Looking to reduce the length and cost of the more than 300 clinical trials it conducts each year, pharmaceutical giant Pfizer is turning to a cloud platform to manage and monitor every aspect of that process, from trial creation to data collection, cleansing, and submission.
The logistics of clinical trials, which are designed to evaluate the efficacy and safety of new drugs or other treatments in test subjects, are highly complex. They can involve multiple phases; the screening of thousands of patients; the training of various personnel in tens or hundreds of countries, each with their own acceptable-care and safety practices; shipments of perishable goods to sites in those countries; coordination with various contract research organizations (CROs) and other partners; and the analysis of a variety of original and third-party data.
“The complexity shouldn't ever be underestimated because there are so many moving parts and it all revolves around the patient,” says Steve Rosenberg, senior vice president and general manager of Oracle Health Sciences. “And then you factor in the science, which is changing so rapidly right now in terms of the way we target drugs.”
- Related: Oracle Health Sciences
For example, drug development is informed by the still-fledgling therapeutic model known as “precision medicine,” whose practices take into account each patient’s individual variability in genes, environment, and lifestyle, as well as the specific variant or mutation of the disease or other condition a patient has contracted. It’s one of the biggest new frontiers in data science.
Time, Cost Savings
Pfizer has selected Oracle Health Sciences InForm Cloud Service and Siebel Clinical Trial Management System Cloud Service to help simplify how it manages both operational and study data, for in-house as well as outsourced trials, says Rob Goodwin, Pfizer vice president of global product development. Pfizer is known for Celebrex, Lipitor, Viagra, Xanax, and other prescriptions drugs, as well as such over-the-counter brands as Advil, Centrum, and Robitussin.
“Clinical teams will be able to access study data through Oracle’s single platform cloud service, eliminating the need to send data back and forth to CROs, saving us time and reducing the cost of our clinical studies,” Goodwin said in a statement.
Pharmaceutical companies use Oracle Health Sciences InForm Cloud for several purposes. Among them: to collect trial data, as well as data from third-party clinical, genomic, medical-record, and other sources; to consolidate, cleanse, and deliver data to various internal parties; to transform data into a format presentable to the US Food and Drug Administration and other agencies; and to monitor risks to the study represented by each site and the type of data it’s collecting. (A clinical trial site consists of many different people, including investigators and research staffers.)
- Related: Oracle Technology for Healthcare Now
Pfizer will be able to take advantage of more than 100,000 clinical trial sites already trained on the platform, which is integrated with Oracle Health Sciences Central Designer and Oracle Health Sciences Data Management Workbench. In addition, study templates and library management capabilities within Oracle Health Sciences InForm will help Pfizer accelerate the time it takes to build studies and complete trials.
Complementing that platform is Oracle’s Siebel Clinical Trial Management System Cloud Service, whose automated workflows and notifications will help Pfizer track various activities, monitor progress, and keep tabs on costs across all 300 of its clinical trials a year.
Data Challenge
Data analytics are a critical aspect of clinical trials. Consider the process for developing a statin to try to lower high levels of cholesterol, a known cause of heart attacks.
The way the typical phase III clinical trial works, a set of randomly selected people are given the drug and another set are given a placebo—who gets what is unknown to those parties at the time. The pharma company will then apply statistical routines to that data to determine whether or not the drug has made a significant difference in lowering cholesterol levels.
At the same time, trial researchers look at the statistics on any adverse effects of the drug to figure out if the drug itself caused them. Once the researchers are satisfied that the statistics prove that the drug is effective and safe within a certain range of parameters, the company submits its results to the FDA or some other national agency for drug approval.
“Stats rule the world,” Rosenberg says. “But what companies try to do is collect just enough data to prove both safety and efficacy. They don’t want to know what they don’t need to know.”
One reason: Clinical trials cost a small fortune. A phase III clinical trial costs a pharma company anywhere between $30 million and $60 million, Rosenberg says.
“And before or after the clinical trials, the amount of data analytics that goes into these trial databases that they buy or have created or archived is enormous,” he says. “They’re doing epidemiological studies and data analytics across populations, looking for variances in genomic makeup and all kinds of other factors to get an idea of what they should be looking at.”
And those data volumes are getting much larger, moving clinical trials into the realm of specialized data science. “They are collecting all kinds of data and buying all kinds of data from various providers to figure out what to do,” Rosenberg says, “It's huge.”
Cloud computing is thus emerging as a key facilitator, allowing companies such as Pfizer to increase the pace at which trials are concluded while controlling costs.
