Clinical trials require the highest possible data quality. Only then can they deliver clinically meaningful results that lead to the development of effective drugs. Cyntegrity is one of the leading providers of technologies for risk-based quality management in the biopharma industry. The medium-sized enterprise helps pharmaceutical companies comply with this process, which is driven by the regulatory authorities for clinical trials. This involves identifying and evaluating critical data and processes as the basis for subsequent decisions on preventive and corrective measures—and it requires a lot and fast computing power. Cyntegrity wanted to meet the growing demand with scalable and secure technologies and reduce IT costs despite a higher number of configurations—which is why it relies on a suite of Azure services for its software-as-a-service solution.
The challenge: Large quantities of data and little scalability
Artem Andrianov, co-founder and CEO of Cyntegrity, compares risk-based quality management to driving a car: “Modern vehicles feature a growing number of tools that provide risk indicators. For instance, the navigation system warns the driver of traffic jams and problems on the roads by collecting, analyzing, and presenting information in real time. This smart, data-driven basis for decision-making helps people drive more safely. Risk-based quality management is also about making data-driven decisions—while bearing the risks in mind.”
When it comes to clinical trials, that is exactly what Cyntegrity enables its customers, whether large pharmaceutical companies or small biotech companies, to do. The biggest challenge here is dealing with data that is inaccurate, incomplete, or inconsistent—known as dirty data. “On average, each pharmaceutical company operates five to seven database systems. This is where they store the data collected from patients on side effects, blood and heart values, vital parameters, documentation, lab data and much more. But these systems don’t talk to each other,” Andrianov says. “What’s more, companies usually have several studies running in parallel over a period of one to three years, and data is sometimes collected decentrally by different institutes and laboratories or by patients themselves. This can lead to inconsistencies and leaves lots of room for errors. In the worst case, clinical trials have to be terminated or can’t be used for drug applications as a result.”
Cyntegrity’s software-as-a-service (SaaS) solution deals directly with this problem by monitoring all potential risks and guaranteeing high data quality. The solution takes the data provided by customer companies via interfaces from their own back-end systems and harmonizes it in the MyRBQM® Portal before analyzing it for performance, speed, risks, and quality. Highly sensitive patient data is always anonymized before processing. If the system detects anomalies that are about to exceed preset thresholds, it triggers an alarm and provides information via an interactive dashboard. That way, customers are alerted sooner to issues such as problems with a test group’s composition or data collection, so they can proactively mitigate risks and let the study continue.
“As the technical foundation for our platform, we relied from the outset on a solution that offers particularly compelling data security: Azure Virtual Machines,” Andrianov says. “But we quickly realized that one can’t build a business on that, because a virtual machine can only run a maximum of five studies at a time before its capacity is exhausted.” Cyntegrity’s first task for new customers is usually to analyze a single study. Once customers have seen the benefits, the studies soon multiply. “For smaller companies it can be 25 studies, and for large companies it can be up to 250 studies. All that data needs processing,” says Dr. Vasian Cepa, Head of Research & Development at Cyntegrity. “So our decision to go with the Azure Cloud was primarily a strategic one."
The solution: More capacity for further studies thanks to Azure Functions
Today, Cyntegrity uses everything Microsoft Azure has to offer: Azure DevOps services simplify deployment, while all applications run on Azure Functions and Azure App Service. “Regulatory requirements are now easier for us to meet,” Dr. Cepa says. “The new authorization management based on Azure Active Directory lets us document all changes to the system and make them traceable. We can always see the status of a deployment and also when it was last changed and by whom. Meanwhile, Application Insights provides simplified and improved monitoring on our MyRBQM® Portal. Overall, this gives us better quality with less effort because we don't have to constantly upgrade hardware or organize infrastructure for our internal configurations anymore. We can simply access Azure's unlimited resource pool. This allows us to save around 90 percent of IT costs.”
In addition, Cyntegrity benefits from the new solution’s scalability and flexibility, as resources are now no longer limited. “We can process multiple studies or just one using a single Azure Function. We can easily add more Azure Functions,” Dr. Cepa says. “This means customers can decide for themselves how many studies they want to run simultaneously.” This ability to scale provides not only additional resources but also new opportunities for internationalization. Since the data for each study is processed in isolation from any others in accordance with regulations, Cyntegrity can deploy its solution in different regions around the world. Data from customers in the United States is processed in the US, while European customer data is processed in Europe.
Above all, however, the new solution ensures transparency within the company. “We built all the processes in our Research & Development team around Azure DevOps. The work all our various teams do, from customer service to change management and quality assurance, takes this into consideration,” Dr. Cepa says. “That ensures we’re ready for new demands and lets us improve customer service, support, and the user experience.”
The first innovation in the portal has already been implemented: customers can directly see which resource pool is available for specific studies, configure Azure Functions themselves, and get faster and more reliable insights. “Except for two hours, we’ve been online continuously for the past five years. Our availability is 99.9 percent,” Andrianov reports. “That’s what people expect of us, along with a high level of security. In our industry, given the sensitivity of the data, trust is a key issue, and that’s where Microsoft is the best certified vendor.” Various security tools such as Microsoft Defender for Cloud are an integral part of the new infrastructure. Azure Front Door secures data coming into the portal, while Azure Virtual Network isolates data from the internet. “The big advantage is connectivity,” Dr. Cepa says. “Standard Azure components can be combined with any Azure security tools as desired.”
Andrianov is enthusiastic: “There are so many benefits in the cloud, it’s hard to identify one that stands out. What is certain is that we’ll continue to rely on Azure.” In the future, it will be possible to share information on hazard detection methods and occurrence. For example, when the protocol of a study is uploaded, the system will automatically provide any risks for the study that machine learning algorithms have identified based on previously collected studies—smartly and accurately. This will enable even faster identification of obstacles in clinical studies as well as further increases in data quality.
“Naturally, our customers expect a high level of security. Within our industry, given the sensitivity of the data, trust is a key issue. And Microsoft Azure helps us build precisely that trust in the security of data.”
Artem Andrianov, PhD, Co-Founder and CEO, Cyntegrity Germany GmbH
Follow Microsoft