Center News

Custom chemo via chemistry

March 6, 2003
Dr. George McDonald and Research Nurse Michelle Bouvier

Dr. George McDonald (left), head of Gastroenterology/Hepatology in the Clinical Research Division, discusses patient reactions to cyclophosphamide with research nurse Michelle Bouvier.

Photo by Todd McNaught

Cancer patients embarking on a course of high-dose chemotherapy steel them selves against the side effects: nausea, hair loss or even organ damage.

Yet among those who receive identical drug doses, only some will suffer all of these symptoms. Other patients - who count themselves among the lucky - sail through therapy with a minimum of serious complications.

Doctors now say that chance has little to do with such disparity. Instead, what predicts the occurrence and severity of a drug's side effects is an individual's genetic blueprint, said Dr. George McDonald, head of Gastroenterology/Hepatology in the Clinical Research Division.

Led by McDonald and Drs. John Slattery, Ted Gooley, and Claudio Anasetti, also of Clinical Research, a study published in the March 1 edition of Blood finds that innate differences in how patients metabolize a common transplant chemotherapy drug have a profound effect on whether an individual suffers liver damage, a significant cause of transplant-related complications and death.

With these findings in hand, the researchers are developing modified dosing regimens for the drug, called cyclophosphamide, that ultimately could reduce such mortality by as much as 20 percent.

Unexpected toxicity

Although their analysis was limited to patients receiving cyclophosphamide (also known as Cytoxan) and total body irradiation in preparation for stem-cell transplant, the scientists speculate that similar phenomena may explain unexpected toxicity from chemotherapy agents among patients with many other cancers.

"When chemotherapy is given to patients with cancer, some percentage have bad reactions," McDonald said. "In past years, we put this to bad luck. But it's clear that there is a biologic reason, a very good biologic reason, for bad reactions to cyclophosphamide."

Ultimately, he said, doctors hope to tailor a drug dose to a patient's unique capabilities for breaking down the medication, rather than the standard practice of dosing according to body size. Ideally, doses could be determined before a patient is ever given the drug.

"The holy grail would be to give patients a genetic test that would enable us to provide them with the optimal exposure to the drug based on how their metabolic enzymes break it down, which in some individuals leads to the production of harmful intermediates."

The results of their study are the culmination of a three-year effort that required a team of patient-care nurses, a research nurse, laboratory technicians, statisticians, graduate students and work-study students, many of whom worked around the clock to collect and analyze the chemical composition of patient blood samples.

In addition to the study team, McDonald expressed gratitude to the patients, whose consent to have blood samples drawn did not result in direct benefit to themselves but likely will improve the quality of life and survival of future transplant recipients.

"This was altruism at its finest," he said. "Without patient participation, this observational study would not have been possible."

The study followed 147 Fred Hutchinson patients who underwent hematopoietic (blood) stem-cell transplants between 1997 and 2000 as part of a treatment protocol overseen by Anasetti.

Dose based on body size

Prior to transplant with tissue-compatible stem cells, patients received intravenous cyclophosphamide for one to two hours each day for two consecutive days prior to undergoing total-body irradiation (a standard preparative regimen). The cyclophosphamide dose was calculated based on body size.

Blood samples were collected from patients prior to and at different time points during and after cyclophosphamide infusion, frozen and then analyzed in Slattery's laboratory for metabolic breakdown products of the drug. Kalhorn, a staff scientist, developed tests for quantifying the metabolites.

The researchers correlated levels of each chemical intermediate with the presence or absence of liver toxicity in the patients. Of the 147 patients in the study, 38 percent suffered liver disease. Among those with liver complications, 18 percent had severe disease, 23 percent had moderate disease and 59 percent had mild disease.

Following analysis of the first 50 patients, a chemical marker - a compound derived from cyclophosphamide known as O-carboxyethyl-phoshoramide mustard, or CEPM, for short - emerged as the most likely indicator of liver toxicity. Based on what was known about CEPM, Slattery said the finding was puzzling.

"When George showed me the outcome data from the first group of patients, I thought there was a mistake," said Slattery, also a UW professor of pharmaceutics. "Increasing levels of CEPM were clearly associated with increased liver toxicity, yet this chemical product is not known to be toxic, nor did we find it to be toxic in any of our laboratory analyses."

A closer look at the chemical pathways resulting from cyclophoshamide's breakdown in the body suggested that the high levels of CEPM might be a "reporter" for a different toxic metabolite in another branch of the breakdown pathway. Animal studies by Slattery and Ruolun Qiu, a UW graduate student, lend support to this hypothesis.

"It's exciting to find a novel aspect of cyclophosphamide disposition that is key to its toxicity in this setting," Slattery said. "This is an ancient chemotherapy drug, so to discover something new about how it is eliminated was very satisfying."

Based on the results, the researchers have designed a new clinical trial in which stem-cell transplant patients receiving cyclophosphamide and total body irradiation will have their blood chemistry analyzed after their first treatment with cyclophosphamide. Depending on their CEPM levels, the second day's dose will be adjusted accordingly. This dose-adjusting approach is now common with busulfan, another chemotherapy agent commonly used in preparative regimens.

"Forty years ago, when the field of cancer chemotherapy was new, optimal drug doses were determined by slowly escalating the dose in groups of patients and then backing off when too much toxicity was seen," McDonald said. "But even with that approach, there's always a subset of patients who do badly. By tailoring the dose to an individual's metabolism, we hope to be able to ensure that the drug is at a high enough level without causing excess toxicity."

Possible decrease in deaths

McDonald speculated that if effective, the modified cyclophosphamide drug dosing they propose could result in a 20 percent decrease in deaths following transplants that used the cyclophosphamide and total body irradiation regimen.

With an eye toward developing a genetic test to determine susceptibility to cyclophosphamide toxicity, McDonald and Slattery have begun collaborations with Dr. Maynard Olson and colleagues at the UW Genome Center to begin to identify gene variants that are associated with differences in the drug's metabolism.

McDonald said the study would not have been possible without grants from the National Cancer Institute that foster collaboration among scientists with diverse strengths.

"My focus is on complications from cancer therapy, John is a pharmacologist who looks at a drug and assesses its outcome in therapy, Claudio is an oncologist, and Ted Gooley is a statistician whose data analysis was key to our findings," he said.

"This kind of study is exactly why we all exist at the same institution."

Fred Hutchinson Cancer Research Center is a world leader in research to prevent, detect and treat cancer and other life-threatening diseases.