Fully Synthetic Trioxacarcins for Use in Antibody-Drug Conjugates

Abstract

Dissertation advisor: Professor Andrew G. Myers Ethan L. Magno

Fully Synthetic Trioxacarcins for Use in Antibody-Drug Conjugates Abstract Antibody-drug conjugates (ADCs) have emerged as an effective, targeted therapeutic modality in which a small molecule warhead is coupled to a cancer cell-selective monoclonal antibody. This conceptually simple design strategy belies the careful coordination of toxic payload, linker chemistry, and antibody engineering that has proved to be a formidable challenge in the discovery of safe and effective ADCs. Advances in the field of ADCs are enabled through the expansion of synthetically accessible payloads differentiated with respect to their mechanisms of action, as well as innovation in the means by which these warheads are coupled to the antibody. This dissertation presents the development of fully synthetic analogs of the anticancer compound trioxacarcin A (1.1) as novel payloads in ADCs. Trioxacarcin A is the most potent member of a class of bacterial metabolites exhibiting sub-nanomolar 50% growth inhibition (IC50) against a broad panel of human cancer cell lines. Its antiproliferative activity arises from intercalation of DNA and subsequent guanine N7-alkylation by the spiro-epoxide. In this work, I employ a diversifiable synthetic route to trioxacarcin A recently completed in the Myers lab to develop novel fully synthetic trioxacarcins (FSTs) with enhanced potency, stability, and structural simplicity. The convergent assembly of diazoketone 1.18, cyanophthalide 1.20, and cyclohexenone provided dideoxy-FST analogs containing functional handles suitable for ADC attachment while retaining the minimal trioxacarcin pharmacophore and nanomolar potency against cancer cells. Structural modifications to the C13 hemiketal were found to have a direct influence on DNA alkylation efficiency, which was correlated with cytotoxicity. Two fundamental strategies for the conjugation of potent trioxacarcin analogs to the anti-HER2 antibody trastuzumab are presented. In the first, FSTs bearing amine substituents were attached to antibody thiols via dipeptide linker systems to provide trioxacarcin-ADCs such as Trastuzumab-R02-D2. These drug-linkers were both stable under physiological conditions and were readily cleaved by the lysosomal protease cathepsin B, facilitating rapid intracellular delivery of free drug upon ADC internalization into cancer cells. In the second strategy, the critical epoxide of the trioxacarcin payload was replaced with an unreactive bromohydrin prodrug to prevent off-target alkylation. ADCs such as Trastuzumab-R02-G3 were synthesized using a methylene alkoxy carbamate linkage to the bromohydrin alcohol, whereby lysosomal cleavage of the glucuronide linker system was necessary prior to reversion to the active epoxide. ADCs employing these novel linker strategies are currently under investigation for selective delivery of FST payloads to HER2-expressing cancer cells. We anticipate these constructs will serve as candidates for the safe and effective treatment of cancers refractory to existing therapies.