Engineering of affibody affinity proteins for cancer immunotherapy

QC 2025-11-04

Abstract

Antibodies and antibody-derived fragments have long been the conventional choice for the development of tumour-recognising affinity proteins in cancer therapy and diagnostic applications. This thesis investigates the potential of using small, non-immunoglobulin affibody affinity proteins as alternatives, or complements, to conventional antibody-based formats. To this end, the main focus of this work has been the development of affibody-based Natural Killer (NK) cell dual engagers for potential cancer treatment. By simultaneous engagement of tumour-associated antigens (TAAs) and the NK cell-activating receptor CD16a, the aim is to use these affibody-based dual engagers to re-direct NK cells to the cancer cells and trigger NK cell-mediated tumour killing. This approach could offer a novel and modular affibody-based platform for engaging NK cells, while also expanding the repertoire of NK cell-based immunotherapies currently in development.

In Paper I, we developed affibody binders targeting B-cell maturation antigen (BCMA), which is an overexpressed and clinically validated target in multiple myeloma (MM). Following two consecutive phage display selection campaigns, post-selection characterisation of one candidate anti-BCMA affibody clone, denoted 1-E6, demonstrated its high binding affinity to BCMA, with an approximate K_D of 1.4 nM as measured by surface plasmon resonance (SPR), and ability to bind specifically to the BCMA-expressing cell line MM.1s, using both flow cytometry and microscopy techniques.

In Paper II, phage display selection was used to develop affibodies targeting CD16a. SPR-based evaluation of two anti-CD16a affibody candidates, denoted H09 and A10, demonstrated moderate binding affinities to CD16a and, more interestingly, recognition of two non-overlapping CD16a-binding epitopes. Notably, the bi-paratopic H09-A10 combination demonstrated a markedly enhanced binding to CD16a through avidity. Further combination with the anti-BCMA affibody 1-E6 yielded a BCMA × CD16a dual engager, with a bi-paratopic CD16a-binding arm, which demonstrated ability to bind both CD16a and BCMA simultaneously in SPR. In vitro functional cell assays using MM.1s cells and donor-derived NK cells showed that the BCMA × CD16a dual engager efficiently activated NK cells in a BCMA-dependent manner, and mediated triggering of potent NK cell cytotoxicity.

Paper III focused on the development of CD16a binders with improved properties, using phage display and a methionine-free second-generation combinatorial library based on the anti-CD16a affibody A10. SPR analysis of one isolated second-generation clone, denoted 34a-A11, demonstrated a modest improvement in CD16a affinity, with a K_D of 14 nM for 34a-A11 compared to 24 nM for A10. A bivalent (BCMA)₂ × CD16a dual engager was constructed by genetic fusion of 34a-A11 with two anti-BCMA 1-E6 affibody units. Cell microarray-based target selectivity screening of this bivalent dual engager against > 6,000 human membrane and secreted proteins demonstrated selective binding to BCMA and both CD16a and CD16b isoforms. Further, functional cell assays using co-cultures of the dual engager, with MM.1s cells and either engineered CD16a-expressing Jurkat cells or donor-derived NK cells, showcased BCMA-specific NK cell activation and NK cell-mediated cell killing. In a blocking assay using IgG as the blocking reagent, the activity of the dual engager was not inhibited in the presence of IgG, thus highlighting the dual engager’s non-competitive binding with IgG to CD16a. This result was also supported by AlphaFold 3 modelling.

Finally, in Paper IV, the anti-CD16a 34a-A11 affibody was combined with previously generated anti-human epidermal growth factor receptor 2 (HER2) affibody or ABD-Derived Affinity ProTein (ADAPT) binders, Z_HER2:2891 and ADAPT6 respectively, to generate HER2 × CD16a dual engagers. Both dual engagers were investigated in a 3D spheroid breast cancer model, using HER2-expressing BT474 cells grown as ~300 µm spheroids, which showed that the dual engagers penetrated deeper into the spheroids than the anti-HER2 mAb trastuzumab. Furthermore, in co-cultures with donor-derived NK cells, the dual engagers demonstrated complete lysis of the spheroids, with faster kinetics than trastuzumab.

Taken together, the work presented in this thesis demonstrates the potential of using affibody molecules as both the cancer cell-binding arm and the NK cell-binding arm in small, affibody-based NK cell dual engagers, offering a novel and modular platform for NK cell-based cancer immunotherapy.

urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-372270