Measuring Heart Toxicity of Cancer Drugs
A stem cell-derived heart muscle cell. Proteins that are important for muscle cell contraction are highlighted in red and green, and cell nuclei are blue. Credit: Joseph C. Wu, M.D., Ph.D., Stanford Cardiovascular Institute
Using human heart cells generated from adult stem cells, researchers have developed an index that may be used to determine how toxic a group of cancer drugs, called tyrosine kinase inhibitors (TKIs), are to human cells. While 26 TKIs are currently used to treat a variety of cancers, some can severely damage patients’ hearts, causing problems such as an irregular heartbeat or heart failure.
For the study, reported February 15 in Science Translational Medicine, the researchers used stem cell-derived heart cells from 13 volunteers to develop a “cardiac safety index” that measures the extent to which TKIs kill or alter the function of heart cells. They found that the TKIs' toxicity score on the index was generally consistent with what is known about each drug's heart-related side effects.
This work follows on the heels of an earlier study from the same research team, published in Nature Medicine, in which they assessed the heart cell toxicity of doxorubicin, a chemotherapy drug that also causes heart-related side effects, including heart failure. In that study, the researchers used stem cell-derived heart cells from women with breast cancer to correctly predict how sensitive each woman’s heart cells were to doxorubicin.
Such tests could ultimately help the pharmaceutical industry identify drugs that cause heart-related side effects earlier in the drug development process and help the Food and Drug Administration (FDA) during the drug review and approval process, said the study's senior author Joseph C. Wu, M.D., Ph.D., director of the Stanford Cardiovascular Institute.
“I hope this research will be helpful for individual patients, once we further implement precision medicine approaches,” he added.
Ranking Heart Toxicity
To assess the potential risk of heart toxicity for drugs in development, pharmaceutical companies use laboratory tests involving animals (usually rats or mice) or cells from animals or humans that are engineered to artificially express heart-related genes. Drug candidates that appear to have an acceptable balance of benefits and risks typically proceed to testing in human clinical trials.
But there can be biological differences between these existing models and humans, so non-clinical lab tests can have significant limitations, explained Dr. Wu.
Currently, “the first time humans are exposed to a new drug is during clinical trials,” he said. “We think it would be great if you could actually expose patients’ heart, brain, liver, or kidney cells to a drug” in the lab, prior to clinical treatment, allowing researchers to determine whether the drug has any toxic effects.
Dr. Wu, a cardiologist by training, studies toxicities cancer drugs cause in heart cells. Human heart muscle cells (called cardiomyocytes), however, are hard to obtain—requiring risky heart surgery that may be of no direct benefit to the patient—and are notoriously difficult to grow in the lab.
As an alternative, researchers have developed a method to produce heart cells from human induced pluripotent stem cells (hiPSCs). hiPSCs are created by genetically engineering normal human skin or blood cells to express four specific genes that induce them to act like stem cells. Chemical treatments can prompt hiPSCs to develop into “mature” cell types, such as heart muscle cells.
A large body of research has established that human adult stem cell-derived heart cells, which function and grow in cell culture, can be used as an initial model to screen drug compounds for toxic effects on the heart, said Myrtle Davis, Ph.D., chief of the Toxicology and Pharmacology Branch of NCI’s Division of Cancer Treatment and Diagnosis, who was not involved in the studies.
For the Science Translational Medicine study, Dr. Wu and his colleagues set out to determine if a panel of human stem cell-derived heart cells could be used to evaluate the heart toxicity of 21 different FDA-approved TKIs.
They generated hiPSC-derived heart endothelial, fibroblast, and muscle cells from 13 volunteers: 11 healthy individuals and 2 people with kidney cancer who were being treated with a TKI. Using drug concentrations equivalent to what patients receive, the investigators next determined how lethal each TKI was to the heart cells.
They found that several TKIs were very lethal to endothelial, fibroblast, and heart muscle cells from all 13 individuals, while others were more benign.
Stem cell-derived heart muscle cells grown in a dish spontaneously contract as a beating heart does, so the researchers also analyzed the effects of TKIs on the cells’ beat rate, or contractility. They found that several TKIs altered the cells’ beat rate before they were killed by the drug treatment. If severe enough, an irregular heartbeat (called an arrhythmia), can disrupt normal heart function.
From these lethality and contractility experiments, the team developed a “cardiac safety index,” a 0-to-1 scale that identifies how toxic a TKI is to heart cells (with 0 being the most toxic). They then used the index to rank the 21 TKIs. The control treatment scored a 1, while a few TKIs that are labeled by the FDA with boxed warnings for severe heart toxicity scored close to 0.
“Safety indices like this one can be very useful during drug discovery,” said Dr. Davis, and the applicability of the index developed by Dr. Wu and his colleagues “will become clear when they evaluate its performance with more compounds.”
And for the safety index to be applicable to more patients, the panel of cells used to develop it would need to be gathered from a “sufficiently representative” population of people reflecting different ages, races/ethnicities, health statuses, and other characteristics, said Lori Minasian, M.D., deputy director of NCI’s Division of Cancer Prevention, who was not involved in either study.
For example, “the study did not include cells derived from patients with [pre-existing] cardiac disease,” said Dr. Davis.
A Personalized Approach
In addition to their potential application during drug development, Dr. Wu believes that stem cell-derived heart cells could potentially be used to predict toxicity risk for individual patients. He and his colleagues explored this possibility in their Nature Medicine study.
Doxorubicin, used on its own or in combination with other drugs, is an effective treatment for breast cancer and several other types of cancer. Like TKIs, however, it is known to cause heart toxicities, such as arrhythmias and heart failure, in a small proportion of patients. But there has been no way to predict which patients will experience these side effects.
The researchers developed stem cell-derived heart cells from eight women with breast cancer who had been treated with doxorubicin—half of whom experienced cardiotoxicity from the treatment and half who did not.
In several different lab tests, the heart cells from women who had experienced cardiotoxicity were more sensitive to doxorubicin than those from women who had not. More specifically, in heart cells from women who had experienced cardiotoxicity, doxorubicin treatment caused more severe irregularities in cell contractility, and even low concentrations of the drug killed the cells.
An Improved Model
While the stem cell-derived heart cell model “may be an improvement over the current [drug testing] system, it’s not perfect,” said Dr. Minasian. For example, the model does not capture contributions of other organs and cells to the toxic effects of a drug, she explained. The drug may be broken down in the liver, for instance, and side products (called metabolites) may also cause toxic effects.
In addition, the lab-grown stem cell-derived version of someone’s heart cells are “not going to be exactly the same” as the cells found in that person’s heart, Dr. Wu noted. “Nevertheless, they reflect the same genetics and they are pretty good at predicting drug response,” he said.
Looking forward, Dr. Minasian said, “figuring out how to best use this approach is going to take more work, but being able to better predict human response [to cancer drugs] is important.”
The research team’s next steps include conducting prospective studies to determine whether they can use a patient’s stem cell-derived heart cells to potentially predict if that person will develop heart toxicity before they actually receive cancer treatment.
Sharma, A., Burridge, P. W., McKeithan, W. L., Serrano, R., Shukla, P., Sayed, N., ... & Matsa, E. (2017). High-throughput screening of tyrosine kinase inhibitor cardiotoxicity with human induced pluripotent stem cells. Science translational medicine, 9(377), eaaf2584.