A second paper about a major randomized trial in Japanese patients with heart disease is being retracted, after an investigation reportedly found multiple problems with the paper.
As predicted by Pharma Japan, Hypertension Research is retracting a 2011 paper, already the subject of two errata. Although a spokesperson said she couldn’t say why the paper was being retracted, as the notice was still in production, editor Toshihiko Ishimitsu told us:
The paper contains honest errors which cannot be corrected accurately by existing data.
Corresponding author Issei Komuro told us more about the problem with the paper:
We have decided to retract the paper because we cannot correct one mistake. We had decided that the follow-up period is 3 years at first, but we have tried to follow up only events as long as possible because the number of events was very small. We made the figure 3, where the Kaplan Meyer curve of events was drown until 4 years, which means the follow-up period is not 3 years. Although we have kept almost all records including events, we could not find the records of follow-up period of each case, and thus we cannot calculate the precise follow-up period.
Komuro added:
The conclusions and key findings are not changed at all.
This retraction has a bit of a backstory. In 2013, an anonymous blog raised suspicions about the work of corresponding author Issei Komuro, suggesting numerous image manipulations in Komuro’s research. Last year, another one of Komuro’s VART papers was retracted, over concerns regarding conflict of interest and the reliability of the data.
Komuro has frequently collaborated with Hiroaki Matsubara, who has logged nine retractions, by our count. Matsubara resigned his post at Kyoto Prefectural University in 2013, after his work on valsartan (sold by Novartis as Diovan) was shown to be riddled with data errors and undisclosed conflicts of interest.
In 2014, Cardiobrief reported that Chiba University had investigated allegations against Komuro, and concluded the 2011 paper in Hypertension Research should be retracted:
The Chiba University investigation obtained testimony from VART investigators and found multiple problems with the paper, including the surreptitious involvement of a Novartis employee…The investigation concludes that the VART paper in Hypertension Research should be retracted.
“Effects of valsartan and amlodipine on cardiorenal protection in Japanese hypertensive patients: the Valsartan Amlodipine Randomized Trial” has been cited 21 times since it was published online in 2010, according to Thomson Reuters Web of Science.
The paper fixed a data point in one table in 2011. In 2013, it added that one of the researchers listed in the acknowledgments was affiliated with Novartis:
The authors of the manuscript would like to correct affiliation of Mr Nobuo Shirahashi, who was mentioned in the acknowledgement section of this article. The correct affiliation of Mr Shirahashi was Novartis Pharma K.K.
VART is a major trial; the newly retracted paper included 1021 patients, who each received one of two blood-pressure drugs, valsartan or amlodipine (also known as Norvasc, now available as a generic). They were followed for an average of nearly 3 1/2 years, and the study concluded:
…although BP levels were well controlled and remained equal in the two groups, valsartan had more protective effects on the heart and kidney than amlodipine in Japanese hypertensive patients.
Last year, the Journal of Human Hypertension pulled a 2011 paper over “problems with management of conflicts of interest and with the reliability of the published data.” That paper was also corrected in 2013 to add that Nobuo Shirahashi, listed in the acknowledgments, was affiliated with Novartis.
Accusations of influencing clinical trial data in Japan aren’t the only problem Novartis has faced in recent years: The company is currently involved in 22 legal cases, investigations and class action lawsuits, according to finenews.com.
We’ve followed up with Komuro to ask additional questions about the retraction, and will update if we receive any more information.
Update 10/11/16 9:31 a.m. eastern: In response to follow-up questions, Komuro told us:
…other investigations by third party organizations have concluded that there is no problem in my clinical studies (VART) and even the investigation by Chiba University did not mention that the involvement of an employee at Novartis was itself problem. All anonymous accusations that have been lodged about my research have been done by the publishers and/or university investigations. All of them are honest errors or misunderstandings by anonym.
Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.