Japan Clinical Research Scandal Prompts Calls for Stricter Rules

By Kana Inagaki

Japan needs to bring its guidelines on clinical research more into line with international standards. That’s what some medical experts are saying, in the wake of findings by two Japanese universities, that data in studies their researchers conducted on Novartis AG’s blockbuster heart medicine had been altered to produce inaccurate results.

Novartis denies involvement in the alleged research distortions, and stands by the efficacy of its product, a popular blood-pressure lowering drug called Diovan.

But Iwao Kuwajima, chair of the Japanese Organization of Clinical Research Evaluation and Review, said the findings have damaged the reputation of Japanese research, and stronger guidelines on how to conduct research properly are needed to restore credibility. Starting late last year, multiple Diovan-linked studies have been retracted from prominent journals such as the European Heart Journal.

“The rule-making is critical. The problem in this case is the significant loss of global confidence,” Mr. Kuwajima said.

At the time the Japanese studies on Diovan started in 2001, for instance, Novartis said there were no guidelines in Japan that clearly regulated conflict of interest in clinical research; those guidelines were later set up by the education ministry in 2006.

Novartis has said one of its employees took part in all five studies, but was identified in published articles only as a lecturer at Osaka City University, where he had a part-time position. The company says it had a “conflict of interest” in the studies, which it should have disclosed but “inappropriately” did not.

The company did not say whether it felt the undisclosed conflicts played any role in affecting the tests.

Experts say the tests may have been vulnerable to tampering because researchers didn’t apply precautions that are standard for similar research in the U.S. and Europe.

Studies by the two universities that found altered data were so-called “open” trials, where both the researchers and participants knew which drug was being given. In the U.S., regulators encourage researchers to conduct “double-blind” tests– where no one in the study knows which participants will receive the experimental drug–to avoid bias in research results. In Japan, many medical researchers conduct open trials, according to Mr. Kuwajima.

In the U.S., statistical analysis of clinical trial data is usually done by a third party separate from statisticians hired by the sponsor of the research. Europe also has rigid guidelines on clinical trials that were introduced in 2001. Japan’s guidelines on statistical analysis don’t call for independent analysis of the data.

Also contributing to the difficulty, doctors say, is a shortage of experts in Japan, especially in the area of statistical analysis–a critical final phase of determining the findings in clinical studies. In both probes by the two universities, the research teams claimed the statistical analysis was carried out by the Novartis employee, who has now left the company.

“Data checks in Japan are behind, compared with the U.S.,” said Tetsuya Tanimoto, a visiting researcher at the University of Tokyo’s Institute of Medical Science.

In a probe commissioned by the Tokyo-based Jikei University School of Medicine, doctors involved in the Diovan study told investigators that they “had no knowledge or capability for statistical analysis.” The Jikei panel concluded that the Novartis employee had carried out the statistical analysis, though he told those investigators that he only gave advice and other university doctors did the allegedly faulty analysis. A separate Novartis panel of lawyers said it did not find any evidence of “willful manipulation or falsification of data” based on interviews with the former employee.

The controversy could lead to some changes in research rules in Japan. Japan’s health minister recently said a 12-member panel probing the studies — which includes Mr. Kuwajima–will consider revising ethical guidelines on clinical research. Separately, the Japanese government also plans to require universities and hospitals to retain their research data over the longer term so figures can be traced back if research results are later in doubt. Now, there are no guidelines on how long Japanese medical researchers must keep data.