T cell-based immunotherapy offers a radically different treatment strategy for numerous cancer types and can be effective at treating cancers with previously poor prognoses. Chimeric antigen receptor T cells (CAR-Ts) target surface expressed antigens with antibody-like binding domains fused to a T cell receptor signaling domain, allowing recognition and specific killing of antigen expressing tumor cells. CAR-Ts targeting the B cell surface marker CD19 have shown great efficacy against B cell malignancies in clinical trials. However, one consistent result from CAR-T therapy has been outgrowth of tumor cells that no longer express the target antigen. Antigen escape remains a barrier for the success of T cell therapies, and many strategies have been employed to overcome this challenge, each with their own unique caveat. For example, mixing CAR-Ts targeting multiple single antigens can be effective against heterogenous tumors but adds significant cost to T cell product manufacturing for patients. Additionally, bi-specific and tandem CAR-Ts have been developed but these are often limited by the packaging capacity of the lentiviral construct necessary for delivering the CAR, leading to lower expression and activity of CAR-T cells. In an effort to develop better multi antigen targeting CAR-Ts, the Riddell group (Clinical Research Division) employed a new strategy making use of an alternative antigen-binding protein called designed ankyrin repeat proteins, or DARPins. Their work was recently published in Clinical Cancer Research.
DARPins are small, stable, antibody mimetic proteins consisting of just 33 amino acids, making 2 a-helices and a b-turn. Libraries of DARPins, in which the contact residues have been randomly mutated, have been used for discovery of DARPins with unique antigen-binding specificities. The Riddell group used previously described DARPins with specificities for three model antigens (EGFR, EpCAM, and HER2) to develop novel multi antigen targeting constructs. First, they tested whether a single DARPin CAR-T against EGFR could function with similar efficacy compared to a CAR-T using cetuximab, an EGFR-targeting antibody used clinically. DARPin CAR-Ts recognized and lysed a panel of EGFR-expressing tumor lines as well as the cetuximab CAR-Ts, showing that DARPin-based CAR-Ts are functional. In order to optimize construct sizing, the authors tested constructs where the T cell signaling domain and DARPin were separated by flexible linkers of various sizes. CAR-Ts bearing these constructs were tested using in vitro T cell functional assays and an in vivo model for their ability to control tumor growth. Separation of DARPin and signaling domain with a short spacer and a Gly-Gly-Gly-Gly-Ser (G4S) linker made for the most effective CAR-T construct, controlling tumor growth as well as a cetuximab-based CAR-T.
Next, the authors set about targeting multiple antigens simultaneously. They developed bi-specific DARPin CAR-T constructs targeting both EGFR and EpCAM. Bi-specific CAR-Ts recognized and killed tumor cells bearing each single antigen but were the most effective against tumor cells bearing both antigens. Finally, the authors sought to design CAR-T constructs targeting three antigens simultaneously, only possible due to the compact nature of DARPins. Tri-specific constructs targeting EGFR, EpCAM, and HER2 displayed reduced expression on the T cell surface, and tri-specific CAR-Ts exhibited lower specific lysis of single antigen tumor cells than single targeting CAR-Ts. However, when tri-specific CAR-Ts were tested in vivo against heterogenous tumor cells expressing a mixture of single antigens, they performed better than single targeting CAR-Ts, significantly delaying tumor growth and increasing overall survival. When tri-specific CAR-Ts were tested against tumor cells expressing all three antigens, tumors were completely cleared, highlighting the synergy of multi-antigen targeting.
Outgrowth of antigen negative tumors is a consistent challenge in adoptive T cell therapy, but CAR-T cells targeting multiple antigens are a potential solution to the challenge of antigen escape. The Riddell lab has demonstrated that CAR-T constructs can be designed with compact DARPins and CAR-Ts expressing these constructs are functional and effectively target multiple antigens.
This work was supported by funding from the National Institute of Health and Fred Hutch Walker Immunotherapy Fellowship.
Fred Hutch/UW Cancer Consortium members Stanley Riddell, Michael Jenson, and Qian Wu contributed to this work.
Balakrishnan A, Rajan A, Salter AI, Kosasih PL, Wu Q, Voutsinas J, Jensen MC, Plückthun A, Riddell SR. 2019. Multispecific Targeting with Synthetic Ankyrin Repeat Motif Chimeric Antigen Receptors. Clinical Cancer Research. 25(24):7506-7516.