Introduction: Carfilzomib (Cfz) significantly prolongs progression-free survival in relapsed or refractory multiple myeloma patients, as highlighted in the ENDEAVOR trial. However, Cfz has high incidences of cardiotoxicity and heart failure, leading to treatment cessation. Thus, there is an imperative need for preventive therapies. The study aimed to i) establish an in vivo Cfz cardiotoxicity protocol, ii) investigate the molecular mechanism, identify molecular targets and iii) based on initial results, investigate the potential protective effect and mechanism of Metformin (Met).

Methods: Male, C57BL/6 mice, were randomized in groups as following: Acute protocol (6 days): Control (n=7), Cfz (n=8); Sub-chronic protocol (14 days): Control (n=5), Cfz (n=8); Pharmacological intervention protocol (6 days): Control (n=8), Cfz (n=8), Cfz+Met (n=8), Met (n=4). Cfz (8 mg/kg, ip) was administered on alternate days and Met (140 mg/kg, po) daily. Glucose levels were monitored following Met administration. Mice underwent echocardiography on baseline and at the end of treatments. Blood and myocardial tissue samples were obtained for histology, proteasome activity, PP2A activity and signaling pathways focused on PI3K/Akt/eNOS axis, NO homeostasis and AMPKα-mTOR-mediated autophagy.

Results: Following acute administration, echocardiography in Cfz group presented a significant reduction in fractional shortening (FS%) vs. Control group (39.87±0.47 vs. 43.03±0.50 respectively, p<0.001), combined with reduced thickness in the left ventricular (LV) posterior wall (LVPW diastole (mm): 0.69±0.01 vs. 0.76±0.01, p<0.01; LVPW systole (mm): 1.17±0.01 vs. 1.24±0.02, p<0.01). Sub-chronic Cfz administration resulted in moderate LV dilation (LV end-diastole diameter (mm): 3.24±0.03 vs. 3.04±0.04, p<0.01; LV end-systole diameter (mm): 1.88±0.02 vs. 1.71±0.02, p<0.01) and borderline FS% reduction (42.07±0.46 vs. 43.52±0.25, p<0.05). Following both protocols, Cfz did not cause major tissue lesions.

Signaling pathways were studied in the acute protocol that demonstrated suppressed myocardial contractility. Cfz resulted in significant inhibition of proteasome in both myocardium (55.5% inhibition, p<0.05) and peripheral mononuclear blood cells (PBMCs) (90.6% inhibition, p<0.001) - inhibitory effect was comparable to clinical practice. Cfz, independently of PTEN expression, reduced phospho-PI3K (p<0.05), phospho-Akt (p<0.001) and phospho-eNOS (p<0.001), and increased iNOS expression (p<0.01). Cfz reduced phospho-AMPKα (p<0.001) and phospho-Raptor (p<0.05) leading to inhibition of autophagy, indicated by reduced LC3-II expression (p<0.01), without affecting phospho-mTOR or Beclin 1.

Co-administration of Met prevented FS% reduction (Cfz group: 41.55±0.43 vs. 43.24±0.50 and 43.39±0.56, Control and Cfz+Met respectively, p<0.05), without exerting glucose lowering actions. Met did not interfere with Cfz-induced proteasome inhibition; Cfz and Cfz+Met groups had significantly reduced proteasomal activity vs. Control group in myocardium (p<0.05) and PBMCs (p<0.001 and p<0.01 respectively). Histology presented mild to moderate vascular congestion in Cfz+Met and Met vs. Control and Cfz groups (p<0.05), appearing to be non-specific finding that did not collocate with haemorrhage, vascular obstruction or tissue lesions. On a molecular level, Cfz+Met group presented increased phospho-Akt (p<0.01) vs. Cfz group, independently of PTEN, PI3K and eNOS/iNOS. Cfz+Met group restored phospho-AMPKα (p<0.05), phospho-Raptor (p<0.001) and LC3-II expression (p<0.05) vs. Cfz group, without inducing changes in mTOR and Beclin 1. Cfz inactivated Akt and AMPKα through increased PP2A activity (p<0.05), without altering PP2A expression; Met did not interfere with PP2A activity.

Conclusions: Cfz exhibits decreased global LV function in vivo, without inducing tissue lesions. The molecular mechanism consists of increased PP2A activity leading to inactivation of AMPKα-mTOR and PI3K/Akt/eNOS pathways - combined with NO homeostasis deregulation. In an intervention approach, Met preserved cardiac function via restoring AMPKα-mediated autophagy. Met administration did not restore NO production and homeostasis; these pathways require further investigation. Met emerges to be a promising preventive therapy for Cfz-induced cardiotoxicity.

Disclosures

Kastritis:Amgen: Consultancy, Honoraria, Research Funding; Genesis Pahrma: Consultancy, Honoraria; Janssen: Consultancy, Research Funding; Takeda: Consultancy, Honoraria; Prothena: Consultancy, Honoraria. Dimopoulos:Takeda: Honoraria; Amgen: Honoraria; Janssen: Honoraria; Celgene: Honoraria; Bristol-Myers Squibb: Honoraria. Terpos:Janssen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: member of steering committee, Research Funding; Takeda: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: travel grant, Research Funding; BMS: Consultancy; Genesis: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: travel grant, Research Funding; Novartis: Consultancy; Celgene: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: member of DMC, Research Funding; Amgen Inc.: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: travel grant, steering committee member, Research Funding.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution