AWS technology—combined with the AWS engagement and relationship model—is unparalleled among cloud providers. Given our company’s small size and ambitious goals, AWS is critical to our digitization strategy.
Marcello Damiani Chief Digital Officer, Moderna Therapeutics

Moderna Therapeutics was founded in 2010 with the goal of delivering on the promise of messenger RNA (mRNA) science to create novel medicines for unmet patient needs. The company’s focus areas include infectious diseases, cancer/immuno-oncology, cardiometabolic diseases, and rare liver and pulmonary diseases. Including both internal and partnered programs, there are 16 mRNA drugs in Moderna’s development pipeline: nine prophylactic vaccines, two therapeutic vaccines, and five therapeutics. Seven of these drugs, including an immuno-oncology therapeutic, are already in clinical, or human, study.

Bringing new drugs to market can take a decade and cost in excess of $1 billion. This is true of both traditional, small-molecule pharmaceutical development and the biologics (such as recombinant proteins and monoclonal antibodies) developed by biotechnology companies. The reasons for such long timeframes and high costs include long experimental cycle times and the need to design custom facilities for individual drugs being studied or produced. Another obstacle is the requirement to comply with complex Good Laboratory Practices (GLP), Good Manufacturing Processes (GMP), and U.S. Food and Drug Administration (FDA) 21 CFR Part 11 regulations—referred to collectively as GxP. Both biotechs and traditional pharmaceutical companies face additional challenges, such as siloed data that thwarts collaboration and development of key insights, as well as the inflexibility of on-premises data centers for supporting scientific workflow and high-performance computation.

Moderna’s founders recognized that using mRNA as a drug-delivery mechanism would open the door to a nontraditional drug-development approach that could overcome many of these challenges. “Traditional drugs have to be created in a factory, and then introduced into patients’ bodies,” says Dave Johnson, director of informatics at Moderna Therapeutics. “Messenger RNA drugs tell the patients’ bodies how to produce the drugs themselves, sort of like biologic software.”

To maximize the potential of mRNA-based drugs, Moderna needs IT infrastructure that supports powerful, scalable computing as well as automation and enterprise-wide data integration. 

With these goals in mind, Moderna started using the Amazon Web Services (AWS) Cloud early, and AWS is a core component powering the company’s drug research, development, and production processes. In the AWS Cloud, Moderna has everything it needs to accomplish an IT strategy based on high-performance computation, data integration, powerful analytics, automation of lab and manufacturing processes, and compliance with applicable laws, rules, and regulations. “AWS technology—combined with the AWS engagement and relationship model—is unparalleled among cloud providers,” says Marcello Damiani, chief digital officer at Moderna. “Given our company’s small size and ambitious goals, AWS is critical to our digitization strategy.”

The powerful, flexible computing capabilities of the AWS Cloud help Moderna achieve the speed, scalability, and lower costs that are the central tenets of its business model. Moderna scientists begin developing mRNA drugs by accessing an online catalog to select proteins as starting points for further engineering. This catalog is part of Moderna’s Drug Design Studio, or DDS—a proprietary, internally developed web application, hosted on AWS. DDS uses a computational algorithm to identify mRNA sequences that are optimal for specific medical effects.

The volume of these workloads varies, so Moderna uses Amazon Elastic Compute Cloud (Amazon EC2) with Auto Scaling to bring new instances up or down as needed. By using Amazon EC2 Spot Instances, Moderna incurs costs as much as 90 percent less than for On-Demand Instances. With Amazon EC2 Spot Instances, if Moderna’s not-to-exceed bid price is outbid by another user, the instances shut down, but this does no harm to DDS computational workloads, which can be paused.

The DDS application submits compute jobs to a cluster master microservice that uses the Amazon Relational Database Service (Amazon RDS) to store them and places each job’s associated data in Amazon Simple Storage Service (Amazon S3). Jobs are next pushed into Amazon Simple Queue Service (Amazon SQS). Amazon EC2 instances pull the jobs, process the work, and use Amazon SQS to return results to the cluster master, which then uses Amazon SQS and Amazon Simple Notification Service (Amazon SNS) to inform the DDS application that the work is done. Once the order for a specific mRNA construct is placed, it is sent to Moderna’s preclinical production facility for manufacturing. This facility, heavily automated for high throughput, can create as many as 1,000 of these constructs each month. Moderna or its partners then evaluate the constructs through preclinical, or animal, studies to see if they can be declared development candidates and advanced to the clinical stage.

To produce clinical-grade mRNA therapeutics and vaccines for use in human studies, Moderna is constructing a 200,000-square-foot Good Manufacturing Practices (GMP) facility, set to be operational in 2018. This facility’s quality, supply chain, and GMP processes are driven by a GxP IT environment Moderna has built in the AWS Cloud that enables a fully integrated ecosystem of applications. The cornerstone of this ecosystem is SAP S/4HANA ERP, which uses AWS services to connect with manufacturing instruments, robotics, and other integral systems such as LabVantage Laboratory Information Management System (LIMS), Emerson Syncade Manufacturing Execution System (MES), and an OSIsoft PI data historian solution.

Moderna is also using Amazon Redshift for simple, cost-effective data warehousing to avoid silos and establish a single source of truth for company data. By using Amazon Redshift as a central repository, the company can capture all its data once, with little manual effort, and use business intelligence data-science tools to give its scientists access to informative visualizations that help them find correlations, develop insights, and make decisions. Backups are pushed to Amazon S3, which is highly scalable and offers 99.999999999 percent annual durability, versioning, cross-region replication, and robust backup and disaster recovery capabilities.

Damiani says that AWS Professional Services consultants with life sciences expertise were crucial to helping Moderna build this fully validated environment in the AWS Cloud. “AWS consultants helped us frame our approach to validating this environment and provided essential documentation to help us do so,” he explains. “With their help, we introduced validated systems to the AWS Cloud that have never been put into the cloud before.”

Achieving the computing power we have on AWS in our own data center would cost us millions of dollars.
Dave Johnson Director of Informatics, Moderna Therapeutics

By relying on the AWS Cloud, Moderna can compress the time needed to advance drug candidates to clinical studies; increase the agility of its research, development, and manufacturing processes; and achieve results—such as personalized cancer vaccines—that would have been impossible even a few years ago.

With its IT infrastructure hosted in the AWS Cloud, Moderna has achieved uncommonly fast preclinical drug development. For example, of the seven vaccines that Moderna currently has in clinical trials, one—for Zika virus—took only a year to reach that stage. “With AWS computational and integration capabilities, combined with our proprietary Drug Design Studio, our scientists can have an idea for a unique protein, order it online that day, and be running preclinical experiments on it in less than a month,” says Damiani. “In traditional pharmaceutical research and development, it would take years to get to the point where you are ready for even preclinical studies, much less testing your drug in humans.”

Using an AWS infrastructure also helps Moderna avoid the high costs of traditional drug development. “Achieving the computing power we have on AWS in our own data center would cost us millions of dollars,” says Johnson, who points out that the company’s AWS infrastructure is managed by just one staff member. “Amazon EC2 Spot Instances provide access to huge computational power at a very reasonable cost.”

AWS services are also making it easier for Moderna to meet its goals for data integration, protection, and quality. “Traditional pharmaceutical companies have huge IT organizations struggling to keep up capacity and force-fit backup and disaster-recovery solutions for legacy technologies,” says Damiani. “In the AWS Cloud, we never have to worry about capacity, and backups and recovery couldn’t be simpler.”

AWS also helped Moderna validate its GxP compliance much more easily than would have been possible in an on-premises data center. “The validation process was greatly helped not only by AWS security controls and isolation capabilities, but also by the assistance of AWS consultants,” says Johnson. “It’s incredibly valuable to be able to work with AWS personnel who not only know the technology but also have strong regulatory experience and really understand our industry.”

Damiani says the AWS Cloud is key to the flexibility and agility that has helped Moderna accomplish so much in such a short time. “AWS is a great platform for innovation, because it’s so simple to just try new things and see how they work for you,” says Damiani. “Using AWS has been an important tool to help us grow quickly to a 550-employee biotech with 16 drugs in our pipeline. And soon we will have the ability to produce mRNA medicines customized for the individual genetic signature of patients. That really shows you the power of the AWS Cloud and the versatile ways we are using it as part of our digitization strategy and to advance our business.”

Learn more about biotech and pharma in the AWS Cloud.