Postdoctoral Research Associate - Statistical Methods for Pediatric Oncology Clinical Trials

Location: Memphis, TN | Department: Biostatistics | Full-time, 40 hours/week, weekday days | Requisition: JR7572

About the role

We are seeking a highly motivated Postdoctoral Researcher to develop innovative statistical methods for pediatric oncology clinical trials. The research program focuses on methodological challenges arising in pediatric and rare-disease settings, where patient populations are often small, outcomes may be delayed or complex, and conventional randomized trial approaches may not always be feasible. The successful candidate will work at the intersection of innovative clinical trial design, causal inference, external controls and real-world evidence, digital twins and counterfactual prediction, and statistical methods for survival, longitudinal, and other complex outcomes. Clinical applications will focus primarily on pediatric solid tumors, including neuroblastoma and sarcoma, as well as emerging cellular and immunotherapy studies such as CAR-T therapy. The position provides substantial flexibility for the postdoctoral researcher to develop an independent methodological research program based on the candidate's background and interests, emerging scientific opportunities, and important problems arising from ongoing pediatric oncology research.

Research areas

• Innovative Clinical Trial Design — Development of efficient and rigorous statistical methods for early- and mid-phase pediatric oncology trials, particularly in settings involving small populations, rare diseases, heterogeneous treatment response, or delayed outcomes.
• Causal Inference, External Controls, and Real-World Evidence — Development of principled approaches for incorporating external information into clinical trials when concurrent randomized control groups are limited or infeasible.
• Digital Twins and Counterfactual Prediction — An emerging research direction is the development and evaluation of digital twins and counterfactual prediction methods for clinical trials. Rather than viewing a digital twin solely as a prediction model, we are interested in understanding when model-based predictions can provide clinically and statistically credible information about outcomes under alternative treatment strategies.
• Survival, Longitudinal, and Complex Clinical Outcomes — Many pediatric oncology trials involve delayed, longitudinal, multistate, or otherwise complex outcomes that motivate new statistical methodology.

Responsibilities

The balance between methodological development and applied collaboration can be tailored to the candidate's background, interests, and career goals.

• Develop new statistical methodology motivated by important pediatric oncology problems
• Conduct simulation studies to evaluate statistical operating characteristics
• Analyze clinical trial, registry, and real-world datasets
• Develop statistical software in R and/or Python
• Collaborate closely with pediatric oncologists, clinical investigators, statisticians, and data scientists
• Participate in the design and