More than 500 million people on the planet have chronic lung disease — all eager for better therapies. Unfortunately, the wait for novel treatment is a long and expensive one — it takes 14 years and more than $1 billion, with a failure rate of 95%, to get a single drug from research labs to pharmacies.
Many of the issues that contribute to these high-cost failures and delays in clinical trials have something in common: poor data. Here are two examples:
- Weak endpoints – Common tests used to assess treatment response, such as the six-minute walk test, are often indirect measures of lung health, are insensitive, or both.
- Poor data quality – Data quality issues can occur in data acquisition, transcription, storage, and more. To measure the impact of a drug, it is critical that the collection of data from time point to time point is consistent and comparable.
Because of weak endpoints and poor data quality, clinical trials require large population cohorts to show a statistically significant therapy impact. This drives high cost and long timelines.
There is good news for those 500 million patients with lung disease. We are in the midst of a lung intelligence revolution that is vastly improving the efficiency of clinical trials by addressing the underlying issue of poor data. The combination of rich imaging data, called an “imaging biobank,” with advances in artificial intelligence (AI) is a powerful tandem.
With lung intelligence, trials don’t need to rely on weak endpoints. Sponsors are increasingly leveraging imaging biomarkers that deliver direct and precise measures of a therapy’s impact on lung structure and function. Now that sponsors can measure sub-millimeter changes to airway walls, for example, the impact of a therapy can be seen much quicker and with fewer subjects compared to conventional measures (e.g. exacerbations, death, or a six-minute walk test).
While precision imaging biomarkers are incredibly powerful, there has been a hurdle for trial sites to overcome. Collecting high resolution, consistent imaging data can be more challenging than administering conventional tests such as patient surveys. For example, CT scanners must be calibrated at regular intervals, patients must be coached on breath holds, and other elements of a trial protocol must be followed carefully to ensure high quality data.
Cloud services, AI, and refined data: A powerful trio
Across life sciences and healthcare, the ability to capture, share and analyze data in the cloud has transformed drug discovery and clinical trials. With that cloud infrastructure, the transformative value of lung intelligence truly shines.
Cloud-based portals now orchestrate imaging operations for clinical trial sites, allowing staff to easily overcome the hurdle of executing precision clinical trial imaging. For example:
- AI-powered services perform quality control checks and can intervene instantly to prevent poor data from entering the system.
- Reminders are issued to sites at appropriate intervals for scanner calibration, with step-by-step guidance.
- Staff training is conducted through portal-based eLearning modules, with certification status maintained centrally.
- Data management, including auto-anonymization, secure exchange, and data organization is now managed effortlessly and centrally by cloud technologies.
- Imaging biobanks and operational data is available in the cloud for high-value analysis, such as retrospective analyses that could identify candidates for future trials.
- Sponsors have real-time trial business intelligence dashboards to monitor the health of their trial sites, data quality, and overall trial operations.
The combination of cloud and AI/imaging-related biomarkers has dramatically reduced the friction historically caused by sponsors introducing imaging endpoints to a trial. Cloud services put AI-generated biomarkers just a few clicks away. Now, for trial sponsors who want to capitalize on the significant benefits of bringing imaging into their respiratory clinical trial, operational complexity for sites is no longer a time or cost barrier. Furthermore, sponsors can be confident in the quality of the data that will be collected by cloud-enabled sites.
Decentralized trial imaging
Now that cloud portal technologies are removing friction from imaging-based clinical trials, an interesting beneficiary has emerged: independent imaging centers. These centers traditionally have not participated in clinical trial imaging, largely due to the complexity that was involved. As a result, trials have been centralized, often at large academic centers. With cloud-based trial platforms enabling a broader set of sites, we are seeing radiology imaging groups embrace clinical trials, which expands access to patients and diversifies trial populations. The concept of decentralized clinical trials—trials that allow for data collection outside of large medical centers—is now becoming a reality even for trials that involve advanced imaging.
A bright future
Thanks to the convergence of cloud technologies, AI, and refined lung data, we are in a new era for those with chronic lung disease. In this new world, lung intelligence provides fast, precision measures of treatment efficacy so new drugs can blaze through the trial pipeline. Trial sites can now conduct precision imaging with ease, collecting pristine data for sponsors. Imaging data can be collected in hundreds of locations in both urban and rural areas to foster diverse trial cohorts.
The biggest winners in this new era are patients with chronic lung diseases. Clinical trials are more accessible and new life-extending therapies are finding their way to pharmacies faster.
After seeing how quickly the healthcare community could introduce new vaccines to market for Covid-19, there’s a renewed sense of enthusiasm for innovations that accelerate timelines. This enthusiasm, combined with lung intelligence, gives hope for a bright future of fast, efficient, clinical trials.
Photo credit: Jackie Niam, Getty Images