FDA Fellowship in Statistical Methods for New Approach Methodologies (NAMs)

About the role

This research opportunity is available immediately with the U.S. Food and Drug Administration (FDA), Center for Drug Evaluation and Research (CDER), located in Silver Spring, Maryland. CDER regulates over-the-counter and prescription drugs, including biological therapeutics and generic drugs, to ensure safe and effective therapies are available to the public.

Research Project

This fellowship encompasses interrelated research projects addressing methodological challenges at the intersection of statistical science, in vitro pharmacology, and regulatory decision-making under the New Approach Methodologies (NAMs) framework. Potential project topics include:

• Investigating statistical methods for evaluating parallelism of dose-response curves between test and reference standards through simulation studies.
• Evaluation of multiple statistical approaches — including logistic regression, principal component analysis, and artificial neural networks — to determine optimal modeling frameworks for high-dimensional in vitro data generated by platforms such as organ-on-a-chip systems and MEA-based stem cell models.
• Contribute to a robust statistical framework to support the regulatory use of NAMs in proarrhythmia risk assessment, with a focus on characterizing cardiac safety signals such as delayed repolarization and QT interval prolongation using in vitro assay data from CiPA reference compounds, ultimately informing regulatory decision-making under ICH S7B/E14 guidelines.

Learning Objectives

Upon completion of this fellowship, participants will have gained the following competencies and practical experiences:

• Develop proficiency in statistical methods for evaluating parallelism of dose-response curves with direct application to regulatory bioassay development and drug product quality control.
• Gain experience in building, selecting, validating, and calibrating predictive models appropriate for small preclinical drug safety datasets, with attention to overfitting, uncertainty quantification, and regulatory defensibility across varying sample sizes and data variabilities.
• Develop applied artificial intelligence (AI) and statistical programming skills in R and/or SAS by implementing end-to-end machine learning pipelines, including data preprocessing, model training, cross-validation, simulation automation, and regulatory-standard reporting.
• Master quantitative risk assessment principles, including the statistical and scientific basis for safety margins, and design simulation studies to evaluate statistical methodologies for establishing safety bounds in regulatory pharmacology applications.
• Acquire practical experience analyzing in vitro assay data characterized by significant inter-laboratory variability, and develop strategies for identifying, quantifying, and mitigating its impact on statistical inference and regulatory decision-making.
• Gain skills in creating reproducible, audit-ready analytical outputs, communicating findings effectively to multidisciplinary scientific and regulatory audiences, and co-authoring a manuscript for publication in a peer-reviewed statistical or regulatory science journal.

Qualifications

• Currently pursuing or have received a master’s degree in mathematics, statistics, biostatistics, or a closely related field.

Citizenship

• U.S. citizen or Lawful Permanent Resident (LPR).

Appointment Details

• Start date: October 2026 (flexible).
• Initial appointment length: One year, renewable upon recommendation and funding availability.
• Level of participation: Full time.
• Stipend: Monthly stipend commensurate with educational level and experience.
• Health insurance: Proof of health insurance is required.
• Background check: Completion of a successful background investigation by the Office of Personnel Management is required.