Cystic Fibrosis (CF)—a common genetic disorder, especially among Caucasians—has been a primary focus of medical research for years. With genetic variations that demand unique treatments and the chronic progression of the disease, designing and executing clinical trials for CF is a complex task. Yet, it does come with a hefty price tag. The question is, how can we reliably estimate the cost of a CF clinical trial right from the protocol stage?
A research published in the Nature Reviews Drug Discovery reveals that any clinical trial’s cost depends heavily on the number of interventions involved, the nature of these interventions, the number of participants, the location, and the duration of the trial. For instance, gene and cell therapy are inherently expensive interventions. The median cost of clinical trials between 2015 and 2016 was a staggering $19 million, with noteworthy cost variations.
For CF, the costs escalate primarily due to its genetic nature and the kind of treatments involved. Vertex Pharmaceuticals’s revolutionary treatment bypasses the genetic errors causing CF, but it costs the NHS well over £100,000 per patient per year. Given the increasing focus on therapies that restore the function of the CF transmembrane conductance regulator (CFTR) protein, the clinical trial costs for CF are expected to rise even further.
An innovative solution to this challenge is provided by Fast Data Science’s Clinical Trial Risk Tool, which employs machine learning to predict a clinical trial’s cost from the protocol text. Machine learning algorithms are trained on historical data to understand the relationship between a trial’s protocols and the associated costs. By interpreting the schedule of events, interventions, patient numbers, locations, and more, these algorithms then provide cost estimates for new clinical trials.
This approach helps identify and mitigate potential cost overruns from the early stages of a trial. It also provides valuable insights to tweak protocols and optimize them, helping to funnel monetary resources more effectively.
The future of cost-effective CF clinical trials lies in leveraging machine learning, which can provide reliable cost estimates right from the protocol-writing stage. This way, researchers and funders can better prepare for the financial aspects of these trials rather than grappling with hidden costs later on. By refining our approach to estimating and managing clinical trial costs, we can accelerate the development of better therapies and, ultimately, improve the quality of life for those burdened with CF.
