Clinical trials are becoming increasingly complex highlighting the need for efficiency. At the same time, electronic storage and computing power have increased dramatically and there are more opportunities for data sharing across industry stakeholders as patient data awareness is increasing. With these conditions, AI is poised to make scientific decisions easier and more reliable, therefore, making new drug development more efficient.

The clinical trial landscape in Asia Pacific

Over the last few years, Asia Pacific (APAC) has experienced rapid growth in clinical research, and the region’s contributions to the global clinical trial landscape are expected to increase even further. In 2022, APAC accounted for more than half (58%) of all global Phase I clinical trials, and the region’s potential is a result of its makeup of diverse countries, with strong research and development capabilities, a wide talent pool, world-class healthcare infrastructure, and robust regulatory and ethical standards.

Its large population to tap on for participants also contributes to both opportunities and challenges. On one hand, access to diverse cultures and communities ensures representation in clinical trials. On the other, cultural differences amongst the population could present challenges in trial design and execution, potentially affecting data collection and result consistency.

AI’s role in powering clinical research

Clinical trials have traditionally been very manual, and in some parts of the world, this is still the case. So, where can AI come in? Everywhere from asset discovery to data collection and cleaning, to data analysis and scientific decision-making.

It is estimated that only 1 in 5,000 to 10,000 new drugs successfully make it from early drug discovery to market and can take on average up to 15 years to develop – a lengthy and costly process with a low success rate. There’s a huge opportunity here to leverage AI to accelerate this process and improve success rates.

AI can assist in identifying targets and predict efficacy or toxicity thereby allowing consideration of many more candidates than previously possible and prioritizing them for further assessment. According to Morgan Stanley Research, the integration of AI and machine learning could pave the way for the discovery of 50 novel treatments within a decade. Furthermore, it has been predicted that big data and AI could reduce the discovery of candidate substances to 1 to 2 years, while the span from Phase 1 clinical trials to approval could be shortened to 5 to 7 years – this would halve the average duration of new drug development while reducing overall costs.

How AI and historical clinical trials data can help with operational and scientific aspects of clinical development

Historical clinical trial data can help more efficiently operationalize the conduct of clinical trials. For example, accelerating patient recruitment by learning from site-specific recruitment rates in past trials and cleaning data more efficiently by targeting data values that appear dissimilar to the rest of the dataset. Patient diversity can be advanced by

leveraging previous trial datasets and AI to identify sites that based on past performance are likely to recruit diverse populations for future trials.

AI and historical clinical trial patient-level data are also impacting the scientific design and interpretation of clinical trials. For example, Medidata’s Synthetic Control Arm (SCAs) uses previous clinical trials data and AI to simulate control groups when prospective recruitment is challenging. And regulatory bodies, including the United States Food and Drug Administration (FDA), are beginning to recognize the potential, especially in life-threatening or rare diseases with inadequate standard of care. With phase 1 single-arm trials, SCA can help make scientific estimates of the differential impact of a new product over the standard of care sooner, sometimes years sooner.

The future of AI in APAC clinical trials

Clinical trials and the process of new drug development generate some of the highest-quality data in the world. The use of AI in these represents a transformative shift that promises to make studies in drug development more efficient, accurate, and patient-centric. In APAC and globally, it continues to be an area to watch, with Stratistics Market Research Consulting forecasting that the global AI market for clinical trials will hit $1.88 billion in 2023, projecting growth to reach $9.28 billion by 2030.

In 2023 alone, approximately 25 percent of global studies started utilizing Medidata solutions, including electronic data capture. Medidata’s systems contain insights from over 35,000 global trials and 10 million participants with data captured and validated in the electronic data capture (EDC) system. For these reasons, we are in a unique position to deliver solutions that leverage the collective insights from past trials.

Thoughtful consideration and collaboration between regulators, technologists, and clinical professionals is also required to ensure that AI’s integration into clinical trials is safe, effective, and ethical. As AI technology continues to evolve, so too must the frameworks and strategies for its application in such a critical field.

#AIinHealthcare #ClinicalTrials #DrugDevelopment #MedicalInnovation #HealthTech