By Allison Proffitt 

August 10, 2017 | After years of development, Oracle Health Sciences released the Clinical One platform last month. It’s a “reimagining of the way that our… clinical trial software is consumed by our customers,” Steve Rosenberg, general manager at Oracle Health Sciences told Clinical Informatics News. And it’s driven by Oracle’s view of the cloud.

Since the beginning of data capture, pharma companies licensed software tools for each step in the clinical trial process, Rosenberg says. “They would piece together different applications from potentially different vendors, and put all that together and have different teams that implemented that on behalf of a study or a program. It took a lot of time.”

Rosenberg was a senior vice president at PhaseForward when Oracle acquired the company in 2010. He remembers when vendors began bringing tools together and integrating them, hoping to build a one-stop-shop. The market, though, did not catch on.

“There was a still a mentality of, ‘Where should I get my EDC tool?’ ‘Where should I get my randomization tool from?’ etc,” Rosenberg said.

Things changed as cloud computing took off and platform-as-a-service and infrastructure-as-a-service became well-known acronyms. Now Oracle—“big Oracle” Rosenberg says, distinguishing it from the 2,000-person Oracle Health Sciences Global Business Unit—is working to redefine the cloud as well as clinical trials.

The New Cloud

Oracle has invested in infrastructure, hardware, networks, tools, and data centers. It’s a different model from what Amazon, IBM, and Microsoft do, Rosenberg argues. Oracle not only offers compute, but includes tools for security, identity management, performance, disaster recovery, and availability, he says. Built on Oracle’s cloud infrastructure, Clinical One has a common interface and handles all data security and disaster recovery.

With this “new cloud” at its disposal, Oracle Health Sciences can offer a set of microservices in its Clinical One platform that are easily assembled to support a program or trial. Microservices are small, modular services—much more granular than an EDC tool or a randomization tool. Instead a microservice might be a type of form, or an image processing step.

Rosenberg believes that cloud architecture will prompt sweeping changes in how trials are designed and how pharma works.

“You’ll be able to map out the behavior you want of a study, and [the platform] will assemble the appropriate microservices,” Rosenberg says. He acknowledges that it will require a shift in thinking, but also believes the shift will remove so much time and cost from the system. “The idea with this is companies will be able to design their own trials by configurations, and not have to rely on the service industry if they so choose,” he says. “Of course,” he adds, “we’ll continue to offer the services.”

Over time, Rosenberg predicts a change the way the pharmaceutical industry is structured. “Rather than have separate teams [for randomization and drug supply; EDC; enrollment; etc.]… these teams will be blended, sort of like how agile development is done in the IT industry,” he says. “You’ll have cross-functional teams that’s working on a study and working together.”

Oracle isn’t the only company hoping to ease the design of trials, and cover the full process from start to finish, but Rosenberg is skeptical of any other company’s ability to acquire the pipeline pieces and integrate them effectively.

“In this new world of true cloud computing, it’s one thing taking older software and just sticking it in a datacenter and saying, ‘I now have cloud.’ It’s a whole other thing delivering on the whole elasticity, rapid deployments, rapid upgrades, automatic validation, and all the other things that come with it… It’s hard to imagine that someone out there has software that’s architected properly to do this.”

Above all, Oracle is a software engineering company, Rosenberg argues. He led a team of over 100 people working exclusively to build Clinical One, he says. “We’re big. … we get to make investments that other players can not in this space.”

Starting Point

But the paradigm shift Rosenberg anticipates is not here yet; Rosenberg calls the first release from Clinical One a module, not unlike competitive terminology. The platform’s foundation is completed but all the tools aren’t yet built out, he says. The changes in thinking that he expects may come after more tools are available.

The first Clinical One offering is the Randomization and Supplies Management module. “We think that’s the area most in need,” Rosenberg says. “That alone could be disruptive because of the way pharma now consumes and deals with IRT [interactive response technology]… They treat it as a service; they have multiple vendors.”

Even as pharma companies have begun building parts of the clinical trial process in house, that hasn’t happened with randomization and drug supply, Rosenberg said. “Some of that is because it’s risky. If you don’t get [randomization and drug supply] right, it can cost you a lot of money and jeopardize patient safety. But we think there’s a huge opportunity to change how these systems are fielded, to have pharma companies standardize on one.”

The randomization module for Clinical One is self-service. Everything – from initial study design and start-up to mid-study changes – can be done independently, in real-time, by study team members. “Unlike our… current IRT,” Rosenberg said, “[Clinical One] has a whole design line where you can configure a study to do what you need it to do, as opposed to having to rely on the vendor to do it.”

Study designers can model multiple design options before deciding, and make mid-study changes, for example, protocol amendments, on their own. The tool can define and configure a randomization strategy by study, site, or region, and assign it to specific events in the study. It allows study teams to constrain depot staff who manage the drug inventory and ship drug kits to patients and sites, limiting the data they can see and manage.

Clinical One will be cheaper than existing solutions, Rosenberg argues. It will be a great fit for small CROs and big pharma both, he says, because the microservices architecture scales so efficiently. “We think it’s going to be cheaper than what the industry is buying now because we’ll get scale for almost no more money.”

And this is just the starting point. “We’ll continue to add functionality over time that rounds out the functionality of everything you need in a clinical trial. That will happen rather rapidly,” he says. “The other thing about this new cloud, microservice architecture, you don’t have to wait a year for the next release. You can just start to add microservices that do pieces of functionality that can be taken advantage pretty quickly. You really start to snowball increased functionality and get much more value for the customer.”