Date: 30th April 2021
Chimeric antigen receptor (CAR) T cells have been successfully used to treat B cell malignancies but their use for solid tumours has been hampered by the lack of specific target surface antigens. As few antigens are truly specific, the cross-reaction of such CAR T cells can cause lethal toxicity. Furthermore, as cancerous cells often display antigens heterogeneously, this allows antigen-negative tumour cells to evade targeting. Now, two publications describe the use of ‘smart’ immune cell therapy, which use synthetic notch (synNotch)–CAR T cells which integrate recognition of multiple imperfect but complementary antigens, increasing specificity and providing superior control of solid tumour burdens in mice. In addition, tonic signaling and exhaustion which often occurs with traditional CAR T cells was averted.
There is an urgent need for the development of more specific and potent next-generation engineered T cell therapies. In particular for solid tumour types, which are inherently hard to treat with traditional modes of cancer therapy, and often have strong immunosuppressive microenvironment which renders many solid cancers refractory to treatment. By using previously developed ‘prime-and-kill’ circuits in which synNotch receptors activate a transcriptional output upon recognition of the ‘priming’ antigen –the subsequent expression of a CAR directed against a ‘killing’ antigen is activated. The theory is that the resulting circuits function as Boolean AND gates, requiring the recognition of both priming (synNotch) and killing (CAR) antigens, and provide a new level of specificity using imperfect targets.
Two related studies published this week in Science Translational Medicine from researchers at University of California San Francisco, UCSF, US, have demonstrated how to engineer smart immune cells that are effective against solid tumors, opening the door to treating a variety of cancers that have long been untouchable with immunotherapies.
The first study was led by Wendell Lim and Hideho Okada, and described the use of synNotch-CAR T cells for the treatment of glioblastoma (GBM). Glioblastoma is the most aggressive primary brain tumour in adults, and has extremely poor prognosis with less than 5% of patients surviving 5 years following diagnosis.
Here, the team engineered T cells that firstly recognised a cancer-specific but heterogenous antigen such as glioblastoma neoantigen epidermal growth factor receptor splice variant III (EGFRvIII) or the central nervous system (CNS) tissue-specific antigen myelin oligodendrocyte glycoprotein (MOG). This then induced the local expression of a tandem CAR which simultaneously targeted two killing antigens, EphA2 or IL13Rα2 – which are more homogeneous GBM antigens. However, as both of the killing antigens are expressed outside the tumour tissue it makes them non-ideal antigen targets for conventional, single-target CAR T cells.
The team tested their smart cell therapy in mice bearing GBM patient-derived tumours, they found EGFRvIII-triggered synNotch-CAR cells to kill tumors in vivo only if they contained primed cells. They were able to effectively and completely clear tumours, without killing surrounding normal tissue or EphA2- or IL13Rα2-positive cells in other parts of the body that lacked colocalised priming antigen. Once the T cells moved away from the priming environment, CAR expression was lost within hours, reducing the toxicity in off-target tissues. The strategy also had the unexpected side-effect that the cells were able to sustain long-term proliferation and killing responses. Traditional CAR T cells often become exhausted rendering them ineffective, here the smart cells were kept in a naïve-like state that was more durable and therefore less subjected to exhaustion.
In a similar fashion, the second publication led by Kole Roybal and Bin Liu demonstrated an improved specificity and persistence of antitumor activity using therapeutic T cells with synthetic Notch (synNotch) CAR circuits. This time the team identified alkaline phosphatase placental-like 2 (ALPPL2) as a tumor-specific antigen expressed in a spectrum of aggressive solid tumors, including asbestos-driven mesothelioma, pancreatic, testicular and ovarian cancer. Whilst, ALPPL2 could act as a sole target, the team also used it as the ‘priming’ sensor and increased the specificity by combining it with tumour-associated antigens such as melanoma cell adhesion molecule (MCAM), mesothelin, or human epidermal growth factor receptor 2 (HER2).
The smart cell therapy was tested in human mesothelioma and ovarian cancer mice models and was found to have superior efficacy and persistence relative to conventional CAR T cells. In the majority of animals a complete response was observed. Once again T cell exhaustion was eliminated, likely due to the SynNotch regulation of CAR expression, which maintained cells in a ‘standby’ mode allowing conservation of energy. The therapy also led to the maintenance of T cell memory, leading to superior tumour control upon rechallenge.
Both teams here have demonstrated that engineered smart immune cells, using synNotch-CAR circuits to perform a two-step prime-and-kill function, effectively controlled a range of solid tumours, eliciting long-lived memory T cells, and averting T cell exhaustion. These discoveries will open new avenues to treating a variety of cancers, that have been untreatable with more traditional immunotherapies. With this in mind, both labs are moving towards clinical trials in the near future.
We are starting to see the next generation of therapies evolving for the treatment of solid tumours. A new immunocytokine therapy showed promise for GBM recently, where engineered antibody-cytokine protein fusions showed anti-cancer activity in mice and the first few human patients in a pilot trial showed promising early responses. Others have been investigating a combo immunotherapy, combing oncolytic virus with CAR T cell therapy to target and eradicate solid tumours in mice. Here, the virus targeted tumour cells forcing them to express CD19, which then traditional CD19-directed CAR T cells could recognise.
However, the ability to harness new or existing CAR targets that are clinically imperfect will be a powerful tool, and is likely to accelerate the translation of such therapies into the clinic. To note, Lim and colleagues previously developed a machine learning approach to assemble a catalogue of antigen combinations that can precisely target cancerous cells whilst leaving normal cells alone. This will enhance the identification of other antigens and will open up the possibilities of more sophisticated circuits that could further improve targeting of smart cells.
Form more information please see the press release from UCSF
Choe, J.H., Watchmaker, P.B., Simic, M.S., Gilbert, R.D., Li, A.W., Krasnow, N.A., Downey, K.M., Yu, W., Carrera, D.A., Celli, A., et al. (2021). SynNotch-CAR T cells overcome challenges of specificity, heterogeneity, and persistence in treating glioblastoma. Science Translational Medicine 13, eabe7378.
Hyrenius-Wittsten, A., Su, Y., Park, M., Garcia, J.M., Alavi, J., Perry, N., Montgomery, G., Liu, B., and Roybal, K.T. Ibid.SynNotch CAR circuits enhance solid tumor recognition and promote persistent antitumor activity in mouse models. eabd8836.