2018 was a big year with respect to big data and analytics. Not just for clinical trials, but the field overall as it relates to the collection of data, the analysis and deep-mining for patterns and the interpretation thereof to inform strategy. Banking and manufacturing are two industries that are particularly adept in the field but health sciences is not far behind.
2018 was a big year because the global spend on big data and analytics crossed the $150B mark. IDC forecasts that it will cross the $250B mark within the next few years. Forbes® conducted its annual survey on big data and analytics and revealed yet another milestone for 2018: over 50% of enterprise companies (57%) now have a Chief Data Officer who upholds the responsibility of complying with data policies and regulations as well as ensuring its fair and accurate distribution, essentially democratizing it. The same survey also revealed that more than half of companies (52%) are now using predictive analytics to source contextual intelligence which can then be applied to drive changes in business operations.
And, most relevant to this audience is the announcement by FDA Commissioner Scott Gottlieb, PhD, that the FDA will be increasingly looking to big data and artificial intelligence (A.I.) to improve clinical trial design and outcomes. We can likely expect additional funding and new ventures in clinical trial data management. To this end, the FDA had a busy year – in September they published their Draft Guidance on Civil Money Penalties Relating to the ClinicalTrials.gov Data Bank. In a nutshell, the FDA will levy fines against any parties who fail to properly register their trials and disclose their data. Several responses to the Draft Guidance were received prior to the November 30 deadline and a formal directive is expected from the FDA sometime in the Spring of 2019.
Indeed, clear definition of clinical trial data management requirements (and a fine-based approach to reinforcement) will go a long way towards improving patient safety and reducing operating costs. This is an important effort given that fewer than 1 in 4 (22%) of trials are compliant under current policies and regulation based on our own analysis with BrackenData™ Compliance software. The EU is faring better with a reported data compliance rate of 49.5% based on the most current retrospective analysis of compliance across 7,274 trials.
The conventional wisdom that “two immunotherapy drugs are better than one” received a few setbacks this year. Perhaps best shown through the lack of results obtained in the Phase 3 melanoma study combining Merck’s Keytruda® with Incyte’s epacadostat (part of a new class of IDO inhibitors: IDO is an enzyme secreted by tumors which suppresses the activity of immune cells). Although the failure provided valuable lessons for the role of IDO, some experts cite this is an example of how preclinical (animal) model data looks promising but doesn’t translate to human patients.
As with everything, there is balance – conversely, the combination of revlimid®, the myelodysplastic (blood & bone cancers)) drug by Celgene and the rheumatioid arthritis drug, rituximab® by Roche, were shown have an additive effect in the treatment of melanoma. Patients on the combination showed superior progression-free survival. Final data for this Phase III trial are pending.
In an analysis of eligibility criteria for Swedish heart patients, the findings were surprising: women were being excluded from cardiovascular clinical trial participation at a much higher rate than their male counterparts. Given that female heart patients were typically older (81 vs. 75) than male heart patients and had reduced kidney function and overall lower body weight (69 vs. 85 kg). Patient data were not meeting the stringent exclusion criteria, especially for women. Based on the criteria, which was applied to 1,924 heart patients, only 15 women (16%) of the 95 patients eligible to participate in a heart-failure trial in Sweden were selected.
Perhaps we missed some of the key stories on clinical trials data for the year 2018? If we did, we would welcome hearing from you. Feel free to send a response on social media!
Industry teams use BrackenData Analytics to create tables and visuals of clinical trial data. Signup to get a personal demonstration of our tools and dashboards: