About
The R&E group Biostatistics significantly contributes to improving health and healthcare based on research design, modelling and data analysis firmly grounded in statistics and mathematics. It supports developing new treatments for rare diseases, innovation in drug regulatory science, modelling dynamics of infectious diseases and advancing novel data analytical approaches to clinical practice (data driven innovation).
Contact
Scope and area of work
The R&E group biostatistics holds a strong research portfolio focussing on application of biostatistics to health and health care related problems that are of high interest for patients and society.
Read more
Scope and area of work
The R&E group biostatistics includes a strong research portfolio, engages in education from BSc, Msc to post-academic level and includes an internal consultancy service for researchers and PhD students. Across the portfolio the R&E group focuses on application of biostatistics to health and health care related problems that are of high interest for patients and society. This ranges from enabling new treatments being developed for highly unmet needs (e.g., rare diseases), to methods to assess complex prediction models for use in clinical practice to improve individualised treatment. The research is covered in the following themes:
- Clinical trials & Real World Data for Regulatory Science.
- Innovative clinical trials for rare diseases.
- Advanced modelling of longitudinal and survival data, including federated inference.
- Models and methods for infectious diseases.
- Statistical inference and use of AI in practice.
Clinical trials & Real World Data for Regulatory Science.
Our R&E group has a strong research track record improving methodology for clinical trials, ranging from cluster randomised trials for complex health interventions to subgroup analyses in trials, hybrid Bayesian – frequentist approaches and multi-arm trials such as basket and umbrella designs. There is particular focus on drug development and regulatory decision making, with long standing collaboration with CBG-MEB and the EMA. A second core of research in this context is the use of Real World Data to augment the evidence from randomised trials. Hybrid designs, use of Real World Data to establish drug effects in target populations and use of real world data for extrapolation of trial results to broader populations are key themes.
Innovative clinical trials for rare diseases.
As part of the Radboudumc’s concerted effort to bring novel treatment options to rare disease patients, this research theme focuses on trial designs and analyses that meet the challenge that the numbers of patients that can engage in clinical trials are small. Methodologies cover a broad range: Leveraging mechanistic modelling and extrapolation for efficient trials, augmenting trials with external control data, basket trials across mechanistically similar diseases for repurposed drugs, n-of-1 designs and Bayesian approaches. It also includes improving and validating novel outcomes (such as Goal Attainment Scaling) that can deal with the heterogeneity between patients that is also typical for many rare diseases.
Advanced modelling for longitudinal and survival data, including federated inference.
Medical research questions often involve complex modelling of survival or longitudinal data. The R&E group Biostatistics has key expertise and research in these areas with broad potential for application: from clinical trials, observational data to public health data. Research and expertise incudes novel methods that can efficiently apply complex modelling with federated data: Data is available at different locations, but they cannot be merged to one dataset due to legal and practical limitations. Bayesian federated inference as developed by our group and the Radboud University Faculty of Science can construct results from local inferences in separate centres what would have been inferred had the datasets been merged. Especially if datasets are small across multiple countries, this has huge practical advantages.
Models and methods for infectious diseases.
Our main aim is to advance the understanding of infectious diseases through mathematical and statistical models. In particular, our group focuses on developing innovative approaches to address the multifaceted challenges posed by infectious diseases within clinical and epidemiological contexts. Research encompasses a wide range of infectious diseases including both directly transmitted pathogens (e.g. SARS-CoV-2) and vector-borne diseases (e.g. malaria, schistosomiasis). Specific activities are transmission models and scenario analysis to estimate the impact of control interventions, longitudinal and survival models, Bayesian spatial and latent variable models.
Emerging theme: Statistical inference and use of AI in practice.
Pathologists are exploring the potential of Artificial Intelligence (AI) to enhance breast cancer treatment decisions. AI-driven software can accurately extract vital information from images, enabling predictions about disease recurrence and treatment outcomes. In a new collaborative research project, i.e. COMMITMENT project, funded by the KWF Dutch Cancer Society, and led by Jeroen van der Laak and Francesco Ciompi, researchers aim to apply this AI technology to benefit three distinct groups of breast cancer patients. Our group’s contribution focuses on the validity and reliability of the translation of the AI solutions for clinical practice. It is a first project in its kind, but it is estimated this will be a topic of increasing interest for biostatistical contributions.
Chairs/RGL's
The R&E group Biostatistics is headed by prof. dr. Kit Roes, who holds the chair in Biostatistics (link will follow).
Projects
-
The aim of the project is to establish the value of registry-based RWD in augmenting RCT data and to enable the more effective and ethical use of registry data to support patient-centred regulatory and HTA decision-making.
Biostatistics leads the Methodology development Work Package, which (a.o.) includes the following key areas: statistical methods to incorporate registry data in early drug development decision making; methods to extrapolate clinical trial data to target populations of different composition; methods for identification of subgroup for which treatment effects differ. As founding element, innovative methods for Bayesian federated inference are implemented and extended, allowing advanced modelling in federated databases with only one pass of the data. -
Dr. Marleen van Gelder is seconded at Oslo University, with research focusing on using observational data to gain more insight into the safety of medication use, in particular for populations in which trials are often not feasible (pregnant people, children). Through this collaboration, it is at the forefront of this research domain, where (a.o.) Oslo University is preparing to join DARWIN-EU, one of the key initiatives of EMA to leverage Real World Data for regulatory decision making.
-
Biostatistics has a long standing collaboration with the Dutch Medicines Evaluation Board (CBG-MEB), with three biostatisticians seconded to the CBG-MEB for advanced Scientific Advice and assessment. In addition, the third CBG-MEB funded Biostatistics PhD project is undertaken, focusing on improving Scientific Advice for early dose finding in oncology, given the significantly different mechanisms of action of modern anticancer drugs (compared to more traditional chemotherapy).
-
The main objective is to accelerate drug repurposing for rare neurological, neurometabolic and neuromuscular disorders. SIMPATHIC’s main accelerating innovation is the simultaneous drug development for groups of patients with different genetic diagnoses but overlapping neurological symptoms and molecular pathomechanisms.
The R&E group Biostatistics contribution is focused on designs of innovative basket clinical trials to which patients with different disorders are recruited, utilizing and aggregating personalized clinical endpoints.
-
The objective of the project is: to provide clinical trial stakeholders, trialists and regulators with a generalisable framework encompassing methods, workflows and evidence-tools to improve the level of evidence in regulatory decision making in rare diseases.
The R&E group Biostatistics is Work Package lead, focusing on integrating the novel methodologies into a credibility framework to guide regulatory use of advanced modelling solutions. A key asset in the project is the large amount of clinical trial and modelling data that will be shared by pharmaceutical companies, as well as real world data from French rare disease registries.
-
Over a period of 8 years, two consecutive PhDs students will support collaborative research across the two RU faculties to act as seed for larger collaborations. In the first PhD project the focus is on the development and application of new mathematical, statistical and computational techniques for inference in case the sample size is small compared to the number of covariates in a statistical model. The second project will focus more on implementation of novel methodology in clinical research practice.
-
This project aims to develop and apply novel methodology for analysing small data sets: correction methods for model overfitting and methodology for doing federated analyses. The latter aims to construct from local inferences in separate centres what would have been inferred had the datasets been merged. Especially if datasets in medical centres are small and cannot be combined due to e.g., privacy legislation, applying this approach improves the accuracy of parameters estimates and prediction. The methodology will be applied to data on salivary gland cancer and other clinical data to obtain new insights in possible predictive factors of these diseases.
-
Whitin this project, our group developed SchiSTOP, a hybrid stochastic agent-based and deterministic modelling framework to reproduce S. mansoni transmission in an age-structured human population, including the parasite dynamics in the intermediate host (freshwater snails). SchiSTOP is flexible in the implementation of the assumptions of regulating mechanisms for S. mansoni transmission. It has been developed to explore the interplay of different regulating mechanisms and their ability to explain observed patterns in S. mansoni epidemiology. Its formulation is suitable to answer diverse research questions about the epidemiology and control of schistosomiasis. For instance, the presence of a specific module for the dynamics in snails allows for a thorough assessment of the impact of snail control interventions.
Read more on Cordis EU wbsite
-
The COVID19 pandemic has brought to the fore long-standing inequities that resulted in already-vulnerable groups bearing a disproportionate burden of the disease. Poor and minority groups are more likely to be infected and to experience severe outcomes. This may be due to biological factors like susceptibility and infectiousness or behavioral like contacts rates. Inequalities may have been exacerbated by non-pharmaceutical interventions (NPIs) along with differences in the ability to comply to them. These factors are all likely to differ by socio-economic status (SES). Indistinguishable in the reported disease figures, inequality factors have different implications in terms of expected effectiveness of NPIs. Mathematical models informing epidemic control policies have not accounted for equity as it is hard to disentangle the key drivers. In this project, we develop a novel approach that builds on existing and novel data sources to i) resolve the relative impact of the key drivers. We use improved stochastic transmission models to ii) estimate the implications of NPIs on the underlying inequalities and design effective and equitable interventions while reducing disease inequality.
-
In this project the aim is to apply AI technology to benefit three distinct groups of breast cancer patients (individualised treatment). The R&E group Biostatistics focuses on the validity and reliability of the translation of the AI solutions for clinical practice.
Initial education
Biostatistics is core for medicine and biomedical sciences Bsc students and the group engages thoughout the Bsc phase. At Msc level we offer several courses on methodology, close to our key focus areas: clinical trials, survival and longitudinal data analyses, experimental design and causal inference. These are either part of the Biomedical Sciences master program at Radboudumc (read more).
Post-initial education
It is also possible to follow post-academic courses, especially 'Statistics for clinical researchers' to be well prepared for research projects. More specialized courses (e.g., in survival or longitudinal analyses or Bayesian methods) are also provided, depending on demand.
Collaborations
Long standing collaborations include the Dutch Medicines Evaluation Board (College ter Beoordeling van Geneesmiddelen – CBG-MEB), the European Medicines Agency (EMA), the Faculty of Science of Radboud University and Erasmus Medical Center (Prof. van der Vlas) and Oslo University. Through the portfolio of (international) projects, collaborations included are (a.o.) University Medical Center Groningen, Inserm, Karolinska Institute and Hospital, Hannover Medical School, University of Lisbon, University Medical Center Utrecht and Utrecht University. The group also participates in ERDERA (European Rare Disease Research Alliance).
Consultancy
The R&E group Biostatistics engages in the Radboudumc Research Technology Center Biostatistics and Health Economics. The RTC offers internal consultancy to researchers and PhD students within or associated with the Radboudumc. The main objective of this service is to enhance the quality and integrity of Radboudumc research. It is therefore very accessible and promoted to engage early – at research design and/or research proposal stage.
If you are interested in consultations or services from an R&E group, please fill out our application form.