Heart failure and cancer are leading causes of morbidity and mortality worldwide. Beyond the cardiotoxic side effects of many anti-cancer drugs, these two disease entities are interdependent, with each serving as a risk factor for the development of the other. To investigate the direct effects of cancer cells on cardiomyocyte contractile function, we established a novel preclinical in vitro model that enables the assessment of cancer cell–induced changes in the contractile properties of engineered heart tissues (EHTs) generated from human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs).
In co-culture with prostate carcinoma PC-3 cells, hiPSC-CM EHTs developed significantly lower contractile force (at 2.5 Hz, PC-3 group: 0.15±0.01 mN; - 21.1%), compared with EHTs cultured alone or with non-malignant PNT2 cells. Analysis of the co-culture medium revealed a distinct cytokine release profile of PC-3 cells and identified biomarkers associated with human cardiac dysfunction, such as growth/differentiation factor 15 (GDF15), which may contribute to the observed loss of contractile function.
Pharmacological treatment with recombinant GDF15 led to a dose-dependent decrease in contractile force (at 2.0 Hz, 2.5 ng/ml GDF15: 0.13±0.02 mN; - 10.8 %), accompanied by a metabolic shift toward glycolysis indicated by an increase of glucose consumption (2.5 ng/ml GDF15: 1.6±0.2 mmol/l; + 65 %) and lactate production (2.5 ng/ml GDF15: 1.3±0.07 mmol/l; + 34 %). Modelling the Frank–Starling response in an organ bath demonstrated increased passive stiffness after GDF15 treatment, evidenced by faster elevation of diastolic force in response to increasing stretch (at 1250 µm, 0.6±0.05 mN; + 106 %). Consistent with these findings, proteomic analysis indicated suppression of translational and mitochondrial components, along with activation of immune response and extracellular matrix production.
In contrast, treatment with a GDF15-neutralizing antibody (GDF15-Ab) prevented contractile dysfunction and even increased contractile force compared to untreated controls (at 2.0 Hz, 250 ng/ml GDF15-Ab: 0.19±0.01 mN; + 32.5 %). Proteomic analysis supported these results, showing reduced immune activation and enhanced cytoskeletal organization. Ongoing experiments suggest that this preventive effect of the GDF15-Ab is maintained in co-culture with PC-3 cells. These findings align with current and independent clinical trials employing GDF15 antibodies to counteract cancer cachexia.
Overall, this study reinforces the concept that GDF15 is a key mediator in the crosstalk between energy metabolism, inflammation, heart failure, and cancer, supporting its potential as a unifying biomarker across these conditions. Furthermore, this in vitro co-culture model of hiPSC-CM EHTs and cancer cells provides a promising platform to study cardiotoxic effects of cancer therapies and to elucidate mutual paracrine interactions between cancer cells and cardiomyocytes.
