Targeted therapy is an emerging treatment for a variety of cancers. Small- sized scaffold proteins are an alternative to conventional antibody-based targeting molecules. Two small scaffold proteins—the 58-amino-acid protein class, the affibody molecules, and the 46-amino-acid protein class, the Albumin binding domain Derived Affinity Proteins (ADAPTs)—have previously been engineered to bind to a large variety of tumor-associated molecular targets with a high affinity.
The human epidermal growth factor receptor 2 (HER2) is a membrane-bound receptor for growth signal transmission. Expression of a high level of HER2 can cause cells to proliferate and may ultimately lead to cancer. It has earlier been shown that HER2 is involved in several different types of cancers, e.g., breast, ovarian, bladder, and gastric cancers.
HER2-targeted affibody and ADAPT molecules have previously been developed, such as ZHER2:2891 and ADAPT6 with strong affinity to HER2 with equilibrium dissociation constants of 76 pM and 2.5 nM, respectively. Their small size and high specificity have rendered these two scaffold proteins promising candidates for imaging of HER2-positive breast cancer tumors in clinical trials.
Delivery of cytotoxic agents to cancer cells, using a cell-targeting domain, may potentially precisely kill the cancer cells while having very low cytotoxic effects on normal cells. Many cancer-targeted antibody drug conjugates (ADCs) and toxic proteins (immunotoxins) have advanced the field of cancer treatment. Small-sized scaffold proteins hold promise as alternative targeting domains to build novel drug conjugates or fusion toxins for cancer treatment.
In this thesis, I first investigated an affibody-based drug conjugate (AffiDC) composed of an anti-HER2 affibody and an anti-mitotic maytansine-derived drug (DM1) for treatment of HER2-overexpressing cells. I studied a variety of targeting domain formats for efficacy optimization. All ZHER2:2891-based AffiDCs showed specific anti-tumor activity on HER2-overexpressing cancer cells in vitro as well as in mouse tumor xenografts. The hepatic uptake of the AffiDCs could be reduced by shielding the hydrophobic DM1 using a poly-glutamic-acid spacer, which might help to reduce potential liver toxicity allowing for administration of higher doses. In addition, tuning the valency of the affibody-targeting domain (ZHER2) from a divalent domain to a monovalent domain showed increased potency and reduced liver uptake. We also investigated the influence of the number of drug payloads on the pharmacokinetic profile of the AffiDCs. An AffiDC bearing three DM1s showed higher delivery of DM1 to the cancer cells in vivo, but fast blood clearance and an elevated liver retention was also observed.
With regards to fusion toxin design, we constructed a variety of recombinant toxins. The targeting domains were ZHER2:2891 and/or ADAPT6, which were genetically fused with truncated versions of the highly cytotoxic Pseudomonas Exotoxin A (PE). All fusion toxins we studied showed potent HER2-specific anti-tumor activity. The results suggested that both ZHER2:2891 and ADAPT6 could direct the PE-based cytotoxins specifically to HER2- overexpressing cancer cells.
In this work, we have demonstrated the potential of using ZHER2:2891 and ADAPT6 as targeting domains to carry the small molecule drug DM1, or cytotoxic PE-derived peptides to cancer cells. It can be concluded that careful molecular design of the targeting domain may considerably improve the potency and minimize the off-target uptake.