If you have an ACS member number, please enter it here so we can link this account to your membership. (optional)

ACS values your privacy. By submitting your information, you are gaining access to C&EN and subscribing to our weekly newsletter. We use the information you provide to make your reading experience better, and we will never sell your data to third party members.



For pharma services firms, big assets can pose major risks

Despite a spotty track record, contract development and manufacturing organizations push on in their acquisition of drugmakers’ factories

by Rick Mullin
October 13, 2019 | A version of this story appeared in Volume 97, Issue 40
A photo of two plant workers walking down the midway of a huge fine chemical factory.

Credit: Evonik | Eli Lilly and Company recently renewed a service contract that accompanied the sale of its Lafayette, Indiana, manufacturing complex to Evonik Industries in 2010. Evonik now also serves about 20 other customers at the facility.


In brief

Drug services providers began acquiring drug companies’ manufacturing sites 20 years ago as, one by one, big pharma firms pared back in-house production. Acquisitions often appeared to be win-wins: the drugmakers continued to be served by the plants, and contractors got to bring in new customers. But converting large factories, some built to produce one blockbuster drug, to multiclient ventures has proved a huge undertaking. While many service providers have built stable businesses at sites acquired from drug firms, ventures continue to fail when the owners aren’t able to bring in new clients.

News 5 years ago that a little-known Chinese company, UniTao Pharmaceuticals, had bought a Boehringer Ingelheim drug plant in Petersburg, Virginia, for $22.5 million raised eyebrows in the pharmaceutical services industry. Few had heard of the company or its parent, Shanghai Tenry Pharmaceutical. But UniTao seemed ready to rescue the 240 jobs the German drugmaker had said it would eliminate when it revealed that the plant would close.

Several months later, though, UniTao announced its decision not to reopen the facility “until business conditions improve.” This left an opening for another buyer, and in 2016 Ampac Fine Chemicals showed up and acquired the plant.

A similar story unfolded this year when Avara Pharmaceutical Services, a new firm that has acquired eight drug company facilities in North America and Europe since 2015, abandoned plants in the UK, Ireland, and Canada. In each case it was unable to secure enough new business. The original owner of the Canadian facility, Novartis, bought it back a year after Avara acquired it.

That sale to the original owner echoed Merck & Co.’s agreement to buy back a manufacturing site in Riverside, Pennsylvania. Merck had sold it in 2008 to the start-up drug services firm Cherokee Pharmaceuticals but took it back 2 years later after Cherokee ran into financial trouble.

And just last month, Famar, a Greek company that ran into financial difficulties following the collapse of its parent firm, said it is in talks to sell five sites in Europe and Canada that it purchased from drug companies. The proposed buyer is Delpharm, a French company that already operates 12 finished-drug facilities in Europe.

These events illustrate the challenges faced by contract development and manufacturing organizations (CDMOs) when buying drug company assets. It’sbeen a trend in the pharmaceutical services field since the major drug firms began to off-load their large plants in favor of outsourcing production nearly 20 years ago.

Selling plants to CDMOs makes sense from a drug company’s perspective. If a drugmaker is going to outsource production, what better place to be served from than the sites where it once operated? From the CDMO’s standpoint, the opportunity to buy drug company assets offers a quick expansion or entry into a new part of the business.

But there are fundamental differences between a major drug company’s manufacturing site and one operated by a CDMO, starting with size.

Big pharma operates huge plants dedicated in many cases to one blockbuster product, whereas CDMOs traditionally rely on agile fine chemical facilities geared toward scaling up and ramping down projects as contract work demands. The foremost challenge to any CDMO buyer, one that has led to a steady parade of failed deals, is the task of converting plants from the big pharma model to the pharmaceutical services model.

At first glance, the proposition of acquiring a large plant from a drug company looks attractive, given the quality of the assets and the contract to supply the seller that is usually included in the deal. “You have time to build your own capacity, build your product line, and not spend a lot of money doing it,” says James Bruno, president of the consulting firm Chemical and Pharmaceutical Solutions. “And these pharmaceutical guys didn’t just build plants. They built Cadillacs while everyone else was building Chevys.”

But drug companies run their plants in a way that would put service companies out of business. “Keep in mind that pharma company procedures at these sites were over the top,” Bruno says. “They were making one product internally and didn’t have to be all that efficient. They weren’t profit driven. Now all of a sudden you’ve got to do it right. You’ve got to lower costs.”

These pharmaceutical guys didn't just build plants. They built Cadillacs while everyone else was building Chevys.
James Bruno, president, Chemical and Pharmaceutical Solutions

Those who have succeeded thus far say that bringing in new business, even when a substantial contract with the seller is in place, is a priority. “Converting these pharma sites to service businesses is a substantial task,” says Andreas Meudt, head of Evonik Industries’ exclusive synthesis business. “Some people are underestimating the effort and what it means to introduce a service culture mentality.”

Evonik took the plunge in 2010 when it acquired Eli Lilly and Company’s Tippecanoe Laboratories plant in Lafayette, Indiana, a huge facility with 860 m3 of pharmaceutical chemical capacity and 2,000 m3 of fermentation capacity. Evonik renewed a supply contract with Lilly last year and says it currently serves more than 20 other customers at the Indiana site.

And Evonik continues to invest in the plant. The firm recently announced spending $40 million in plant improvements across its exclusive synthesis business, with much of it going to Tippecanoe.

Despite a generous 9-year contract with Lilly, Evonik had to get to work immediately to bring in new business, which meant making a lot of changes, explains Clive Whiteside, general manager of the Tippecanoe facility. “One of the features that is particularly challenging when you acquire a larger facility is that each individual customer you take on expects a response that is very flexible, very creative, very engaged,” Whiteside says. Teamwork and project management must be taken to a level that is not typical in a large pharmaceutical plant.

“We invested heavily in recruiting the right people, in training and development, helping all of our employees understand financially how we work,” Whiteside says. Staff need to consider the operational impact of the most routine activities, he explains. “When I make a piece of equipment more reliable, what difference does that make? If I have a deviation in production, what does that cost us?”

Today, the facility has a staff of about 600, compared with 700 when Evonik bought the site. A significant number of Lilly employees were near retirement age when Evonik took over, Whiteside says, leading to an initial wave of departures and recruitment. “I would say that probably at least half of the folks at Tippecanoe now have only known life at Tippecanoe as a CDMO.”

There are still a lot of opportunities, but it has to be the right deal.
Clive Whiteside, general manager, Tippecanoe Laboratories, Evonik Industries

Michael Quirmbach, CEO of CordenPharma, the pharmaceutical services division of International Chemical Investors Group (ICIG), agrees that time cannot be wasted in making the changes necessary to bring new business into a former big pharma plant. “The clock is ticking, and 3 years is a very short time,” he says, referring to a standard duration for a service agreement with a drug company seller. “You need to quickly get people on board who understand the CDMO business.”

This includes marketing staff. In CordenPharma’s case, a team of 20 people coordinates contracts for a network of plants in the US and Europe acquired over 15 years. In that time, Quirmbach says, the company experienced just one unsuccessful venture: the purchase of an active pharmaceutical ingredient (API) site in Cork, Ireland, from Cambrex, another contract manufacturer, in 2006.

Aerial photo of a drug chemical plant in Colorado with mountains in the background.
Credit: CordenPharma
CordenPharma operates a network of sites acquired from major drug companies, including this peptide manufacturing plant in Boulder, Colorado, purchased from Roche in 2011.

Quirmbach says CordenPharma has learned a lot through experience. “Initially, ICIG just acquired these assets thinking all these plants can operate by themselves,” he says. “Then they realized they need a different strategy, a marketing strategy and platform for how they are going to market these services.”

The company also learned that the conversion of big pharma sites to service ventures requires real investment. The Plankstadt, Germany, site that CordenPharma purchased from AstraZeneca, for example, manufactured the cholesterol drug Crestor in finished-dose form. “But it had no development assets,” Quirmbach says. To attract new business, the firm built development labs and small-scale manufacturing capacity for finished drugs. It also built three smaller plants to accommodate the production of drugs for Phase I and II clinical trials.

You need to quickly get people on board who understand the CDMO business.
Michael Quirmbach, CEO, CordenPharma

Similarly, at its Caponago, Italy, site, where AstraZeneca manufactured the anesthetic propofol before CordenPharma bought the site in 2009, the company has invested $25 million in a lab and fill-and-finish plant necessary to attract new customers.

Thermo Fisher Scientific is also operating a network of plants that were mostly acquired from pharmaceutical companies, according to Michel Lagarde, senior vice president of pharmaceutical services.

This month, the firm closed a deal to acquire a plant in Cork, Ireland, from GlaxoSmithKline for approximately $100 million. Thermo Fisher will continue to manufacture for GSK under a “multiyear” contract. The plant, which employs 400 people, produces APIs for cancer drugs, including the breast cancer treatment Tykerb, as well as for drugs for depression, Parkinson’s, and other diseases.

“We are well versed in this strategy of successfully transforming sites and making them viable multicustomer sites,” Lagarde says, noting that of the 45 plants in Thermo Fisher’s pharma services division, 25 are involved in making APIs and finished drugs. Adding a plant means enhancing a global network, he says.

Aerial photo of a drug chemical plant on the coast in southern Ireland with an astonishingly large wind turbine.
Credit: Thermo Fisher Scientific
Thermo Fisher Scientific’s acquisition of a GlaxoSmithKline plant in Cork, Ireland, this year came with a multiyear supply agreement. GSK manufactured the active ingredient for the breast cancer treatment Tykerb and other drugs at the plant.

“If you add a dot on the map, it needs to make strategic sense,” Lagarde says. Plants are added to fill gaps in technology, allow for growth in areas where the company is running at full capacity, or establish a presence in a geographic market that enhances the network. In the case of the Cork site, Thermo Fisher was interested in having API production in Ireland, which has a sizable pharmaceutical industry and maintains a tax advantage.

Lagarde emphasizes that there is no formula for success in acquiring such plants. “I think there is an element of a similar approach, and then there is a little bit of bespoke,” he says. “We are not believers in the random accumulation of sites. We always first apply the lens of ‘Will we be a better owner?’ ”

One consistent element, he says, is a contract with the seller. “We don’t buy empty sites because we don’t think the math works,” he says. “And then we unleash our commercial organization on that capacity.” Lagarde wants to avoid being the owner of what he calls a boomerang plant—a facility that is returned to its original owner when a buyer fails to bring in new business.

Novartis’s repossession of the Canadian finished-drug plant it sold to Avara is the most recent example of a plant boomeranging back to its original owner. The future of Avara’s plants in Ireland and England remain in the balance, though the fate of the latter may soon emerge.


Drug companies have sold dozens of manufacturing sites to contract development and manufacturing organizations in recent years.

Sources: C&EN research. 
Note: List is not comprehensive.

“We received five firm offers for the site and all its contents,” says David Stephenson, senior manager of David Rubin & Partners, a British law firm specializing in insolvency that is overseeing the sale. “We are currently in detailed contract negotiations with the two best bidders.” He adds that there is an ongoing program of layoffs and that all remaining employees will be gone by the time the sale is completed.

Administrators of the Irish site are “seeking offers for the business as a going concern” and dealing with potential buyers, says Edwin Kirker of Kirker & Co., a law firm overseeing the plant’s liquidation.

Avara declined to be interviewed for this article. After jettisoning the three plants, the firm will operate at five sites, all former pharmaceutical company facilities that make finished drugs.

If acquiring a site that comes with contracts carries a risk of collapse, taking on a plant that is already shut down would seem to be a formula for failure. But Aslam Malik, CEO of Ampac when the company acquired the Petersburg site, sees it differently.

Photo of a worker at work at a vessel in a drug chemical factory.
Credit: Ampac Fine Chemicals
Ampac Fine Chemicals acquired this former Boehringer Ingelheim facility in Petersburg, Virginia, from a Chinese firm that purchased the site 2 years earlier.

When Ampac took over the former Boehringer Ingelheim plant, it had ceased operation and had a staff of just 16. “It didn’t come with any business or any contract or anything at all,” Malik says. “It came to us as a plant.”

But the dormant assets were a blessing in disguise, Malik argues. A contract with a seller would have kept the site running, “but in a weird way, that would have been a curse, because that business always runs out,” he says. “When you’re down to a skeleton crew and you don’t have any business, then you’re very hungry. When you’re hungry, you make things happen.”


Malik says his small crew was highly skilled; in addition, he brought back former employees that wanted to stay in the area. Transfers from the company’s California and Texas operations helped establish a service mentality and Ampac’s corporate culture, he says.

One of the plant’s enticements for Ampac was controlled-substance production capabilities. They also appealed to Noramco, another pharmaceutical chemical maker. In need of additional capacity, Noramco hired Ampac last year to manufacture controlled substances in Petersburg for uses such as treating drug addiction and attention deficit hyperactivity disorder. “Noramco was a start,” Malik says, “but a lot more products are coming in.”

After Ampac was acquired last year by the South Korean conglomerate SK Holdings, it began networking with another ex–big pharma plant—a facility in Swords, Ireland, built by Bristol-Myers Squibb and later bought by SK Biotek. Malik, who now heads SK’s pharmaceutical chemical division, notes that the Ireland site was BMS’s launch facility for new drugs. It maintains the original staff.

“Those people know how to do scale-up,” Malik says. “Their quality and technical engineering are a plus.”

SK continues to produce the Swords plant’s portfolio for BMS on a contract basis. In addition, Malik says, since SK took the plant over, more than a half-dozen new customers have been brought in.

Flamma, an Italian CDMO, also took on a shuttered pharmaceutical chemical plant by acquiring Teva Pharmaceutical Industries’ chemical synthesis center in Malvern, Pennsylvania, earlier this year. According to Kenneth Drew, senior director of North American sales and business development, the company was looking for a development-scale facility in the US. The site was also attractive for its high-potency API production capacity.

Drew explains that the site had been shut down for a year when Flamma acquired it. “The smart thing is we hired back six people, pilot-plant operators,” he says. Flamma currently has 10 employees at the facility and hopes to have 60 by 2022. The company will have staff from its facilities in Italy and China work temporarily in Malvern, Drew says.

He says the deal netted capacity at a lower cost than building a new pilot plant, which Flamma considered its only option before the Malvern site came on the market. The up-front investment in a fully operational plant is higher, he says, but there is an advantage to being able to put it right to work. “You’re already where you need to be,” Drew says.

There are other variants on the theme of acquiring big pharma assets. Some companies target technology that might boost a service offering. Hovione’s 2008 acquisition of a Pfizer plant in Cork is a prime example. While the Portuguese company benefited from an expansion of API capacity—it recently opened a second production line that Pfizer had mothballed—it had its eyes on a $90 million spray-drying plant that Pfizer had built but never used.

BioVectra, a CDMO based in Prince Edward Island, acquired a Sunovion Pharmaceuticals plant in nearby Nova Scotia in 2014 with the intention of growing a nascent microbial fermentation business. The deal did not come with a contract with the biotech firm—in fact, BioVectra had to convert the plant from synthetic chemical production to fermentation.

Heather Delage, vice president of business development and general manager of the Nova Scotia site, says the 4,600 m2 facility, built with several small-scale synthesis suites, had an ideal layout and utilities for BioVectra’s planned fermentation operation. Its proximity to Prince Edward Island and location on the mainland justified a major conversion of assets.

The company went into the venture with a production contract that proceeded to fall through when demand for the client’s drug fell short. BioVectra pushed forward, landing three other contracts for fermentation. It’s now scaling up a multiclient services facility for complex biologics. The site currently has 70 employees.

CDMO executives agree that the market for pharmaceutical plants remains buoyant. “We still see opportunities where pharma companies want to off-load these assets,” CordenPharma’s Quirmbach says. And many potential buyers still line up when major assets come on the market. Last month it was reported that six potential buyers have toured a 400-employee site in Grimsby, England, that Novartis is selling.

CordenPharma has not ruled out further acquisitions of pharmaceutical company assets, Quirmbach says. “Today, if we could find an injectable facility in the US, it is something we would definitely look into,” he says. “Or if we could find another good high-potency API plant in Europe.”

Whiteside at Evonik’s Tippecanoe site says the firm is also reviewing a steady flow of proposals to acquire big pharma plants. “There are still a lot of opportunities, but it has to be the right deal,” he says; for example, the plant should offer in-demand technologies that complement the company’s current operations. Acquiring a large site merely to bulk up on capacity could set a service firm up for a fall.

Bruno, the consultant, sees enthusiasm for acquisitions continuing, but he also sees an end to the trend. “I think it’s the preferred way of getting assets,” he says. “But we are going to eventually run out of plants to sell.”


This article has been sent to the following recipient:

Chemistry matters. Join us to get the news you need.