By Deborah Borfitz
November 5, 2020 | An international team of researchers has demonstrated that an organ-on-a-chip model can accurately measure the effects of a chemotherapeutic on breast cancer tissue and simultaneously identify the unintended side effects on healthy heart tissue. In practical terms, that means “it may be possible to identify not only what drug is going to be effective for which patient but also whether that person has had previous issues, such as a heart attack, which can affect their outcome as well,” according to team member Ali Khademhosseini, director and CEO for the Terasaki Institute for Biomedical Innovation.
A detailed description of the dual-compartment system, which was used to show the differing impacts of the widely used chemotherapy drug doxorubicin (DOX) on healthy and diseased cardiac tissue, recently published in Small. The “heart-breast cancer-on-a-chip” platform is designed so that induced pluripotent stem cell (iPSC)-derived cardiac tissues interact with breast cancer tissue and it incorporates a sensor to monitor multiple, cell-secreted biomarkers.
The rationale for the study is that individual patients respond differently to a particular therapeutic regimen, including susceptibility to side effects, says Khademhosseini. While first-generation use of organs on chips focused on toxicity during the drug development process, subsequent iterations of the technology are now being used to create disease models to start exploring what happens, good and bad, when a patient is given a drug—in this case, someone with breast cancer.
Chemotherapeutic drugs are particularly harmful to the heart, but current detection approaches are either highly invasive (myocardial biopsy) or miss the early warning signs (echocardiography and MRI) when it’s still possible to reverse cardiac failure, says first author Junmin Lee, a postdoc at the Terasaki Institute and first author of the study. As the researchers discovered, the widely employed biomarkers cardiac troponin T and creatine kinase-MB isoenzyme (CK-MB) could be used for earlier identification and prediction of chemotherapy-induced cardiotoxicity (CIC).
Relative to the conventional enzyme-linked immunosorbent assay, the platform’s electrochemical immuno-aptasensors proved to have much higher sensitivity and lower detection limits for early monitoring of cardiotoxicity as well as breast cancer progression using human epidermal growth factor receptor 2 (HER-2) as a biomarker. The system also showed that the production rates of biomarkers from cardiac and breast cancer tissue improved when a nanoparticle-based DOX delivery system (previously described GYSM-NPs, DOI: 10.1016/j.msec.2018.07.060) was used as the drug carrier.
Individualized Disease Monitoring
The heart-breast cancer-on-a-chip model is revolutionary in several respects, says Lee. For starters, it is larger and more comprehensive than most other organ-on-a-chip technologies because it was developed as a dual-organ system with compartments for culturing healthy and fibrotic heart cells as well as breast cancer cells.
The model is also engineered to be physiologically relevant. The compartments are connected to microfluidic channels that pump and circulate oxygen- and nutrient-enriched media to help keep the tissues alive, Lee explains. For the recent study, a computational model simulated the distribution of flow rate, oxygen level, and drug concentration where the three tissue types were located.
Human, iPSC-derived breast cancer spheroids, as well as cardiac spheroids comprised of cross-talking cardiomyocytes and fibroblasts, first get cultured in hydrogels intended to create a more native-like microenvironment, he continues. Otherwise healthy cardiac tissue was made fibrotic with a supplement of transforming growth factor-beta 1, and breast cancer tissue grew from a cell line that overexpresses HER-2.
Breast cancer and heart cells get interconnected using a microfluidic-based perfusion system, Lee says. Incorporated sensors allow the secretion rates of multiple biomarkers to be measured at the same time. Importantly, tissues did not have to be destroyed in order to monitor their behavior and the biomarkers they release, notes Khademhosseini. Traditionally, drug toxicity is assessed using two-dimensional cultures, where cells quickly stop functioning because they have no means to get gradients of oxygen and nutrients or to interact with other cell types in the human body.
Among the study findings were that the interactions between breast cancer and cardiac tissues were required to accurately reflect the levels of biomarkers associated with cardiac functionality and cancer progression, says Lee. The tissue models on the dual-organ system could also be useful for the study of CIC, although biomarkers other than troponin T may be needed for early detection among breast cancer patients with preexisting cardiac fibrosis.
The next step is to use the heart-breast cancer-on-a-chip model on iPSC-derived cardiac tissue and biopsy samples from the same patient, and look at the variability across individuals, says Lee. Further validation studies will also be needed to see, among other things, if biomarkers for monitoring cardiotoxicity and breast cancer progression can be confirmed by well-known markers of cardiomyocyte dysfunction, says Lee. The research team also plans to scale the model to encompass more organs, notably the liver since it is involved in drug metabolism.
Clinical validation studies of the heart-breast cancer-on-a-chip model with many different patients will be needed before the system becomes a standard of care with the FDA’s seal of approval, says Khademhosseini. Those trials would correlate patients’ cells and medical history and assess how predicted and actual outcomes compare.
Ultimately, the goal is for the platform to become useful in real-world clinical settings for the early detection and prediction of CIC—and individualized disease modeling for breast cancer and beyond.