Intern Biostatistics (m/f/x)

With over 120 years of experience and more than 17,000 employees in over 20 countries, Daiichi Sankyo is dedicated to discovering, developing, and delivering new standards of care that enrich the quality of life around the world.

In Europe, we focus on two areas: The goal of our Specialty Business is to protect people from cardiovascular disease, the leading cause of death in Europe, and help patients who suffer from it to enjoy every precious moment of life. In Oncology, we strive to become a global pharma innovator with competitive advantage, creating novel therapies for people with cancer.

Our European headquarters are in Munich, Germany, and we have affiliates in 13 European countries and Canada.

For our headquarter in Munich we are seeking highly qualified candidates to fill the position, Working at Daiichi Sankyo is more than just a job - it is your chance to make a difference and change patients' lives for the better. We can only achieve this ambitious goal together. That is why we foster a culture of mutual respect and continuous learning, with a strong commitment to inclusion and diversity. Here, you will have the opportunity to grow, think boldly, and contribute your ideas. If you have a proactive mindset and passion for addressing the needs of patients, we eagerly await your application.

Ihre Aufgaben

Indirect treatment comparisons (ITC) are commonly used to compare the efficacy and safety of different interventions that have not been directly compared in head-to-head clinical trials. These comparisons are particularly relevant in the context of Health Technology Assessments (HTA), where the available treatment options in specific countries may differ from those included in the clinical trial of the intervention being evaluated.

Standard ITC methods include network meta-analysis (NMA) and Bucher ITC, which assume that effect modifiers are balanced across all populations in the network. When this assumption is not met and discrepancies in effect modifiers are present, population-adjusted methods such as matching adjusted indirect comparison (MAIC) and simulated treatment comparison (STC) are frequently used. These approaches account for differences in effect modifiers but are limited to pairwise comparisons, with the target population being that of the comparator trial.

Multilevel network meta-regression (ML-NMR) is a recent population-adjusted method that extends the NMA framework by using available individual patient data (IPD) in conjunction with aggregate data and adjusting for differences in effect modifiers [1,2]. Compared to other population-adjusted methods, such as MAIC and STC, ML-NMR can be applied to larger networks allowing for multiple treatment comparisons. Furthermore, ML-NMR enables treatment comparisons for any target population, rather than limiting the analysis to the comparator's population. However, the shared effect modifier assumption is often required when IPD is limited, meaning that parameters from the ML-NMR are identical among a set or class of treatments.

As the landscape of HTA evolves, ML-NMR is increasingly recognized as a viable tool for evaluating treatment effectiveness. HTA bodies are beginning to adopt ML-NMR in their assessments due to its ability to provide more tailored estimates than traditional methods when effect modification is present. However, additional research is needed to determine optimal settings for ML-NMR, to investigate the impact of certain assumptions on the results, and to understand when it performs similarly to other standard or population-adjusted ITC methods.

The aim of this internship is to assess the benefits and constraints of ML-NMR compared to other ITC methods using various scenarios. The work will involve analyzing how these methods perform under different conditions, identifying potential limitations, and suggesting enhancements to optimize their application in complex data analyses.

The duration of the internship is 6 months and can start now or according to agreement.

References

D. Phillippo, S. Dias, A. Ades, M. Belger, A. Brnabic, A. Schacht, D. Saure, Z. Kadziola and N. Welton, "Multilevel network meta-regression for population-adjusted treatment comparisons.," J R Stat Soc A Stat Soc, vol. 183, no. 3, p. 1189-1210, 2020. D. Phillippo, S. Dias, A. Ades and N. Welton, "Multilevel network meta-regression for general likelihoods: synthesis of individual and aggregate data with applications to survival analysis," arXiv [pre-print], 2024.

Ihr Profil

• Bachelor's degree in statistics, mathematics, or similar (ongoing).

• Good programming skills in R and SAS

• Good knowledge of indirect treatment comparisons

• Excellent oral and written communication skills

• Good written and spoken English

Kontakt

For more information: www.daiichi-sankyo.eu