The future of clinical development: AI, real-world evidence and new research models

Artificial intelligence is changing how clinical trials are designed

AI is rapidly evolving from a supporting technology into an instrument capable of strengthening multiple aspects of the development process. From protocol optimisation and feasibility assessments to patient identification and predictive analytics, AI helps organisations generate insights more rapidly and make better-informed choices.

Particularly during clinical trial preparation, AI can support:

• The identification of suitable study populations;

• The prediction of recruitment challenges;

• The optimisation of study protocols;

• The analysis of large volumes of clinical and operational data.

  
  

Although AI supports decision-making, human expertise remains essential. Clinical development requires scientific interpretation, clinical insight and regulatory judgement that cannot be fully automated.

Real-world evidence is becoming an increasingly important source for decision-making

Traditional randomised clinical trials remain the gold standard for demonstrating safety and efficacy. At the same time, demand is growing for insights into how treatments perform in routine clinical practice.

Real-world evidence, based on sources including electronic health records, registries and other routine clinical data, provides additional information on treatment outcomes, patient populations and long-term effects.

RWE is consequently playing an increasingly important role in:

• Clinical development strategies;

• Expansion into new indications;

• Post-marketing research;

• Health technology assessment;

• Engagement with regulators and healthcare authorities.

  
  

The challenge is therefore shifting from collecting more data to generating reliable and actionable insights.

New research models make clinical trials more flexible

Alongside technological advances, the design of clinical research is also changing. Decentralised and hybrid trials make it possible to conduct parts of a study outside traditional research centres, supported by digital monitoring, telemedicine and wearables.

These developments can contribute to:

• Greater accessibility for patients;

• Improved patient retention;

• More efficient data collection;

• Wider geographical distribution of participants;

• Better representation of diverse patient populations.

  
  

At the same time, these models introduce new challenges concerning data quality, cybersecurity, logistics and international regulation. Successful implementation therefore requires a careful balance between innovation and control.