Cancer research requires high-quality data which is resource-intensive and expensive to obtain. Almost a third of the total cost of cancer clinical trials is spent on non-clinical activities, including data collection, cleaning, and management. If researchers could use information from national data sets, gathered during patient consultations rather than a clinical study, the cost and time allocated to data collection in cancer research could be reduced. However, it’s essential that this data is accurate and timely.
The new study, published in The Lancet Oncology, compared cancer diagnosis data from national cancer registries in England and Wales with data collected during the SYMPLIFY study – an NDPCHS and Department of Oncology study assessing the performance of an early detection blood test. These national cancer registries, such as the National Cancer Registration Dataset (NCRD) in England and the Welsh Cancer Intelligence and Surveillance Unit (WCISU) in Wales, are considered the gold standard registries because of their rigorous processes to ensure that the data they collect is accurate, comprehensive, and validated through multiple quality control checks, making them reliable sources for clinical research
In the study, the researchers looked at how complete, accurate, and timely the information was from both sources, focusing on details like where in the body the cancer was found, what the cancer cells look like under a microscope and how far the cancer has spread.
The researchers discovered that:
-
Both sources had similar levels of complete information on the cancer, with over 95% of the details filled in.
-
The information about where the cancer was located, matched very well (over 90% agreement) between the two sources.
-
More complex details about the cancer cells and how far the cancer had spread, matched reasonably well (60-85% agreement).
-
The registry data was available almost as quickly as the data collected directly by researchers.
‘To our knowledge, this is the first study to comprehensively compare cancer outcome data collected on-site by research staff during a prospective clinical study with centrally held registry cancer data in England and Wales,’ said Ashley Jackson, a master's student in the Department of Oncology working with the NDPCHS Cancer Research Group and lead author of the study. ‘Previous studies in the field of cardiology have found significant cost savings so, given the cost burden of obtaining data for cancer studies, we wanted to explore if the same savings could be made.’
Dr Brian D. Nicholson, Associate Professor at NDPCHS and an author of the study, said: ‘Our findings suggest that central registry data can help alleviate the resource burden in clinical trials and improve cancer research efficiency. However, early engagement with cancer registries and adequate resourcing are crucial to realising these benefits. For example, registries might need additional resources to manage the increased administrative demands for provide data for a larger volume of registry-based trials. As many researchers currently face long delays in access to data, we hope this study will provide further impetus in support of improving data access.’
Even with the additional funding and resources that registries would need to provide data for research trials, the authors explain that they remain an economic alternative to the costly and time-intensive process of collecting data on-site. However, they also warn that while the registries were comparable to the study regarding the location of the cancer, other details weren’t as comparable. These include how far the cancer had spread and how fast it might grow, meaning before relying on registries for trial data, researchers need to think carefully about the kind of information they need.
Dr Stephanie Smits, Head of Population Cancer Research at the Welsh Cancer Intelligence and Surveillance Unit, Public Health Wales and an author of the study, said: ‘We are pleased that these findings highlight how cancer registries can contribute to cancer research. When involved in the research process, and adequately resourced, cancer registries can support the delivery of timely, and high quality cancer research.’
Sarah Stevens, Director of the National Disease Registration Service, NHS England, said: ‘This study further demonstrates the value of high quality and timely national cancer registration data to support cancer research. Adequately resourced, this approach can significantly reduce the burden on clinical teams and minimises the cost of establishing additional data collections for research.’
The authors explain that further research is needed to understand if factors like the type of cancer, its stage at diagnosis, or treatment options used impact the accuracy of information available in the registries. Moreover, they explain that future studies could explore if the findings from this study also apply to other types of cancer information, like how well treatments work or how long patients survive.