Case studies – Cambridge Biostatistics Consultancy
Read about some of the work we have done with our clients.
Developing bespoke software to improve genetic drug target validation (Novo Nordisk)
In statistical genetics, Mendelian Randomization is a technique that uses naturally occurring genetic variation to investigate whether a risk factor has a causal effect on a disease outcome. In recent years, the approach has increasingly been applied in drug development to provide genetic evidence for target validation.
Scientists at the Novo Nordisk Research Centre in Oxford requested the development of bespoke statistical software to perform large-scale multivariable Mendelian Randomization analyses using genetic summary data. Ashish Patel carried out a secondment at Novo Nordisk to complete this work. The project focused on improving the reliability of results in complicated genetic regions by using genetic principal components as tools in the analysis, and by dealing with situations where the genetic variants used might be weak or unreliable, which could otherwise distort the findings.
The methods were extensively evaluated through simulation studies and subsequently applied in case studies relevant to Novo Nordisk's drug target discovery programme. A particular emphasis was placed on integrating Mendelian Randomization with colocalization analyses, enabling more robust assessment of whether related traits are influenced by shared causal genetic variants. A write-up of the statistical methodology used is currently being prepared for academic publication in collaboration with Novo Nordisk.
Methodological Leadership and Regulatory Engagement (Novartis)
Our consultants have long-standing expertise in cutting-edge statistical methodology for clinical trials. This includes pioneering frameworks for response-adaptive randomization [1] as well as the development of valid point estimation and confidence intervals following adaptive designs [2,3]. In parallel, the theme actively shapes global regulatory policy, notably by submitting formal commentary on the ICH E20 draft guidance on adaptive designs (read our response). To bridge theory and trial ethics, the group also recently organized a multi-stakeholder workshop examining the ethical dimensions of adaptive trials [4,5]. This unique combination of foundational research and proactive engagement with regulatory and ethical questions ensures that guidance remains both mathematically rigorous and directly applicable to contemporary clinical practice.
To translate this expertise into industry practice, we conducted an intensive three-day visit to Novartis in Basel, delivering bespoke, hybrid short courses spanning the advanced trial design methodologies outlined above. The program was carefully structured to mirror the practical priorities of the participating industry statisticians. It featured an interactive panel discussion that provided a valuable forum to examine the operational and statistical trade-offs inherent in adaptive methods when applied to specific pipeline problems.
A series of targeted, one-to-one consulting sessions enabled individual Novartis researchers to troubleshoot complex methodological questions of direct relevance to their active protocols. This visit facilitated a highly productive, bilateral exchange of expertise and ideas, serving to strengthen existing academic-industrial ties and laying the groundwork for ongoing collaborative research projects. These active ongoing initiatives include developing methods for adjusted estimation following adaptive designs, extending the MCP-Mod approach to adaptive design frameworks [6], and integrating digital outcome measures into contemporary trial protocols [7].
References
1 Robertson, D. S., Lee, K. M., López-Kolkovska, B. C., & Villar, S. S. (2023). Response-adaptive randomization in clinical trials: from myths to practical considerations. Statistical Science, 38(2), 185-208.
2 https://onlinelibrary.wiley.com/doi/full/10.1002/sim.9605
3. https://onlinelibrary.wiley.com/doi/full/10.1002/sim.70174
4 Baas S, Pin L, Robertson DS, et al. Adaptive versus fixed designs in confirmatory clinical trials: centering the choice on ethics. Clinical Trials. 2026;0(0). doi:10.1177/17407745261468001
5.Baas S, Pin L, Robertson DS, London AJ, Sheehan M, Villar SS. Rejoinder: “Adaptive versus fixed designs in confirmatory clinical trials: Centering the choice on ethics”. Clinical Trials. 2026;0(0). doi:10.1177/17407745261469874
7 https://link.springer.com/article/10.1186/s12916-024-03590-