Research published Thursday shows that a common genetic disorder ― one that many people don’t even know they have ― is linked to a more than twofold increase in death rates among patients treated with a particular cancer drug.
Scientists estimate that between 3 percent and 10 percent of people worldwide have Gilbert’s syndrome, which alters the way the liver processes one of the byproducts that results from the body’s recycling of dead red blood cells. Gilbert’s typically causes no ill effects; in fact, it’s even been linked to long life and good health in the general population.
This study found, however, that what typically seems to be a Dr. Jekyll turns into a Mr. Hyde for patients who receive a chemotherapy drug called busulfan, which is commonly used as part of a chemotherapy regimen prior to bone marrow transplantation. Bone marrow transplantation is a standard therapy for people with serious blood disorders like advanced leukemia.
The research overturns decades of conventional wisdom in the transplant field about the importance of Gilbert’s [“zhil-BARE’s”] syndrome, said Dr. George McDonald, a lead researcher on the study.
“In the medical textbooks, it’s always said to be totally benign, nothing bad ever happens,” said McDonald, a clinical researcher at Fred Hutchinson Cancer Research Center. This study’s results “came out the total opposite of what I expected.”
This study is just the latest example of how variations in metabolism from patient to patient can have huge impacts on the toxicity or effectiveness of a drug treatment, McDonald said.
“People are born with certain hair color, certain skin color, certain eye color — and certain metabolizing enzymes. You can tell people’s hair color … But you can’t tell what their enzymes are,” he said. “Right now, we assume every human being is going to metabolize drugs in the exact same way. But we know that that’s not true.”
McDonald’s hope is that this study, an analysis of 20 years of data on several thousand patients who were transplanted at via the Fred Hutch/University of Washington Cancer Consortium, will swiftly be validated and implemented to save lives.
“Like all science, it needs to be replicated,” McDonald said. “If it’s replicated with the same dimensions of risk, it’s something that should be universally applied in practice.”
A surprising and mysterious result
Gilbert’s syndrome typically has no outward signs unless a person is under particular physical stress, when they can develop mild jaundice, tiredness or abdominal pain. It’s detected via a blood test that measures levels of bilirubin, the blood-cell-breakdown byproduct whose processing is affected by the genetic disorder. McDonald estimates that about half of people with Gilbert’s are unaware of it; most people have no reason to be.
This isn’t the first study to show that Gilbert’s syndrome can affect how a drug is processed, or metabolized, by the body. Over the last couple of decades, reports have come out about patients with Gilbert’s syndrome who’ve experienced toxic side effects after taking certain drugs that are metabolized by the same enzymes affected by the syndrome.
These reports began to make McDonald wonder about the chemotherapy drugs that patients receive to prepare their bodies for transplantation.
“I’m a big believer in what’s called evidence-based medicine,” McDonald said on a recent day in his office at Fred Hutch in Seattle, behind a desk thoroughly covered in scientific papers.
McDonald was in the middle of moving offices, and a giant canvas mail cart alongside his desk was two-thirds full with outdated textbooks, now destined for the trash, that he collected over his career, which has spanned more than four decades.
For many of those years, McDonald was one of just a handful of people whose research was focused on gastrointestinal and liver complications of transplantation.
With this study, McDonald said, “I just wanted to see if the advice we’d been giving to patients for 40 years really had a factual basis when it came to people getting very high-dose chemotherapy.”
He felt confident. All available evidence on these drugs indicated that they were not processed by the metabolic mechanisms affected by Gilbert’s syndrome. If anything, McDonald reasoned, patients with Gilbert’s syndrome would experience fewer toxic side effects from these chemotherapy drugs than everyone else: Because of Gilbert’s effects on bilirubin-processing enzymes, these people have a higher blood concentration than normal of the preprocessed form of bilirubin ― which just so happens to be an antioxidant, a substance that protects cells from damage. (This antioxidant effect, researchers suspect, could be the reason why people with Gilbert’s tend to live longer and have lower rates of cardiovascular and lung disease than the rest of us.)
To find the answer to his question, McDonald and colleagues turned to a massive database dubbed Gateway, a compendium of medical data from patients treated by Fred Hutch and its consortium partners over 30 years. (The patients had all previously consented for their information to be used anonymously in research.)
Statistician Dr. Ted Gooley of Fred Hutch crunched the numbers from almost 3,500 transplant patients seen between 1991 and 2011, including more than 200 patients the researchers identified as having Gilbert’s syndrome based on records of their blood test results.
What came out of their analysis is rare in science, McDonald said ― a result that is dramatically different from what the researchers expected going into the study. In the first 200 days after transplant, patients with Gilbert’s syndrome who also received busulfan prior to transplant were more than twice as likely to die of any cause than all other patients, and they were nearly three times as likely to die of a cause not related to disease relapse. Gilbert’s seemed to have no effect on the outcomes when any other type of chemotherapy was used.
McDonald pulled in longtime collaborator Dr. Jeannine McCune of Fred Hutch to help figure out why. For years, McCune has been studying how busulfan and other drugs used in cancer treatment are absorbed, distributed, metabolized and eliminated by the body ― a field known as pharmacokinetics ― with the goal of reducing drugs’ toxic side effects and increasing their effectiveness.
In her lab, she analyzed the pharmacokinetics of the drug between patients with and without Gilbert’s. But she did not see any telltale differences in the way their bodies processed the drug that might have indicated a reason for the differences in death rates. Neither could the research team find any difference in the causes of death between the two groups of patients, which might also provide hints as to how the syndrome was interacting with the drug.
“We’re left with a bit of a mystery as to why the striking finding has occurred. But sometimes you don’t come up with a ribbon on the package, where you tie everything together,” McDonald said.
“When you have findings like this, it always leads to another question.”
Tackling the next question
McCune is initiating more experiments in her lab to try to figure out what they might have missed in this study. Perhaps after the body breaks down busulfan, those breakdown products interact with the bilirubin-processing pathway altered by Gilbert’s, she hypothesizes.
Until the finding is replicated by other bone marrow transplantation groups around the world, the research team recommends that transplant physicians carefully consider the use of busulfan in patients with Gilbert’s syndrome. (About 40 percent of transplant patients at Fred Hutch currently receive the drug.)
“If I have a patient with Gilbert’s and I have a choice of similarly effective drugs, with [a recommendation for] caution, I might use the other regimen,” McDonald said.
Another cancer drug whose metabolism is affected by Gilbert’s is irinotecan, which is commonly used to treat advanced colon cancer. McDonald emphasized that every new cancer patient should already be receiving extensive blood work that will reveal this often-hidden genetic condition. But the presence of Gilbert’s syndrome will not necessarily raise a red flag for many oncologists as they determine a patient’s treatment plan, he warned, given the long-standing conventional wisdom about the condition.
“In the oncology setting, it wouldn’t be unreasonable for patients scheduled to have some form of chemotherapy, who knows they have Gilbert’s syndrome, to ask their oncologist, ‘Are the drugs you’re planning to give me affected by my Gilbert’s syndrome?’” McDonald said.
Patients being treated with busulfan through the Fred Hutch/UW Cancer Consortium already have a system in place to lower the risk of getting dangerously overdosed or underdosed with this drug — a clinical lab established by Fred Hutch researchers years ago that tests patient blood samples early in treatment to learn how the medication is being metabolized, allowing the physician to lower or raise subsequent doses.
McDonald stressed that he doesn’t see any reason why people who are healthy should go out and get tested to see if they have Gilbert’s syndrome — especially considering that its protective effects could result in them living a longer and healthier life. And McCune noted that most people will have had their bilirubin levels tested for one reason or another by adulthood.
“Patients should not be freaked out,” she said. Everyone has different factors that alter the toxicity or efficacy of certain drugs, she explained. “No drug is benign. Not to be fatalistic, but it’s just reality. Medications have done a lot for us over the years; we’re just trying to use them to the best of our ability.”
To this end, McCune stressed the importance of research.
“Really, while it is scary to think that we might all have a genetic factor that puts us at higher risk of a drug toxicity or for the drug to not work as well, the important thing is for patients to participate in research so that we can learn as much as we can from each patient to help the next patient,” she said.
Since McDonald started in the field of bone marrow transplantation, at a time when it was an experimental, dangerous and last-ditch procedure, he has seen how studies like this one have added up to dramatic improvements.
“That’s how progress has been made over 40 years. Seldom do you have a moment where it’s an ‘aha’,” McDonald said. “It’s been a series of tiny little dots being connected to one another.”
Talk about this story on the Fred Hutch Facebook page.
Susan Keown is a staff writer at Fred Hutchinson Cancer Research Center. Before joining Fred Hutch in 2014, Susan wrote about health and research topics for a variety of research institutions, including the National Institutes of Health and the Centers for Disease Control and Prevention. Reach her at skeown@fredhutch.org or follow her on Twitter at @sejkeown.
