Abstract
P246
Introduction: The therapeutic window of radioligand therapeutics (RLTs) is often restricted by suboptimal tumor-to-non-tumor ratios. Antibodies can have high affinity and specificity to tumor targets, but their long systemic half-life frequently results in hematological toxicities. Alternatively, low molecular weight ligands are restricted to a limited number of tumor targets and often exhibit insufficient tumor retention and tissue selectivity. Therefore, alternative molecular platforms are urgently needed to exploit the potential of RLTs in a broader field of indications.
DARPins (Designed Ankyrin Repeat Proteins) are small binding proteins (ca. 15 kD) that combine short systemic half-life, ideal binding properties and very high stability. DARPin molecules with very high affinity and specificity can be easily generated against a broad range of tumor targets, and several DARPin-based non-radioactive products are currently investigated in clinical trials. We have previously shown that increased DARPin affinity correlates with elevated tumor uptake and extended tumor retention in mouse models (Zahnd et al., 2010) suggesting that radiolabeled binders with picomolar affinity will lead to tumor uptake levels meaningful for therapeutic applications. The simple and robust architecture of DARPins further provides high thermal stability, which is beneficial for labelling with radionuclides that require harsh conditions, and which enables sequence-engineering approaches that are not compatible with other protein scaffolds.
However, the limitation of small-sized, protein-based targeting agents for RLT originates from their renal clearance pathway, which leads to a strong kidney accumulation of coupled residualizing radionuclides resulting in kidney toxicities. To overcome this problem, we have undertaken an extensive engineering campaign to optimize the surface of the DARPin scaffold for reduced kidney reabsorption.
Methods: Several surface optimized DARPins variants were engineered against different tumor targets and were analyzed for their biophysical properties. Building on the absence of cysteines in the DARPin scaffold, we generated single-cysteine versions for site-specific conjugation to a DTPA chelator using maleimide chemistry. The molecules were radiolabeled with Indium-111 and subsequently analyzed for their in vivo biodistribution properties in different xenografted tumor mouse models. Furthermore, tumor penetration of candidates was analyzed in detail by immunohistochemistry on tumor sections.
Results: Our in vitro characterization showed that DARPin engineering did not impact the affinity to the target antigen and the good biophysical properties were maintained. At the same time the in vivo biodistribution profiles of surface engineered DARPins were strongly improved as compared to parental binders. In fact, kidney accumulation was reduced by up to 90% at four hours post injection while tumor uptake was not affected (Figure 1). This effect was confirmed with different DARPin candidates suggesting a general applicability of the approach. Combined with other orthogonal strategies for reduced kidney accumulation, we were able to improve the tumor-to-kidney ratio from 1:34 to 1:3 for one of our Her2 targeting candidates in preclinical mouse models. Additionally, our immunohistochemistry analysis revealed that the mono-DARPins studied above exhibit a homogeneous and deep tumor penetration, which is highly superior to other targeting agents like antibodies and other DARPin formats with extended size. Refinements of the applied engineering strategies to further optimize the tumor-to-kidney ratio are currently ongoing and an update on the newest findings will be presented.
Conclusions: The presented results show that our proprietary "Radio DARPin Therapy" platform represents an attractive solution for the development of next-generation RLTs. Several programs in indications with high unmet medical need are currently underway.