Merck puts big data to work to solve vaccine manufacturing concern

Big data analytics have become a common consideration in the R&D part of the pharmaceutical industry, allowing researchers to search and share remarkable amounts of data points. But Merck & Co. has turned to the process to solve a manufacturing problem and potentially save hundreds of millions of dollars in the process.

According to InformationWeek, the Whitehouse Station, NJ-based drugmaker was having issues with yield rates on a vaccine in 2012 and was going through the painful process of analyzing spreadsheets of process histories and other data to try to solve the problem. It meant comparing only a couple of batches at a time and was going to take a very, very long time. Jerry Megaro, Merck's director of manufacturing advanced analytics and innovation, thought there had to be a better way and started working with a massively scalable distributed relational database. But then he learned that Merck's R&D operation could hook him up with cloud-based Hadoop computer, which could do all the things big data analysis is known for.

"We took all of our data on one vaccine, whether from the labs or the process historians or the environmental systems, and just dropped it into a data lake," George Llado, vice president of information technology at Merck explained to InformationWeek.

In three months, Merck was able to do 15 billion calculations and more than 5.5 million batch-to-batch comparisons. It discovered there were ties between what was happening in the fermentation phase of vaccine production and what Merck was getting in its yields during the final purification step. "That was pretty powerful, and we came up with a model that demonstrated, quantifiably, that specific fermentation performance traits are very important to yield," says Megaro.

Merck is now working to prove in a test lab what the big data analysis has shown it to look for. If confirmed and it changes processes, Merck will ask regulators to approve a new manufacturing process. It is using the analysis techniques on a version of the vaccine not yet approved and intends to put them into practice for other products.

Analyzing data, even in real time, is something that drugmakers are now playing with as they look for every kind of way to cut manufacturing costs. GlaxoSmithKline ($GSK) since 2011 has been working with McLaren Applied Technologies, part of its Formula 1 racing organization that helps the team sort through thousands of pieces of info to make split-second decisions to help its racers win. The partnership has allowed GSK to achieve a 50% production improvement on its Breo Ellipta inhaler plant in the U.K.

- read the InformationWeek story