How innovative trial designs de-risk drug development in rare diseases

6 October 2025

Estimated reading time: 6 minutes

EpiScalp

Rare disease is a therapeutic area where ad hoc development and access strategies with dedicated analytics are required. From innovative trial designs to data augmentation for regulatory success, CROs and experienced experts and consultants are contributing to more efficient, and smarter drug development, to bring better drugs to patients, faster. Dr Billy Amzal, leading strategic consulting at Phastar, explores some of the current innovations transforming drug development for rare disease

Despite regulatory and health economic challenges, rare disease continues to be a huge area of focus for pharma and biotechs, given its societal and ethical burden. In 2024, more than half of novel drugs approved by the FDA were for orphan therapies¹. Last year, both the number of rare diseases being targeted by pharma, and industry-sponsored rare disease trials were at record highs². This growth seems set to continue with the rare disease market expected to achieve a CAGR (compound annual growth rate) of 11.93% from 2025 to 2030³and the new EU HTA Regulation aiming to address inequities in access to rare disease treatments⁴.

However, traditional clinical trial designs are often unsuitable for the rare disease space, and a new era of innovation requires new ways of working. Smarter and innovative design and analytics offer pharma and biotech companies the opportunity to accelerate and de-risk drug development, particularly for underserved patient populations. By working in partnership with specialist biometrics experts or CROs (contract research organisations), pharma and biotech companies can effectively minimise risk and maximise expected value for their drug development and address a huge area of previously unmet medical need.

Image of two female scientists in white lab coats, looking at data on a computer in a lab — Traditional clinical trial designs are often unsuitable for the rare disease space, and a new era of innovation requires new ways of working

Innovative trial designs

Innovative and effective trial designs are meant to minimise data collection yet maintaining or even increasing probability of clinical trial success. Such approaches include simulation-based optimisation for e.g. adaptive and/or Bayesian designs. This is particularly relevant to the rare disease space where recruiting enough participants to demonstrate efficacy and safety continues to be a major barrier to drug development.

Collectively rare diseases affect a huge number of people—between 263 and 446 million people globally at any one time⁵. However, with more than 7,000 recorded rare diseases⁶ individual patient populations are often small. Bayesian trials require between 30% and 2,400% less participants than frequentist trials⁷ as they allow us to borrow information from past or similar studies^8,9. Such simulation based optimal design enables the reduced sample size that and larger probability of success¹⁰.

External control arms

External control arms (ECAs) offer a valuable approach in rare disease where gold standard randomised controlled trials (RCTs) are not always possible for practical or ethical reasons. The potential benefits of ECAs include reduced development time for new medicines, enhanced generalisability of relative efficacy vs. standard of care for real-world patients and a trigger to accelerate or optimise trial patient recruitment.

ECAs can be built via trial arm emulation using matched controls, simulation using predictive models or synthetic patient generation using artificial intelligence. However, careful methodological planning and thoughtful/critical reporting are essential for actionable and robust results. This is further emphasised in draft guidance from regulatory bodies including the FDA¹¹ and the MHRA¹².

Proper strategised evidence generation sequence for ECA development helps to maximise the chances of approval. For example, a drug to treat refractory precursor B-cell acute lymphoblastic leukemia, a rare and aggressive cancer, was granted accelerated approval based upon findings from a single-arm, open-label phase II trial and supplemental data derived from a historical control arm. A second external control arm study using summary-level outcome estimates from previous clinical trials provided further support for approval¹³.

Data augmentation

In many situations, including rare conditions, standard clinical trial settings fail to generate sufficient evidence to demonstrate effect vs standard of care. The use of external information, processed with fit-for-purpose analytics and patient synthesis, can augment information and maximise power beyond the trial data.

Bayesian borrowing allows us to incorporate information from clinically relevant external evidence, such as historical trial data or real-world data (RWD), without compromising the statistical validity of the trial. This approach has been increasingly used and promoted by EMA and FDA in the context of paediatrics extrapolation, rare diseases and country-to-country bridging.

For example, Chinese patients had not been included in a global study to support drug approval. A separate study was therefore required to support registration with the CDE. A Bayesian dynamic borrowing (BDB) approach was chosen as it utilised the evidence already available from the global study and was more likely to provide robust evidence than an unpowered standalone study. The BDB study was designed to be as similar as possible to the global study, with the objective of assessing the effect of a new treatment compared with a reference treatment for a continuous endpoint in Chinese patients. Use of a BDB approach resulted in greater efficiency of data use and considerable time saving, enabling accelerated availability of medicines to patients¹⁴.

Image of connecting lines and lights to indicate global data sharing — Bayesian borrowing has been increasingly used in the context of paediatrics extrapolation, rare diseases and country-to-country bridging

Leveraging real-world data (RWD)

The past five years have seen unprecedented advances in our access to, and the interoperability of, RWD. This has transformed drug development and evaluation paradigms. Regulators have highlighted particular interest in RWD as a complement to clinical trials for rare disease treatments, as well as recognising its benefits to support wider public health¹⁵.

Leveraging RWD can also unlock insights to optimise and de-risk clinical trial designs. For example, characterising progressor profiles and endotypes, the relevant (sensitive and specific) endpoints to consider for maximal translation of treatment effect and anticipating baseline risks and standard of care (SoC) treatment effects.

Conclusion

In an era of innovation, advanced analytics are contributing to more efficient, faster, and smarter drug development. However, these opportunities come with new challenges and require careful planning and execution to secure impactful results, regulatory approval, and ultimately medical and economic drug value once reaching the patients.

Pharma and biotech companies must work in partnership with expert statisticians and data scientists to understand the opportunities on offer—and how to mitigate the risks. By doing so we can unlock a new era of research and accelerate drug development in rare disease.

Connect with Billy

innovation spotlights the future of RARE through a technological lens. To access more innovation articles click below.

innovation

Subscribe Now