Patricia F. Fitzpatrick Dimond Ph.D. Technical Editor of Clinical OMICs President of BioInsight Communications
Firms adopt technology for varied advantages while they work on adapting to concerns over data privacy.
Pharma and biotech companies have so far tended to use cloud computing services for drug development research, not clinical trials. IBM and other smaller tech companies would like to change all that, though.
Additionally, applying cloud computing for data analysis to aid patient care is catching on. As cloud computing extends to clinical data analysis, which would be considered private information, security will loom larger as an issue for drug development companies.
The benefits of using the cloud to store, manage, and analyze clinical data are similar to those for other drug development work. The cloud offers time on gigantic, dispersed infrastructures on a pay-as-you-go basis. It is estimated that 8–12% of clinical trial costs come from the need to move data around among various trial sites, comparing the data as it is being created, according to Erich Clementi, vp of strategy and GM of enterprise initiatives at IBM.
Additionally, for projects that require heavy data crunching, cloud computing enables vast amounts of processing at a lower cost. For example, Jeffrey T. Leek, Ph.D., assistant professor at Johns Hopkins Bloomberg School of Public Health, and colleagues used an internally developed, open-source cloud-computing platform called Myrna for calculating differential gene expression in large RNA sequencing datasets.
Running an analysis for a single RNA sequence on one laptop could take up to three weeks to complete, Dr. Leek explained. In contrast, by renting computers for cloud computation services for $65, Dr. Leek said he could get results back in an hour and 45 minutes.
Private Data Means Need for Security
One component of using cloud computing resources is the extent to which it can be “anonymized,” Dr. Leek told GEN. “For most of our analyses we have used publicly available data. There are a host of privacy issues that will come into play once clinical patient data comes into the picture. It will be a matter for the FDA, NIH, and institutional review boards to consider when evaluating projects.”
Mika Nuutilainen, director, product development at CRF Health, in an interview with GEN said, “For clinical trial data management, I do not yet see public clouds as a viable solution. The current acceptable way of proving correct installation and operation of clinical trial data management systems is still very traditional and highly focused on hardware units and their location.” CRF Health provides technology for ePRO (electronic patient reported outcomes), a method used in clinical trials to collect information on a patient’s health status directly from the patient.
Nuutilainen predicts that “private clouds” will be used, but added, “that is a solution where a dedicated server environment is boosted with automatic management software to allow, for example, fast disaster recovery from one site to another. ‘Community clouds’ could be used if clinical companies can co-operate that much and there is a technology-oriented company driving the development.
“I personally do not consider the concept of ‘private clouds’ as cloud computing, instead I consider it a ‘virtualized environment.’ You still need to buy, install, and operate your environment, so a majority of the benefits from cloud computing is not achieved,” he pointed out.
Clinical Trial Cloud Solutions
Stuart Henderson, IBM’s Americas life sciences R&D leader, believes that cloud computing for clinical trials has unique advantages. IBM offers “Clinical Cloud” to the drug development industry. The aim was to create a secure environment where partners can work together with quicker and easier access to data.
The company said that the Clinical Cloud has the requisite features for an “intelligent” platform: multitenant security, infrastructure, process integration and orchestration, a clinical application suite, compliance, analytics, collaboration, help and support, data sharing, and file sharing.
Multiple companies smaller than IBM are stepping into the clinical trials arena. Wipro Technologies offers Wipro Clinical Collaboration Portal to drug developers, CROs, clinical sites, and regulators. The aim is to “significantly improve collaboration capabilities for multiregion clinical trials.”
The Wipro platform was thus designed to reduce the clinical trial cycle time by 20–30%, the company reported. The portal speeds up communication and document exchanges among the sponsor organization, CRO staff, clinical site coordinators, and principal investigators.
PharmaPros provides technology solutions for data and workflow management in clinical trials. Its Dataflow Manager™ uses cloud computing, enabling trial managers to make more rapid and better-informed decisions during a trial.
“We now have data streaming in from many disparate sources: Interactive voice response systems, electronic data capture, ePRO, laboratory data, imaging data, safety signaling systems,” PharmaPros’ head of strategic development Brion Regan told GEN. “What we’re left with is a challenge to view clinical trial data holistically, in stream, and in time to make critical decisions about the conduct of the trial.”
“Because the data is broken up into disparate technologies and sources, the ability to quickly see trends, identify safety issues, or simply report on trial progress has become extremely difficult.” As a result, he noted, many life science organizations have taken on the herculean task of integrating all this disparate data into one repository in an effort to gain visibility.
This is not, in PharmaPros’ opinion, a realistic solution for in-stream trial management. It believes the focus should instead be on gaining visibility without moving data. “What we’re seeing is companies attempting to integrate data, only to discover it’s incomplete or not ready; then they try again. It’s become a burdensome cycle for statisticians. It’s also just extremely inefficient.
“For driving decisions and operations, clinical data doesn’t actually need to be integrated,” Regan noted. “Cloud-based solutions can leverage the information that resides in these disparate systems and deliver an integrated view of the data.”
Cloud-based computing has already become integrated into drug R&D at many levels. In spite of security and privacy concerns, the benefits of leveraging cloud computing in the clinical trial process have proven to be enticing to drug developers, CROs, clinical site managers, and others involved.
The ease of information sharing as well as enhanced collaborative and analytical capabilities are very attractive given the international nature of clinical trials. These advantages will prompt many companies to adopt it and adapt to it.
Patricia F. Dimond, Ph.D. (firstname.lastname@example.org), is a principal at BioInsight Consulting.