How to Use Gut Bacteria to Make Chemo Safer
Capecitabine, a chemotherapy drug, changes the gut microbiome (the ecosystem of microbes in the gut). It is found to increase the number of vitamin K2-producing E.coli (a type of bacteria), which may be associated with protecting the gut microbiome and relieving chemotherapy side effects..
By Scarlett Yang
What is the study about?
Chemotherapy uses strong drugs to kill or slow the growth of cancer cells. It has extended and saved many lives, but is also notoriously known for side effects, such as nausea, vomiting, hair loss, mouth sores, and nerve damage. These side effects can sometimes become so strong that doctors even have to reduce, pause, or even stop ongoing treatments.
Doctors try various strategies to combat the side effects, but the options are still limited. For instance, some drugs can reduce nausea and vomiting, but they may trigger additional side effects[1]. There are also non-drug approaches like lifestyle changes, but their effectiveness largely varies from person to person [2,3]. The need for safer and more reliable solutions to chemotherapy side effects raises important questions: Are there processes inside our bodies that determine how well we respond to the drugs? Can we borrow these processes to deal with the side effects more effectively? Can we predict how a patient will react to a drug even before the chemotherapy starts?
Recent research suggests that the answer may be yes. One study conducted by the UCSF Turnbaugh Lab discovered an unexpected relationship between our gut microbiota [the ecosystem of microbes that live in your intestines] and patients’ ability to tolerate chemotherapy.
How was it conducted?
This study focuses on a type of chemotherapy drug called capecitabine (CAP) [used to treat cancers that develop in the large intestine and often cause side effects]. The researchers started by looking at the changes in the patients’ gut microbiome triggered by CAP treatment. This is done through the following steps:
Collecting stool samples and tracking patient symptoms: The researchers chose 56 patients who had colorectal cancer and were scheduled to start three 3-week cycles of CAP treatment. The patients collected and stored their own stool samples at three time points: before treatment, in the middle of cycle 3, and 1 week after treatment finished. Throughout the study, they also reported any side effects they experienced by filling in questionnaires.
Analyzing the samples: The scientists extracted DNA from the collected stool samples. Then, they used a powerful tool called metagenomic sequencing. Imagine you have a cake with various flavors mixed together, perhaps chocolate, strawberry, and mango, and you want to know the exact flavors. Then, you blend the cake into a smooth texture and then taste it. Scientists use metagenomic sequencing in a similar way that this tool “tastes” the blended genetic material and figures out the microbial flavors. By using metagenomic sequencing, the researchers got a genetic map of the microbes in the stool samples. They revealed not only what bacteria were present but also what their functions were.
Finally, the researchers used machine learning models [a way to teach a computer to learn from examples and experiences, just like how people learn]. They trained the computer to find the patterns in the data of the patients’ gut microbiome and then predict future treatment outcomes of certain drugs, like side effects and dosage changes.
What did they discover?
The first big discovery was that CAP treatment increased the number of gut bacteria that carried the gene coding for vitamin K2. This means that the type of bacteria that could produce vitamin K2 became more common. It seemed that CAP treatment created a gut environment advantageous for the survival of such vitamin K2-producing bacteria.
This suggested that K2 might protect the gut from toxicity associated with CAP chemotherapy. With this crucial clue, the researchers then showed that, indeed, vitamin K2 could directly protect gut bacteria from the toxic damage of CAP. More importantly, such protection also applied to the entire organism. In the patients, those with a lower amount of vitamin K2-producing bacteria were more likely to suffer from nerve damage, a painful side effect mentioned earlier. Researchers also confirmed this in mice. After the mice were treated with CAP and vitamin K2, the researchers surprisingly saw that they were protected from nerve damage. That is to say, vitamin K2 acts like a shield that protects both the gut bacteria and the host from chemotherapy toxicity.
The researchers found that the gut microbiome could predict chemotherapy side effects. They trained the model to use data about patients’ gut microbiome to predict how well they can withstand chemotherapy (like what side effects and dosage changes may occur in the future). With high accuracy, the model successfully predicted which patients would go on to suffer severe side effects during cancer treatment.
Why does it matter?
Chemotherapy side effects have long been intimidating. Toxicity not only brings about both physical and mental discomfort, but also compromises the drugs’ lifesaving effects. However, with this new finding in the relationship between the gut microbiome and chemotherapy, scientists may find new ways to tackle the problem.
Firstly, it provides new perspectives to manage chemotherapy side effects. There is no current treatment to prevent nerve pain from chemotherapy. However, this study shows that vitamin K2 may be related to nerve side effects. It sets a strong start and foundation for the subsequent study of the effectiveness of simple vitamin K2 supplements. If follow-up studies are successful, this preventive measure could be safer and more affordable for many cancer patients. In addition, the research may only be the tip of an iceberg. It encourages scientists to discover new preventive pathways, applying them to the development of novel medications that alleviate chemotherapy side effects.
Besides, by improving the predictive model, scientists may help doctors to design cancer treatments in a more personalized way. Currently, it is difficult for doctors to tell how patients will react to chemotherapy. But imagine that one day, a simple sample from the patient can change that. The machine learning model would predict the possible reactions of the patient and dosage adjustments. In this way, patients will avoid the worst side effects from the start. Cancer treatment will thus become more effective, acceptable, and humane.
A link to the full publication can be found here: https://pubmed.ncbi.nlm.nih.gov/40391895/