Providing rapid answers in the evaluation of new treatments without adopting the standard strategy of evaluation appears growing, with uncontrolled trials increasingly used as the design of choice for new treatment evaluation. Such an accelerated process impacts the approval pathway (1). To increase the level of evidence in these early settings, external comparisons, that is to look for historical controls, including real-world data (RWD), appear of interest (2). However, to be valid, this requires a careful implementation of innovative statistical methodologies accounting for between-study variation, depending on the availability and type of external data, either individual patient data or only aggregated data, obtained from previous interventional trials or RWD. Most authors have warned against the misuse of each approach, and others have reported some methodological considerations regarding the use of external controls, but none has reported all the available methods and their
underlying objectives and assumptions for leveraging RWD according to the type of available data. Notably, most methods to address non-overlap, such as trimming or down-weighting data in regions of poor data support, change the estimand so that inference can no longer be made on the sample or the underlying population, resulting in ambiguous target populations that cannot be accurately defined (3).

The objective of this project is, based on an overview of existing methods and the development of new designs, to provide a well-defined methodological framework for the use of external data in single arms trials.

Steps of the project are to:

  • (a) Review existing literature of the design and analysis approaches of single-arm trials with external comparisons.
  • (b) Develop new designs, notably using Bayesian methods incorporating aggregate or individual data from experts elicitation or individual data with propensity score approaches
  • (c) Develop performance measures for such indirect comparisons and evaluate statistical properties of selected trial designs and analysis strategies
  • (d) Illustrate the newly developed efficient designs on case studies (post hoc analyses based on existing trial datasets) or influence the methodologies of future such trials.

At least 2 peer reviewed publications are anticipated in a methodological/clinical trial journal from this project.

The candidate will be based in the ECSTRRA team (Epidemiology and Clinical Statistics for Tumor, Respiratory, and Resuscitation) of INSERM U1153, located at the Department of Biostatistics of Saint Louis hospital in Paris (, France, where they will benefit from real-life experience working with specialists on clinical trials, including early phase trials and causal inference.

This project is part of the SMATCH consortium “Statistical and AI based Methods for Advanced Clinical Trial CHallenges in Digital Health” of the national program PEPR Santé Numérique, Programme #1 ( The supervisory team for this project includes Prof. Jérôme Lambert and Prof. Sylvie Chevret. The project will include consulting with other members of the SMATCH consortium group to design and plan the study, to present results.

The position (PhD or post-doc) will be defined according to the candidate. In case of a PhD program, the position will be a full-time 36-months PhD contract and the student will be affiliated to Graduate School 393 Pierre Louis de Santé Publique – Université Paris Cité – Sorbonne Université ( In case of a post-docthe exact duration of the contract will depend on the candidate’s background and previous experience, and be 36 months at most.

Requirements: Master or PhD level in biostatistics, clinical epidemiology, applied mathematics, with experience and proficiency in R programming.

Deadline for application: The position is available immediately and will remain open until a suitable candidate is selected.

Start date: Fall 2024 (academic year 2024-25).

Duration: 24 to 36 months (depending on entry level)

Salary: Doctoral or postdoctoral level; INSERM salary scale.

To apply, please send your CV and cover letter to, and


  1. Beaver JA, Pazdur R. “Dangling” Accelerated Approvals in Oncology. New England Journal of Medicine 2021; 384(18): e68. Doi:10.1056/NEJMp2104846
  2. Wang CY, Berlin JA, Gertz Bet al. Uncontrolled Extensions of Clinical Trials and the Use of External Controls—Scoping Opportunities and Methods. Clin. Pharmacol. Ther. 2022;111:187-199
  3. Crump RK, Hotz VJ, Imbens GW, et al. Dealing with limited overlap in estimation of average treatment effects. Biometrika 2009;96(1):187–199
Back to top