Measuring the value of clinical cancer trials

Clinical trials are an essential way to test the safety and efficacy of new modes of medical care. They can also prove expensive and don’t always provide value proportionate to cost.

With support from the Victorian Comprehensive Cancer Centre, researchers at the University of Melbourne’s Cancer Health Services Research Unit have developed a new tool to identify which trials generated the highest value, to ensure health systems make the most of limited resources.

Published in Cost Effectiveness and Resource Allocation, their work analysed existing literature and consulted a wide range of stakeholders including people with lived experience of cancer to identify a set of criteria that can be used to rank the value of cancer related clinical trials using a technique called multi-criteria decision analysis (MCDA).

The study identified seven key criteria including unmet need, size of target population, trial accessibility for eligible participants, patient outcomes, total cost, academic impact, and use of trial results. These metrics were weighted and ranked before being applied to the results of six different clinical trials to calculate their value.

“With only limited resources available, clinical trial units need to be able to effectively choose trials that will provide value, which can be defined in a range of different ways,” explains lead investigator Piers Gillet.

“This research establishes one possible definition of value and provides a way to empirically quantify it. It is the first step towards being able to better evaluate trials prospectively, and thus more efficiently use the limited funds available.”

Results of the study were also validated by Dr Kathryn Field at the Peter MacCallum Cancer centre in a separate study published in the Journal of Clinical Neuroscience analysing brain cancer trials.

Despite this promising progress, the research team emphasizes that the tool requires further refining and is intended to complement rather than replace current clinical trial selection processes.

“With further development, either by extending the tool to better evaluate a broader range of trial types and reported outcomes, or implementing machine learning methods to prospectively value trials, this tool could be used to assist in the process of trial selection within clinical trial units,” says Piers.