Novartis Drug Studies in Japan–Tracing Back the Questions

Bloomberg News
A logo sits above windows at Novartis AG's headquarters in Basel, Switzerland.

The scandal over Japanese studies on Swiss drug giant Novartis AG’s blockbuster heart medicine began when one Japanese professor sent out a short letter to British medical journal Lancet, titled: “Concerns about the Jikei Heart Study.”

“This finding seems strange to me,” Yoshiki Yui, a doctor and assistant professor at Kyoto University Hospital, wrote in April 2012.

In that letter, Mr. Yui said that a 2007 article published in Lancet by a research team from Tokyo-based Jikei University School of Medicine seemed “strange,” since both the average and standard deviation for some blood-pressure measurements were identical in the control group of patients as well as the group getting the test medication. The researcher said that result was rare.

That letter prompted scrutiny of more studies on Novartis drug Diovan, and then retractions.

In December, the Japanese Circulation Society retracted two articles based on research by the Kyoto Prefectural University of Medicine, dubbed the Kyoto Heart Study, noting “a number of serious errors in data analysis.” In January, the International Journal of Cardiology retracted two articles based on the Kyoto study. In February, the European Heart Journal retracted an article on the study with a short note saying, “Critical problems existed with some of the data reported in the above paper.”

A total of six papers based on the Kyoto Heart Study have been retracted. Jikei said in July that it would ask Lancet to retract its paper.

Over the past month, special investigative panels at Kyoto Prefectural University and Jikei said their studies were based on faulty statistical analysis. Three other universities–Chiba University, Shiga University of Medical Science, and Nagoya University–have also launched investigations into the Diovan studies they carried out, but haven’t yet reached any conclusions.

What’s unclear from the separate investigations conducted so far is who’s to blame for the falsified results. Both Novartis and the universities say the same Novartis employee–whom they decline to identify–participated in all the university studies.

The Jikei study concluded that the Novartis employee had carried out the inaccurate statistical analysis, though he told those investigators that he only gave advice and other university doctors did the faulty analysis. The Kyoto panel, which was not allowed to interview the Novartis employee, said it was “likely” that the Novartis employee carried out the data analysis, and it was therefore possible that he manipulated the data. But the head of the panel told reporters it was also possible that the university’s own researchers may have manipulated the data–a possibility the head of the team has denied.

In April, Novartis commissioned its own probe by an outside panel of lawyers (in Japanese), who did interview the employee. That panel report, released July 29, said that it “did not discover any evidence of willful manipulation or falsification of data in these studies.”

Novartis says the employee left the company in May, voluntarily, after his contract expired. That’s around the same time the three other universities that had done studies announced they were investigating the Diovan studies, following a request from the Japanese Association of Medical Sciences to reexamine the data involved. The former employee couldn’t be reached for comment.

The company has not itself investigated the accuracy of the studies, saying it doesn’t have access to the data. Instead, it’s limited the investigation of the panel it commissioned to possible undisclosed conflicts of interest in the studies. The company said its Japanese unit found that an employee had a “deeper-than-expected” role in all of the studies under question, noting that his affiliation with Novartis wasn’t disclosed in any of the published studies. In June, the company issued a statement saying it had improperly permitted “an undisclosed conflict of interest” in the retracted research.

In the June statement, Novartis added it was “inappropriate” that it did not ask for a correction in the articles when the employee’s affiliation with the company was excluded.

The Japanese health ministry also launched its own probe into the case this month.

Add a Comment

We welcome thoughtful comments from readers. Please comply with our guidelines. Our blogs do not require the use of your real name.

About Japan Real Time

  • Japan Real Time is a newsy, concise guide to what works, what doesn’t and why in the one-time poster child for Asian development, as it struggles to keep pace with faster-growing neighbors while competing with Europe for Michelin-rated restaurants. Drawing on the expertise of The Wall Street Journal and Dow Jones Newswires, the site provides an inside track on business, politics and lifestyle in Japan as it comes to terms with being overtaken by China as the world’s second-biggest economy. You can contact the editors at japanrealtime@wsj.com

    • Japan Real Time on Twitter
    • Japan Real Time on Facebook