Affibody-mediated targeting of HER-family receptors for cancer imaging and therapy
Time: Fri 2022-04-08 13.00
Subject area: Biotechnology
Doctoral student: Charles Dahlsson Leitao , Proteinvetenskap, Stefan Ståhl
Opponent: Professor Christoph Rader, Scripps Research Institute, Florida, USA
Supervisor: Professor Stefan Ståhl, Albanova VinnExcellence Center for Protein Technology, ProNova, Proteinteknologi; John Löfblom, Proteinvetenskap, Proteinteknologi
Proteins are remarkable molecules with diverse and specialized functions playing essential roles in most biological processes. One such function is protecting us from diseases by the action of antibodies in our immune system that can recognize and mediate the destruction of invading pathogens by binding to foreign epitopes found on non-self proteins. The concept of utilizing specific protein-protein interactions to achieve a therapeutic effect has for several decades been a cornerstone for the development of cancer-directed treatments. While antibodies have formed a basis for the development of such drugs, other protein alternatives may be engineered to complement current antibody-based treatments, and may even prove to possess superior features.
This thesis focuses on the engineering of affibody molecules, a small alternative scaffold protein, for design and development of novel cancer-targeting therapeutic and diagnostic drugs. There are many different strategies that have been investigated for inhibiting cancer progression and tumour growth with perhaps one of the most straightforward involving disruption of dysregulated growth-promoting signalling pathways. Members of the human epidermal growth factor receptor (HER) family is prominently expressed in various cancer types and have been shown to be intricately involved in tumorigenesis. One of the members (HER3) often becomes upregulated in cancer and have been shown to mediate acquired resistance to targeted therapies by the mechanism of ligand-induced activation. We have designed five novel affibody-based HER3-targeting molecules able to prevent ligand-binding and consequently activation of HER3. We investigated the targeting properties and biodistribution profiles of these molecules in vivo and subsequently evaluated the anti-tumour efficacy for the most promising variants in direct comparison to a HER3-targeting antibody with a similar inhibitory mechanism. We observed a large influence of design on both the biodistribution properties and the in vivo efficacy of different affibody molecules. Moreover, we demonstrated that two of the affibody-formats were equally effective as the antibody in inhibiting tumour growth and prolonging survival of mice bearing HER3-positive xenografts. The effectiveness of cancer treatments depends on efficient diagnostic approaches that can reliably stratify patients based on these targetable biomarkers, which is possible using radionuclide molecular imaging. We have performed a direct comparison of the diagnostic potential for visualizing HER3-expressing tumours of affibody- and antibody-based imaging probes. We concluded that affibody molecules provide superior imaging quality with higher diagnostic potential and enable early visualization of HER3-expression in tumours.
Another member of the HER family that is of interest for cancer therapy is HER1 (or EGFR) but due to substantial expression in healthy tissues, targeted therapies may lead to severe side-effects. One possible solution to this is taking advantage of the distinct milieu of the tumour microenvironment to design EGFR-targeting drugs that become conditionally activated at the tumour site, but not in normal tissues, with the aim of drastically reducing systemic toxicity. We have generated an affibody molecule with anti-idiotypic binding specificity for a previously generated EGFR-binding affibody molecule, which we used to construct an affibody-based prodrug. We were able to show that, in a proof-of-concept format, this anti-idiotypic masking domain was able to block the binding to EGFR until removed by protease-mediated cleavage. We subsequently developed and characterized a more refined version of this prodrug, which we call a pro-affibody, and could show that activation by cancer-associated proteases confers binding to EGFR-expressing cancer cells and enables conditional cytotoxic payload delivery in vitro. The pro-affibody was further evaluated in vivo using tumour-bearing mice to investigate the feasibility for masked uptake in healthy tissues while retaining binding-activity in tumours. We observed a substantial reduction in EGFR-specific liver uptake compared to a control construct without a masking domain, and a strong indication of protease-mediated EGFR-binding in tumours.
In conclusion, the experimental work presented in this thesis provides a rationale for designing effective affibody-based cancer therapeutics and diagnostics with different targeting strategies and demonstrates the potential of such drugs from preclinical in vivo data.