Profiles in Cancer Research
Dr. Bryce Reeve
Psychometrician and Program Director, Outcomes Research Branch, NCI Division of Cancer Control and Population Sciences
Not everyone comes to cancer research via the usual path. Dr. Bryce Reeve is a native of Chapel Hill, NC, has a Ph.D. in psychometrics from “the University,” and insists with pride that he “bleeds Tar Heel blue.” When you hear that he was a math whiz in high school and see him now—revolutionizing the field of assessing quality of life and adverse events with an esoteric brand of modeling called item response theory—you may not think there are too many dots to connect. But you would be wrong.
His story emerged through the lens of an intensely personal experience. He expressed his frustration from a recent visit back to Chapel Hill, where the mother of a childhood friend is home after surgery for pancreatic cancer.
“Despite the medications and surgery, she’s still in excruciating pain,” said Dr. Reeve. “On one side of the scale is a mountain of scientific progress in cancer treatment. On the other is this wonderful woman, enduring nobly but ravaged by her cancer, without much hope. She’s not a complainer,” he explained, “but she still should be able to communicate her symptoms and experiences. I think we need to provide patients a voice, a language to express what they’re going through—an expectation that their experience is not just a disembodied ‘side effect’ to be listed on a chart while they face the often-grisly reality alone.”
That is precisely what his work at NCI is about.
When Dr. Reeve headed to engineering school at Virginia Tech in the late 1980s, he was fully prepared to work 60 hours a week to keep up with that regimented discipline. What he didn’t expect was that something as nonmathematical as his own quality of life might interfere. “It was all science and math, and I just wasn’t happy with my life,” he recalls. He dropped out, and for the next 5 years he paid his bills by taking reservations by phone for the airline industry.
He eventually finished his undergraduate degree by taking night courses at the University of North Carolina at Chapel Hill (UNC). In the process he fell in love with psychology. Fascinated by what he was learning about the science of human personality and behavior, he found a way to combine it with math in a graduate program in psychometrics at his beloved UNC. His Ph.D. in 2000 coincided with NCI’s creation of an Outcomes Research Branch (the first at NIH), with Dr. Joseph Lipscomb as its first chief.
“When Joe interviewed me for this job, I told him that all I knew about cancer was that nobody wants it. Finding a mentor like him was the luckiest thing for me,” said Dr. Reeve. “The best minds in the field are now friends and colleagues. My learning curve looks like a rocket trajectory.” And the timing was right, but only when you count those 5 “lost years trying to find my way.”
“We were the lucky ones,” said Dr. Lipscomb, who is now Professor and Georgia Cancer Coalition Distinguished Cancer Scholar at the Emory University School of Public Health. “You just get an instinct about someone. Bryce was out on the leading edge of what we call the ‘modern’ school of measurement approaches. Even on our first phone conversation, he was obviously very bright and quick-witted, and we were hoping he could help us jump the cancer research measurement field into the future.”
Reporting the adverse events (AE) that cancer patients experience has become a critical aspect of clinical trials, but it wasn’t always so. Until the 1970s, AE reporting in retrospective studies was limited, often ignoring the severity and nuance of what patients were experiencing. As clinical trials began to assume a more central role, the process became more formalized with safety (toxicity) profiles, standardized terminology, and a grading system.
The current system in oncology, devised by NCI, is the Common Terminology Criteria for Adverse Events (CTCAE), used throughout the United States and beyond to report and grade both acute and late effects from therapeutic oncology interventions, including surgery, radiation, and chemotherapy.
“But we have a long way to go,” said Dr. Reeve. “It’s not only clinical trials, but patients everywhere in the health care delivery system. We make the physician the gatekeeper and rely on them to communicate with patients about any symptoms they may be experiencing. But it’s a huge burden for oncologists,” who are often focused on treatment.
“Finding better ways to capture what it’s really like to live with cancer, day-to-day, is what drives me,” he explained. And after a decade of such work, he is now recognized as one of the leading experts on an idea whose time has arrived: patient-reported outcomes (PRO). This approach incorporates the patient’s perspective more directly in assessing health-related quality of life (HRQOL), and can even help patients ascribe personal meaning to the burden of cancer.
PROs would prove to be the springboard, and Dr. Reeve’s contribution was to weave item response theory (IRT) modeling into the art of devising questionnaires that would more clearly capture a patient’s HRQOL. IRT is based on the idea that how a person will respond to a question depends on his or her personal experience of the HRQOL domain being measured by the questionnaire. For example, asking: “Do you feel exhausted?” is an indicator of fatigue reported by cancer patients. Those with severe fatigue are more likely to answer “Yes,” while those experiencing mild to moderate fatigue are more likely to answer “No.” The application of IRT models combined with qualitative methodologies enhances a questionnaire’s ability to capture a cancer patient’s experience with minimal response burden. “Throughout the past decade, IRT methods have increasingly been used to great effect in PRO assessment,” said Dr. Reeve.
“Of course a patient’s disease status matters,” he explained, “but cancer is a journey. Every patient comes to see things differently, re-evaluate which daily activities mean the most, and ultimately to find a new set of priorities, if not values. We can’t overlook the human cancer patient in our crusade to cure the disease. With these new ways of specifying—and honoring—their experience, I think we’re on the right track. It’s really a privilege to do this work.”
In 2006, Dr. Reeve organized a major conference at NCI that produced a consensus: For subjective symptoms, the patient’s own account should be considered the gold standard. Entire domains of oncology, such as survivorship, palliative care, and quality of care, are being reinvigorated with just such a patient-centric emphasis. “We’re really beginning to redefine the basic model of care,” Dr. Reeve said.