While randomized controlled trials (RCTs) have traditionally been considered the gold standard for drug development, it is widely recognized that RCTs are expensive, lengthy, and burdensome on patients. According to some estimates, it takes more than a billion dollars in funding and a decade of work to bring one new medication to market. Despite exciting advances in genomics, patient-centric awareness, decentralized clinical trials, and the application of artificial intelligence (AI), there is a lack of compelling evidence to date that these trends have had a significant impact on the time and cost of the typical oncology clinical trial.

Nor have these developments yet had a significant impact on making clinical oncology trials less burdensome on patients.

The application of AI to oncology clinical trials has the potential to significantly reduce both the time and expense of clinical trials. There is increasing evidence that big pharma companies are using AI. The impressive list of companies with AI enabled initiatives to improve the oncology clinical trial process includes companies such as Novartis, Pfizer, Merck & Co., GlaxoSmithKline (GSK), Sanofi, AstraZeneca, Bristol Myers Squibb, and AbbVie.

AI has significant potential as it can be deployed to ease the burden of oncology clinical trials, which currently involve extensive paperwork and documentation to ensure compliance. Next, AI can be implemented to improve how informed consent forms are designed, as these can be lengthy and bureaucratic, involving detailed explanations of the trial's purpose, potential risks and benefits, and patient rights. Then, AI algorithms can be used to analyze electronic health records (EHRs) and identify eligible patients based on specific criteria such as cancer type, stage, genetic markers, and previous treatments.

For example, Roche, Genentech and AstraZeneca are using an app called Trial Pathfinder, an open-source AI tool which uses electronic health record (EHR) data to simulate clinical trials, integrating EHR data according to different inclusion criteria, and analyzing the overall survival risk ratio.

Successfully deploying AI in oncology clinical trials requires an end-to-end view such that business processes be redesigned and data quality be improved, integrating diverse sources, and dealing with unstructured content. It also calls for an unprecedented focus on the patient experience. However, neither business process redesign nor a focus on patient experience comes naturally to most pharmaceutical firms – and the lack of progress in these two areas is an impediment to the successful use of AI.  

While many pharmaceutical companies are doing something around patient experience and process improvement, in most cases not enough is being done to shift management attention from a vertical view of departmental activities to the flow of work that crosses organizational boundaries in creating value for patients. Shifting management attention from a vertical view of the organization to a patient centered or an “outside-in” horizontal view is a significant leadership challenge. It calls for focused leadership attention and at least two artifacts: a patient journey map and a set of enterprise level process maps – as well as candid discussion of these at the senior leadership team level.

While the health care sector has dabbled with how journey mapping can be used to improve patient experience and outcomes, this technique is not yet in widespread use in clinical trials. A patient journey map must be developed from the patient’s perspective, not that of the clinical team. It should be a living document that needs to be actionable and should be shared with the senior leadership team and known throughout the entire organization. Despite the available research on best practices, many companies continue to examine just the touchpoints. It gets worse. Such efforts are often made department by department and as middle managers tend to hoard data, little or no attention is paid to the improvement action that may be needed to resolve patient pain points.

Similarly, the process improvement and management approach in oncology clinical trials has emphasized an incremental improvement approach and has shied away from end-to-end process redesign. Note that researchers have identified that focusing on parts – rather than the whole is a root cause of failure with major change and yet that is what happens when organizations attempt to improve small parts of RCT processes inside departmental boundaries. Of course, the pharmaceutical sector has long been known for focusing on a vertical, hierarchical view of performance, due to the powerful departmental silos, which stand in the way of the type of cross functional collaboration that’s needed for success with major process change. The powerful departmental silos also adversely impact data sharing and create an obstacle in the effective use of artificial intelligence (AI) in new drug for discovery. Significant data integrity issues and the lack of published Pharmacokinetic/ Pharmacodynamic (PK/PD) data, for competitive or proprietary reasons, are particularly noteworthy hurdles to the deployment of AI.

Forward thinking pharmaceutical companies are invited to consider the following steps if they wish to deploy AI and thereby substantially compress oncology clinical trial cycle time and dramatically reduce costs:

·      Develop a patient journey map from the patient’s perspective – and discuss at the senior leadership team level.

·      Invest in creating a sense of team across organization boundaries.

·      Create incentives to share data and consequences for those who don’t.

·      Engage patients in the design of clinical trials and in drafting the text of the ICF.

·      Incorporate decentralized testing into clinical trial design.

·      Ensure that each arm of a clinical trial provides a clear benefit to patients.

·      Structure the screening and the study process with attention to patient comfort.

·      Focus on data quality and integrity.

·      Focus on transparency in reporting.

·      Appoint principal investigators not just on the research abilities – but also on their patient centric approach.

·      Advocate increased attention to patient education with medical schools.

Cancer is the 2nd leading cause of death globally and the world health organization (WHO) had predicted that global cancer cases will rise by more than 75% by 2050. Traditionally, cancer clinical trials have had just a 3.4% success rate – well below all other therapeutic areas – that should not just be a cause for concern but also a call for immediate action in leveraging AI to substantially compress oncology clinical trial cycle time and dramatically reduce costs.

Andrew Spanyi is President of Spanyi International. He is a member of the Board of Advisors at the Association of Business Process Professionals and has been an instructor at the BPM Institute. He is also a member of the Cognitive World Think Tank on enterprise AI.