Chenqi Fu, a graduate student from Penn State College of Medicine’s doctoral program in Biostatistics, won three honors for her research, which is aimed at improving data synthesis.
Chenqi Fu, a graduate student from Penn State College of Medicine’s doctoral program in Biostatistics, won three honors for her research, which is aimed at improving data synthesis.
February 28, 2022
The American Association for the Advancement of Science selected Fu as a finalist in their 2022 Student E-poster Competition for her research, “Evidence synthesis with reconstructed survival data.” The association announced the winners virtually on Feb. 20.
In addition, Fu’s research has been recognized by the International Biometric Society Eastern North American Region (ENAR) and American Statistical Association Health Policy Statistics Section (HPSS). Fu will receive a Distinguished Student Paper Award and be recognized during ENAR’s Spring Meeting in Houston March 27 to 30. Later in the year, she will receive a HPSS 2022 Student Paper Award and be honored during their Joint Statistical Meeting in Washington, D.C., Aug. 6 to 11.
Fu has introduced a new approach to synthesizing evidence that could help advance clinical research and assist in regulatory decisions with more informative inference on various types of diseases.
In pharmaceutical development, meta-analysis has become an essential tool to integrate the clinical evidence needed for approval by the U.S. Food and Drug Administration. Traditionally with meta-analysis, the standard approach for survival outcome relies on hazard ratio estimates, which gauge the relative risk of a hazardous event occurring. However, data can be deemed deficient when no hazard ratio is reported or if the proportional hazards assumption are violated.
Fu’s research goes beyond prior studies and proposes a new framework — meta-analysis of reconstructed survival data (MARS) — to estimate the hazard ratio and survival probabilities. To help respond to the Food and Drug Administration’s concerns on developing drugs for hematologic malignancies that affect the blood, bone marrow and lymph nodes, the researchers used MARS in a meta-analysis to examine the association between measurable residual disease (MRD) and disease-free survival in patients with acute myeloid leukemia.
After completing cancer treatments, disease-free survival refers to the length of time that patients experience no signs or symptoms of the disease. According to the researchers, the findings suggests that the MRD negativity is associated with superior disease-free survival.
“MARS provides a powerful and comprehensive tool for meta-analysis because it enables scientists to integrate more studies and more types of aggregate data and disseminate results in alternative ways with richer information,” said Fu, who was lead author on recent work published about the approach.
Compared with the standard method, MARS significantly reduces selection bias. Through the simulation study, the researchers demonstrated that MARS provides a valid estimation of treatment effects in the form of various outcome measures.
“Meta-analysis is important in various scientific areas, such as astronomy, psychology, social science and biomedical science because it combines information from multiple sources in the current big data era,” said Shouhao Zhou, corresponding author and Fu’s advisor. “Chenqi’s work overcomes one of the major challenges in meta-analysis by providing a path-breaking gateway and an effective tool to enhance evidence synthesis for better medical practice.”
Researchers from the University of Texas MD Anderson Cancer Center, including Xuelin Huang and Donald Berry from the Department of Biostatistics, and Drs. Nicholas Short and Farhad Ravandi from the Department of Leukemia contributed to this research.