An in vivo model for cardiotoxicity of myeloid cell leukemia 1 inhibition (mcl1)

Z. Papdi (Heidelberg)¹, M. Heckmann (Heidelberg)², N. Frey (Heidelberg)², L. H. Lehmann (Heidelberg)^2
1Universitätsklinikum Heidelberg Klinik für Innere Medizin III, Kardiologie, Angiologie, Pneumologie Heidelberg, Deutschland; ²Universitätsklinikum Heidelberg Klinik für Innere Med. III, Kardiologie, Angiologie u. Pneumologie Heidelberg, Deutschland

Background: Myeloid cell leukemia 1 (Mcl-1) is an anti-apoptotic protein of the Bcl-2 family that is overexpressed in hematologic malignancies such as acute myeloid leukemia and multiple myeloma. Its upregulation is associated with increased therapy resistance and poor prognosis. The Mcl-1 inhibitor MIK665 (also known as S64315) demonstrated promising results in preclinical models with absence of cardiovascular side effects in in vitro models (e.g. hiPSC-derived cardiomyocytes), which led to the initiation of Phase 1 clinical trials for the treatment of Acute Myeloid Leukemia (AML). However, these trials had to be discontinued due to unclear cardiotoxicity evident by troponin elevations. The clinical relevance of elevated troponin and its underlying mechanism remain unclear.

Purpose: Aim of this study is to investigate MIK665-induced cardiotoxicity and to enable the development of cardioprotective treatments, allowing the clinical use of the inhibitor to various cancers harnessing its efficacy against tumor progression.

Methods: In order to improve our understanding of the underlying mechanism and clinical relevance of the cardiotoxic effects caused by Mcl-1 inhibitors in humans, we used a mouse model with the humanized Mcl-1 protein and subjected these mice to MIK665 therapy ± Venetoclax, a potential protocol for AML therapy. Biomarkers and left ventricular function were monitored weekly.

Results: Groups treated with either MIK665 or MIK665 and Venetoclax, a specific Bcl-2 inhibitor, showed significantly elevated troponin T levels after treatment (hs-troponin T [pg/ml] 83.4 ± 31.6 vs 693 ± 160 vs 1276.6 ± 465; placebo vs MIK665+Venetoclax p<0.001). However, mice demonstrated no symptoms of heart failure or cardiomyopathy as the left ventricular pump function remained preserved (EF [%] 75 ± 8.67 vs 74.17 ± 7.03 vs 77.79 ± 6.67). These results resemble the clinical observation of the human phase 1 trial.

Conclusion: Using a mouse model with a humanized Mcl-1 receptor, we were able to mimic the cardiotoxicity seen in phase I trials on Mcl-1 inhibitors, which were overseen during drug development. This builds the basis for a better mechanistic understanding of Mcl-1 inhibitor induced cardiomyopathy and demonstrates that hiPS-cell lines as well as non-humanized murine models might miss important cardiotoxic mechanisms due to the artificial environment of iPS-cell lines and differences in proteinsequences between humans and mice. Further functional and molecular characterization will allow us to give mechanistic insights and open the way for cardioprotective strategies.