A Setback For Chinese Drug R&D

In Shanghai’s Zhangjiang Hi-Tech Park, shiny new R&D centers built by the world’s largest pharmaceutical companies tower near each other, even side by side in some cases. Large drug companies have downsized their Western R&D operations in recent years, but the opposite is happening in China, perhaps the world’s most promising drug market.

Novartis, which already employs more than 400 scientists in Shanghai, is completing the construction of new facilities to accommodate more researchers. Merck & Co., with 200 researchers at work in Beijing, plans to increase its R&D headcount there to 600 in coming years and invest $1.5 billion in research facilities in the country.

Among such companies, GlaxoSmithKline stands out as perhaps the most confident in trusting China-based researchers to conduct world-class, cutting-edge research. Since announcing in 2007 that it would set up an R&D base in Shanghai, GSK has built an organization of 400 scientists in the city. The GSK group in China is responsible for managing the company’s neuroscience R&D around the world, from target identification to clinical trial management.

But last month, after an internal investigation, GSK fired Jingwu Zhang, the manager of its Shanghai research center and the very person it had entrusted to establish R&D in the country, after determining that he had misrepresented data in a 2010 scientific paper. The move shocked pharmaceutical researchers in China, who fear that the country’s growing role in global drug innovation may now be jeopardized.

“This casts a pall over Chinese researchers,” says Yi Zhun Zhu, dean of the School of Pharmacy at Shanghai’s Fudan University. “People will look at what happened here and start to believe that research in China is all about data manipulation.” A cardiologist trained in Germany, Zhu leads a large and well-funded group at Fudan that works on cardiovascular research with both Chinese and multinational drug companies.

Zhang’s dismissal came at a time when researchers in China have yet to prove their worth, according to Greg B. Scott, chief executive officer of ChinaBio, a Shanghai-based company that organizes conferences, provides intelligence on China’s biotechnology industry, and supports Chinese biotech start-ups. “If something like that had occurred in the U.S. or Europe, the company involved would just have said that a researcher made an error,” he says. “But this happened in China, so it affects the perception of the whole country.”

The incident fits into a view of China as the source of many false or erroneous research reports. In a January 2010 editorial, the British journal The Lancet warned that China needed to act against excessive scientific fraud in the country (DOI: 10.1016/s0140-6736(10)60030-x).

It is unclear what exactly led GSK to fire Zhang. The company says he misrepresented data in a paper published in 2010 in Nature Medicine (DOI: 10.1038/nm.2077) that concerned a hypothesized genetic link for multiple sclerosis.

“We are committed to the highest ethical and scientific standards, and regulators, physicians, and patients can have confidence in the research we carry out,” a GSK statement says. In response to questions from C&EN, the firm said, “We are confident in the thorough investigation we conducted and the actions we have taken as a result of our findings. We will not tolerate activity and behavior that falls short of the high standards expected from our employees.” The firm says it requested retraction of the paper, but it was still available on Nature Medicine’s website at C&EN’s press time. A spokeswoman for Nature’s publisher would not reveal whether it was investigating the paper.

In an interview with C&EN, Zhang appears both shocked and puzzled by GSK’s hard line. The company, he says, told him he was fired for “influencing the investigation,” a charge he firmly denies.

In late May, he recalls, GSK instructed him not to contact certain researchers in China but didn’t tell him why. Not aware that he was under investigation as well, Zhang disregarded the directive and met with the researchers—who included coauthors of the paper—along with another manager who was not being investigated. “I was trying to explore what issues might have triggered the investigation,” he says. “This was a manager’s natural reaction.”

Regarding the problems in the paper, Zhang says it is unfair to pin the blame on him. “I do take certain responsibility for the paper; I was the corresponding author,” he says. As a corresponding author, he explains, his role was to set the study’s framework, polish the flow of the paper, and organize its presentation. He did not check the data in minute detail, nor was it his job to do so, he says.

Five data sets concerning animal models are the “substance of the paper,” Zhang claims. A sixth figure is labeled as data from multiple sclerosis patients, but the other authors of the paper inadvertently plugged in data from healthy patients, Zhang says. Baylor College of Medicine, in Houston, provided the patient data, he notes, because multiple sclerosis is uncommon in China. Healthy patients were a control group in the experiments, he explains, adding that he left the details to other researchers.

Zhang has been conducting research on multiple sclerosis since 1986. “My whole career is based on this disease,” he says. Roland Martin, a professor and consultant in neuroimmunology and multiple sclerosis research at University Hospital Zurich, finds it hard to believe Zhang would cheat in a paper. “I know Dr. Zhang as a very thorough scientist and cannot imagine any willful placing of data from healthy controls instead of data from patients,” he says.

Guidelines for journals from the American Chemical Society, which publishes C&EN, hold that corresponding authors don’t have to personally check all the data in their papers but are nonetheless responsible for accuracy, according to John P. Ochs, ACS Publications’ vice president of strategic planning and analysis. “What this means in practical terms is that corresponding authors need to take whatever measures they think best to satisfy themselves that the data underlying their paper are accurate and complete,” he says.

Yet inadvertent mistakes occur even in high-profile journals, Martin observes. “As long as they are corrected and do not invalidate the entire study, publishing of a correction or erratum is the way to handle this.”

When Zhang, 57, joined GSK in 2007, he was a successful academic researcher with many accomplishments to his name. He had returned to China in 2002 from the U.S. where he was a tenured professor at Baylor. “I could have stayed there and enjoyed a secure and interesting life,” he tells C&EN. The momentous changes taking place in China, which in the past few years has emerged as one of the largest contributors to science journals, compelled him to return, he says.

During Zhang’s first five years back in China, he set up two research institutes. He was the founding director of the Institute of Health Sciences in Shanghai, a medical school that is affiliated with both the Chinese Academy of Sciences and Shanghai Jiao Tong University. He was also cofounder and codirector of the Pasteur Institute of Shanghai, a joint undertaking of the famous Paris institute, the Shanghai government, and the Chinese Academy of Sciences. Zhang says he recruited numerous scientists internationally to staff both institutes.

He agreed to join GSK, his first job in the pharmaceutical industry, because he was enticed by its vision of a “global end-to-end R&D center in China, not some small symbolic outpost of the company,” he recalls. He was the company’s first researcher based in the country. Over six years, he hired 400 scientists, half of whom he recruited from outside China.

GSK gave Zhang much leeway to set up the China operation and direct research. “I decided to keep my academic attitude and focus on science and innovation, because I believed that’s the best way to produce new medicine,” he says. “We published about 60 papers in six years.” The rise of GSK’s neuroscience effort in China happened while the firm was closing neuroscience drug development in other places, notably the U.K.

Along the way, Zhang seems to have made enemies within the firm. On May 31, anonymous contributors to a Chinese-language website in the U.S. posted a detailed update on GSK’s internal investigation. Signed by “former GSK employees,” the post revealed that Zhang had been placed on administrative leave, something that indeed happened that day. It ended with a taunt that Zhang would now pay for his arrogance. Zhang says he has no idea who wrote the post but notes that he did dismiss scientists while at GSK.

Zhang’s firing is a blow to all pharmaceutical researchers in China, says the manager of a science park that hosts many biotech firms in eastern China. He asked that his name be withheld because disclosure could harm his career. “Dr. Zhang was assigned huge responsibilities with the management of research in an entire disease area,” he says. “This incident will impact not only researchers at multinational companies, but also those in Chinese research institutes, and even at contract research organizations, because they’re all staffed with people with essentially the same background.”

Multinational companies will be more cautious about expanding their R&D operations in China, predicts Scott, the China­Bio CEO. “A few had already announced that they would slow down the growth of their R&D in China, and they will likely slow down even more now,” he says. Nonetheless, the pressure that big multinationals are under to develop new drugs creates an incentive to involve researchers from all over the world, Scott notes.

Despite the incident, GSK says it remains committed to conducting R&D in China. “The work going on in China remains a key part of our R&D activities,” the company tells C&EN. “It is business as usual, and we continue to have confidence in the work under way by our scientists. What happened with this research was unacceptable, but these are very rare events.” AstraZeneca, Pfizer, and Novartis, all of which have R&D centers in Shanghai, declined to comment about the impact of Zhang’s firing on their activities.

To rehabilitate the image of its R&D center, GSK will likely soon dispatch from headquarters a new director who isn’t ethnically Asian, predicts a manager at a Shanghai-based contract research organization who spoke off the record to avoid jeopardizing his relationship with clients. “The image of the Chinese is tainted, so it will take a non-Chinese to reestablish the credibility of the facility,” says the manager who is himself a native of China.

But the incident will have nearly no effect on the growth of R&D centers operated by multinational drug companies in China, he expects. He reasons that the companies set up the labs primarily to please Chinese regulators who want to see topflight research take place in China. The sophistication of the research conducted in China will continue to increase for the same reason, he predicts.

Unless GSK publishes the findings of its investigation of Zhang’s team, ascertaining what really happened in the Shanghai labs won’t be easy. Zhang says he’s concerned that the incident could delay China’s entry into the top tier of global pharmaceutical innovators. But it’s hard to believe that events at one company—even a company leading the way—could seriously affect the momentum that is so evident in Zhangjiang.