Scarcely a week goes by without news of the sobering state of the global HIV/AIDS epidemic and the challenges faced by those who work to end it. Of the roughly 38 million people living with HIV today, almost 5 million became infected in 2003 alone — yet experts estimate global spending on AIDS-vaccine research for 2003 was $650 million, less than 1 percent of the total spending on all health-product development. There is increasing pressure to advance vaccine research more quickly, but also with more accuracy.
Among those working to carry out this goal are biostatisticians like Dr. Li Qin, a new member of the Statistical Center for HIV/AIDS Research and Prevention (SCHARP). With expertise in designing and interpreting the results of large-scale HIV-vaccine trials, Qin and her SCHARP colleagues play a key role in propelling the most promising vaccine candidates from discovery to testing. One of the major objectives of their work is to ensure that precious resources are used efficiently in the search for an effective vaccine. Although HIV/AIDS treatment options are improving, many experts believe that vaccines are the only way to ultimately gain the upper hand on a disease that claims generations of people during their most productive years of life.
Qin, who joined SCHARP as a faculty statistician last August, will apply her data-analysis skills to early stage clinical trials of HIV-vaccine candidates being conducted by the HIV Vaccine Trials Network (HVTN). The world's largest program for developing and testing HIV-preventive vaccines, the HVTN is one of several large-scale international AIDS-prevention projects at the center that SCHARP supports.
Researchers seek to better understand how vaccines offer immunity to the virus, improve HIV resistance or even simply to slow the progress of the infection.
"If we can learn the mechanisms by which a vaccine we are testing is working, this would help us to improve it," Qin said. She admits, in the quest for an effective HIV vaccine, there is a lot to learn. "People are all learning from the process, and as we learn, we will eventually achieve our goal."
Qin assesses how well a vaccine protects against infection by correlating a vaccine recipient's immune responses to their future rates of infection with HIV.
Dr. Peter Gilbert, one of Qin's colleagues at SCHARP, said, "Li's work includes methods for modeling the profiles of immune responses in individuals over time. The vaccine candidates showing the strongest and broadest immune responses are advanced to Phase III HIV-vaccine trials that test its effectiveness in thousands of volunteers."
Qin's techniques help investigators make a critical distinction when evaluating vaccines: whether immune responses are a direct result of the HIV-vaccine candidate or are immune responses inherent to the individual and not caused by the vaccine.
The ideal outcome of an effective vaccine is, of course, to prevent infection. But another important outcome, Qin said, is to slow the progression of the disease in those who become infected with HIV. Part of her research will involve analyzing data from those who do become infected. This will allow scientists to correlate an individual's immune responses to the vaccine with how much virus is present in their blood and how quickly they progress to needing anti-HIV drugs.
During a vaccine trial, statisticians use a variety of methods to analyze data that has been collected through a sequence of measurements over time. These tools reveal patterns or trends in the data, which then enable statisticians to draw conclusions and, sometimes, even predict future values or effects.
New models, estimate procedures
When existing tools for complex analysis aren't sufficient, statisticians like Qin must develop new ones, which is exactly what she did to earn her doctorate in biostatistics and epidemiology from the University of Pennsylvania last June. There, she developed new models and estimation procedures to improve the analysis of complex data and make the process much more efficient.
For instance, a computer using similar existing algorithms would have to make 1,000,000 calculations for a data set with 100 data points before any patterns could be spotted. Qin's problem-solving tools require the computer to make only 100 calculations for the same-size data set. In studies with thousands of participants each providing dozens or hundreds of data points in the form of immune-response measurements, such a paring down can save enormous amounts of time and computer storage space.
Among the projects to which she applied her work was an examination of the electroencephalograms (EEGs) of epilepsy patients for patterns that might indicate approaching seizures. Qin's analyses pinpointed time periods and areas of the brain in which changes in activity might serve as warning signals. That information enables physicians to design more targeted experiments to enhance treatment options.
Qin's career choice stemmed from what she describes as "family influence and personal interest." Her mother is a professor of statistics in China, and Qin's sister is a biostatistician. Throughout her education, Qin enjoyed math, and especially its real-world applications. Her pursuits took her from Peking University to the Gallup Organization in Beijing, China, and eventually to the University of Pennsylvania.
Along the way, she gained an appreciation for the kind of challenging and motivating academic research environment she found when she first visited the center and SCHARP. The prestigious, joint Fred Hutchinson and University of Washington biostatistics program and its faculty attracted Qin, as did SCHARP's ambitious research scope.
"From the introduction I received from Peter Gilbert and Steve Self (both faculty statisticians in SCHARP) about the center's HIV-vaccine research, I became very interested," she said. And she has found SCHARP to be the supportive and stimulating research environment in which she had dreamed of working. "There are always new projects and new challenges waiting for me," she says, "and that can make my life pretty exciting."
Qin plans to make her own contributions to the statistical literature and the field of global health. She hopes to one day apply her skills and expertise to public- health issues that are particularly important in Asia, including HIV/AIDS treatment and prevention. "Asia has a big population, and the density of the population is also very high, which makes transmission of HIV higher," she said. In fact, there are an estimated 7.4 million people living with HIV in Asia, a region home to 60 percent of the world's population. According to reports, the epidemic in Asia could be the worst yet if infection continues to rise at its current rates.
Qin is also looking forward to collaborating with other scientists throughout the center. "Nowadays many different areas of research need biostatistics. Usually what appeals to me first is the scientific problem, and then I try to formulate it into a biostatistical problem. These two parts cannot be separated, and I think this is the beauty of biostatistics."