In the Age of AI, Reinvention Biopharma R&D Clinical Trials

In the Age of AI, Reinvention Biopharma R&D Clinical Trials
In the Age of AI, Reinvention Biopharma R&D Clinical Trials

Clinical trials are now the most expensive aspect of biopharmaceutical R&D, but they also offer the most opportunities for automation and the use of artificial intelligence (AI). The future of clinical trials will be fully Al-enabled, beginning with protocol and progressing to clinical operations centered on on-site and investigator management, before pivoting to patients and making it easier for them to participate in clinical trials.

History of Clinical Trials

Clinical trials are designed by research groups to answer specific research questions about the safety and effectiveness of a therapeutic strategy by evaluating defined access points, such as diagnostic biomarkers, throughout trial participants. Clinical trials can only begin after a regulatory authority has approved the pre-clinical regulatory submission and an ethics committee has reviewed it. The basic premise of clinical trials is that researchers take data from a small but indicative sample of subjects and extrapolate the findings to the larger patient population. If the sample size is small or poorly chosen, the results will be limited in their applicability. This is a concern not only from a statistical standpoint but also from an ethical and medical standpoint.

Today, it takes an average of 10-12 years to bring a new drug to market, with little change over the last decades in the linear and sequential process used to assess drug efficacy and safety. Currently, drug discovery, the first stage of R&D, take five to six years, followed by clinical trials, which take another five to seven years. Only ten of the 10,000 candidate drugs that were initially screened make it to clinical trials. Only one drug candidate out of every ten that enters clinical trials is approved for use with patients.

Modifying Clinical Trials is critical to improving productivity

The tried-and-true process of discrete and fixed phases of randomized controlled trials (RCTs) was created primarily to test mass-market drugs. On the other hand, RCTs lack the analytical power, flexibility, and speed needed to develop complex new therapies that target smaller and frequently heterogeneous patient populations. Furthermore, the current high-risk, high-cost R&D model is unsustainable, according to a Deloitte survey of biopharma industry leaders conducted in 2018 for the Digital R&D: Transforming the Future of Clinical Development report. Furthermore, clinical development is struggling to keep up with the ever-increasing amount of genomic data, real-world evidence (RWE), and other emerging data sources (such as biosensors).

More particularly, as clinical trials proceed, more patients are needed, but the eligibility and suitability requirements increase as well. A patient may be ineligible to participate due to medical history or a mismatch in the disease stage compared to the trial protocol. Eligible and suitable patients may find the requirements difficult, the recruitment process complex and time-consuming, or they may be unaware of or under-incentivized to participate.

Participation is also limited by the cost of frequent clinic visits. According to recent data, 18% of patients drop out after enrolling. Despite numerous initiatives to address this issue, increasing clinical trial diversity in an effective, scalable, and sustainable manner remains a challenge for clinical research. Furthermore, patients thinking about participating in a new medicine trial must weigh the benefits of early access to a potential new therapy against the risks of adverse events, as well as the inconvenience, potential financial burden, and time requirements.

Ineffective site selection, poor study design and trial execution, safety issues, and dropouts due to practical or financial issues are all challenges that an increasing proportion of clinical trials face. Despite the fact that patient recruitment and retention lengthen the time it takes to complete a trial, four-fifths of trials fail due to the inability to demonstrate efficacy or safety. Furthermore, the time and money required to complete a trial increase with each phase. The total cost of a Phase III failure includes the costs of all previous phases as well as the time that could have been spent on a different drug trial. Each failed trial adds to the ever-increasing costs of biopharma R&D.

AI to improve efficiency and effectiveness

Excessive costs, delays, and failures in clinical trials have a significant impact on patients. Despite numerous advances, two-thirds of classified diseases still lack effective drug therapy.  As a result, finding more effective and efficient methods of conducting clinical trials is critical.

AI, specifically deep learning (DL), machine learning (ML), and natural language processing (NLP), when combined with an effective digital infrastructure, has the potential to improve drug approval rates, reduce development costs, and expedite medication delivery to patients. AI and its applications are being invested in by all large biopharma companies. Novartis, for example, used AI to combine clinical trial data from multiple internal sources in order to predict and monitor trial cost, enrollment, and quality. As a result, the company reported that patient enrolment times in pilot trials were reduced by 10-15%.

AI can also be used to clean, aggregate, code, store, and manage a continuous stream of RWD. This makes data management more efficient, seamless, and dynamic. Furthermore, improved electronic data capture (EDC) can reduce the impact of human error in data collection while also allowing for seamless integration with other databases.


Interoperable data, open and secure platforms, and consumer-driven care are driving large-scale disruption in the life sciences and healthcare industries. Over the next 20 years, there will be a fundamental shift from "health care" to "health." While diseases will never be completely eradicated, advances in data and science, as well as the use of AI and other digital technologies, will allow for the earlier detection of pathological conditions. This will allow for more proactive interventions and a better understanding of disease progression, allowing for more active and effective patient well-being maintenance.

subscribe to our newsletter