S227928: A Novel Anti-CD74 ADC with MCL-1 Inhibitor Payload for the Treatment of Acute Myeloid Leukemia (AML) and Other Hematologic Malignancies

Introduction Overexpression of MCL-1 is a well-known mechanism of resistance to venetoclax (ven), a BCL-2 inhibitor. MCL-1 inhibition has shown promising responses in preclinical models of AML in combination with BCL-2 inhibitors leading to tumor regression and prolonged survival. Despite the enthusiasm in testing MCL-1 inhibitors as a treatment strategy in AML, clinical development of MCL-1 inhibitors has been limited by cardiac and gastrointestinal toxicities, which have precluded the delivery of therapeutic doses and led to the discontinuation in clinical evaluation of multiple agents. Hence, new strategies are warranted to achieve MCL-1 inhibition with higher therapeutic margin. Results S227928 is an ADC that couples an MCL-1 inhibitor payload, S64315 (aka MIK665), to an anti-CD74 monoclonal antibody via a cleavable valine-citrulline polyethylene glycol (PEG24) linker. Analysis of RNAseq expression databases and surface expression analysis by flow cytometry (FC) demonstrates that CD74 is highly expressed in several hematologic neoplasms, including AML. CD74 is expressed on the cell surface as the invariant chain of MHC class II and is the receptor for macrophage migration inhibitory factor. Evaluation of 28 primary AML samples by FC revealed CD74 expression in myeloid cells and leukemic blasts, with >60% of patients showing >50% of myeloid cells positive for the target. RNAseq and immunohistochemistry (IHC) analyses showed that physiological expression of CD74 is mainly restricted to cells of hematologic origin, with no expression in liver, gastrointestinal tract, or cardiomyocytes. S227928 demonstrates CD74-specific binding and cell-killing in CD74-expressing AML cell lines as compared to an isotype control ADC. Cytotoxicity correlates with MCL-1 inhibition. Intracellular payload release and target engagement were confirmed by detection of payload measured by LC MS/MS (liquid chromatography tandem mass spectrometry) and a dose-dependent disruption of MCL-1/BIM complex measured via AlphaLISA assay, respectively. In the AML-derived EOL-1 cell line, the MCL-1/BIM complex was more potently disrupted by the ADC compared to the S64315 payload (IC50 0.03 nM vs 0.41 nM), demonstrating enhanced intracellular accumulation of payload via ADC deliverythan with the small-molecule inhibitor. While single agent S227928 showed modest activity in in vitro and in vivo AML preclinical models, S227928 demonstrated potent anti-tumor activity when combined with ven in AML cell lines and in in vivo tumor models. S227928 activity was dependent on CD74 expression, with the most potent and durable responses observed in cell lines and PDX models with elevated levels of CD74 detected by RNAseq or IHC, and more transient responses or a lack of sensitivity in models with low CD74 expression. Complete responses with durability up to 55+ days were achieved in 5/7 primary patient-derived xenograft models and in 3 cell-line xenograft (CDX) models. Additionally, complete response was achieved in an AML CDX grafted on CD34+ humanized murine model, mimicking endogenous CD74 expression in immune compartment and expected TMDD (Target-Mediated Drug Disposition). While these data indicate that patients with no to low expression of CD74 may experience less benefit, it supports the hypothesis that non-target expressing tissues are likely to be spared toxicity. Conclusion Given these promising data, S227928 should lead to an improved therapeutic index via more selective delivery of S64315 to CD74-expressing tumor cells. S227928 will be evaluated in a Phase 1/2 study as a single agent and in combination with ven for the treatment of patients with relapsed/refractory (R/R) AML or chronic myelomonocytic leukemia (CMML) who are no longer candidates for standard therapies.