experienced cardiac toxicity, and 23% experienced CRS¹. These figures have driven the development of cellular technology that allows medical professionals to better control CAR- and TCR-engineered T cells once they have been infused into the patient, including methods for switching them both on and off.

In 2013 the danger of cross reactivity was highlighted, when two patients died in a clinal trial testing a new adoptive cell therapy, using TCR-engineered T cells targeting the experimental tumour antigen MAGE A3². The normal cellular function of this protein is unknown, however it is a cancer testis antigen expressed in many cancer types, making it an attractive target for adoptive cell therapy. In this study the authors designed a panel of genetically modified TCRs based on the wild type TCR sequence, so that the engineered T cells recognise and bind the same peptide antigen with higher affinity. The engineered patient T cells were then tested in vitro for cross reactivity against other MAGE family peptides, and in mouse models of ovarian cancer to confirm efficacy³.

Despite the TCR-engineered T cells showing no toxicity in mouse models, and no expression of MAGE A3 in human cardiac tissue, the first two patients that received the therapy died within 5 days of infusion due to cardiac injury. Post-mortem and further investigation revealed engineered T cells had infiltrated the heart tissue, suggesting they were able to recognise an unrelated antigen that the lower affinity wild type TCR did not. Complex in silico analyses revealed that the culprit was a peptide from the muscle protein titin, which had the same amino acids at the 5 most important positions for epitope recognition. Further testing revealed that whilst human titin activated T cells by around half that of MAGE A3, mouse titin was unable to induce activation, explaining why the mouse model did not reveal any cross reactivity.

One approach that could help avoid potential toxicities caused by unexpected cross reactivity problems is engineering of a ‘suicide switch’ into the T cells. Bellicum Pharmaceuticals have patented a clever system whereby engineered T cells also express an inducible Caspase-9 (iC9), that can be activated in response to the clinically inert drug rimiducid (AP1903)⁴. AP1903 binds to FKBP domains on iC9, triggering homodimerization and activation of the caspase cascade, resulting in apoptosis. This allows medical professionals to rapidly ablate engineered T cells in patients without causing inflammation or damaging other cell types, which could help reduce serious adverse effects in patients receiving adoptive T cell therapies.

Conversely, the use of AP1903 to switch on engineered T cells has also been explored recently. In the early days of development, costimulatory domains were introduced into first generation CAR-T cells, to increase their potency and persistence in patients and achieve greater tumour control. In a recent study, Duong and colleagues introduced inducible MyD88 and CD40 (iMC) costimulatory domains to CAR-T cells, which possessed the same FKBP domains allowing dose dependant activation of signalling through the addition of AP1903⁵. This system again allows medical professionals to control CAR-T cells after the patient is infused, with costimulation of T cell activity only permitted in the presence of AP1903. This provides the added benefit of enabling CAR-T cells to exist in a dormant state during remission, with the ability to reactivate upon detection of relapse.

Duong and colleagues also demonstrated that by using an alternative suicide switch responsive to rapamycin, they were able to engineer a two-dimensional system where CAR-T cells could be switched on to stimulate proliferation and tumour killing, yet still remain responsive to an iC9 mediated suicide switch. Using orthogonal switches in this way could provide an excellent tool for medical professionals to finely tune the activity of T cell therapies in patients.

This article is but a snapshot of the problems adoptive T cell therapies are facing and the solutions currently in development to address them. Patients experience toxicity due to cross reactivity in all sorts of tissues. Also, CRS, not discussed here, accounts for a similar number of deaths and serious adverse events during clinical trials. So far, use of iC9 suicide switches in patients has been proven safe, and lymphoma patients are actively being recruited into a trial testing an iC9 CD19 CAR-T cell therapy. These switches could have a huge impact in the field; hopefully in the coming years the suicide switch will increase survival rates in patients, while the activation switch could lead to better overall response rates due to increased tumour control.

References 1. Anand K, Pingali S, Ensor J, Neelapu SS, Iyer S. Comprehensive Report of Anti-Cd19 Chimeric Antigen Receptor T-Cells (car-T) Associated Non Relapse Mortality (cart-Nrm) from Faers. Hematological Oncology. 2019;37(S2):313–313.

2. Linette GP, Stadtmauer EA, Maus MV, Rapoport AP, Levine BL, Emery L, et al. Cardiovascular toxicity and titin cross-reactivity of affinity-enhanced T cells in myeloma and melanoma. Blood. 2013 Aug 8;122(6):863–71.

3. Cameron BJ, Gerry AB, Dukes J, Harper JV, Kannan V, Bianchi FC, et al. Identification of a Titin-Derived HLA-A1–Presented Peptide as a Cross-Reactive Target for Engineered MAGE A3–Directed T Cells. Sci Transl Med. 2013 Aug 7;5(197):197ra103.

4. Spencer DM, Bayle JH, Foster AE, Slawin KM, Moseley AB, Collinson-Pautz MR, et al. Methods for controlled activation or elimination of therapeutic cells [Internet]. US20160175359A1, 2016 [cited 2020 Sep 9]. Available from:

5. Duong MT, Collinson-Pautz MR, Morschl E, Lu A, Szymanski SP, Zhang M, et al. Two-Dimensional Regulation of CAR-T Cell Therapy with Orthogonal Switches. Molecular Therapy - Oncolytics. 2019 Mar 29;12:124–37.


bottom of page