A new way to make small batches of biopharmaceutical products on demand

These drugs are normally manufactured in large facilities dedicated to a single product, using processes that are difficult to reconfigure. This rigidity means that manufacturers tend to focus on the drugs that many patients need, while drugs that can help smaller populations of patients may not be manufactured.

To help make these drugs more available, MIT researchers have developed a new method for rapidly producing biopharmaceutical products on demand. Their system can be easily reconfigured to produce different drugs, allowing for flexible switching between products based on their needs.

"Traditional bioproduction relies on unique processes for every new molecule produced," says J. Christopher Love, professor of chemical engineering at MIT and a member of the Koch Institute for Integrative Cancer Research at MIT. "We have demonstrated a unique hardware configuration that can produce different recombinant proteins in a fully automated, hands-free manner."

The researchers used this manufacturing system, which can be installed on a laboratory bench, to produce three different biopharmaceutical products and have shown that they are of comparable quality to commercially available versions.

Love is the lead author of the study, which appears in the October 1 issue of the journal Nature Biotechnology. The main authors of the journal are graduate students Laura Crowell and Amos Lu, and research scientist Kerry Routenberg Love.

A streamlined process

Biopharmaceuticals, which usually need to be injected, are often used to treat cancer, as well as other diseases, including cardiovascular disease and autoimmune diseases. Most of these drugs are produced in "bioreactors" where bacteria, yeasts or mammalian cells produce large quantities of a single drug. These drugs must be purified before use, so that the entire production process can include dozens of steps, many of which require human intervention. As a result, the production of a single batch of a drug can take weeks or even months.

The MIT team wanted to develop a more agile system that could be easily reprogrammed to quickly produce a variety of different drugs on demand. They also wanted to create a system that would require very little human supervision while maintaining the high quality of protein required for use in patients.

"Our goal was to automate the whole process. Once you have set up our system, you press "go" and you come back a few days later.

A key element of the new system is that researchers used a different cell type in their bioreactors: a yeast strain called Pichia pastoris. Yeasts can begin to produce proteins much faster than mammalian cells and can reach higher population densities. In addition, Pichia pastoris secretes only about 150-200 proteins, compared with about 2000 for Chinese hamster ovary (CHO) cells, which are often used for biopharmaceutical production. This greatly simplifies the process of purifying drugs produced by Pichia pastoris.

The researchers also significantly reduced the size of the manufacturing system, the ultimate goal being to make it portable. Their system consists of three connected modules: the bioreactor, where the yeast produces the desired protein; a purification module, wherein the drug molecule is separated from the other proteins by chromatography; and a module in which the protein drug is suspended in a buffer that stores it until it reaches the patient.

In this study, researchers used their new technology to produce three different drugs: human growth hormone; interferon alpha 2b, used to treat cancer; and granulocyte colony stimulating factor (GCSF), which is used to stimulate the immune system of patients receiving chemotherapy.

They discovered that for the three molecules, the drugs produced with the new process had the same biochemical and biophysical characteristics as the commercially manufactured versions. The GCSF product has behaved in a manner comparable to an authorized product of Amgen when it has been tested on animals.

To reconfigure the system to produce a different drug, it is sufficient to give the yeast the genetic sequence of the new protein and replace some modules for purification. In collaboration with colleagues at the Rensselaer Polytechnic Institute, the researchers also designed software to develop a new purification process for each drug they wish to produce. Using this approach, they can suggest a new procedure and start making a new drug in about three months. In contrast, the development of a new industrial manufacturing process may take 18 to 24 months.

Decentralized manufacturing

The ease with which the system switches between the production of different drugs could allow many different applications. On the one hand, it could be useful for the production of drugs for the treatment of rare diseases. At present, these diseases have few treatments, because it is not interesting for pharmaceutical companies to dedicate an entire factory to the production of a drug that is not widely needed. With the new MIT technology, small-scale production of these drugs could be easily achieved and the same device could be used to produce a wide variety of such drugs.

Another potential use is the production of small amounts of drugs needed for "precision medicine", which is to give cancer or other disease patients drugs specific to a genetic mutation or other characteristic of their disease. Many of these medications are also needed only in small amounts.

These machines could also be deployed in areas of the world without large-scale drug manufacturing facilities.

"Instead of centralized manufacturing, you can move to decentralized manufacturing, which allows you to have two systems in Africa, then bring those drugs to these patients rather than do everything in North America, keep cold ", said Crowell.

This type of system could also be used to quickly produce the drugs needed to respond to an outbreak such as Ebola.

Researchers are now working to make their device more modular and portable and to experiment with the production of other therapies, including vaccines. The system could also be deployed to speed up the process of developing and testing new drugs, the researchers said.

"You could prototype a lot of different molecules because you can actually create processes that are simple and fast to deploy, so we could look at many assets in the clinic and make decisions about which ones work best clinically, because we could potentially the quality and quantity needed for these studies, "says Routenberg Love.

Explore more:
Hitchhikers limit the shelf life of drugs

More information:
On-demand manufacturing of clinical grade biopharmaceuticals, Nature Biotechnology (2018). DOI: 10.1038 / nbt.4262, https://www.nature.com/articles/nbt.4262