Novartis Diovan scandal claims two more papers

A complicated story involving Novartis’s valsartan (Diovan) has led to the retraction of two more papers, one cascading from the other.

Last September, The Lancet retracted the Jikei Heart Study after a slew of retractions of related work prompted an investigation of valsartan research. That investigation found evidence of data manipulation and the failure of one researcher to note his Novartis affiliation. The company has apologized.

Here’s one retraction, from Diabetes Care, for “The Shiga Microalbuminuria Reduction Trial (SMART) Group. Reduction of Microalbuminuria in Patients With Type 2 Diabetes: The Shiga Microalbuminuria Reduction Trial (SMART):”

Based on the review and recommendation of the American Diabetes Association’s Subcommittee on Ethical Scientific Publications (ESP), Diabetes Care formally retracts the above-cited article. For reasons described below, the published data are considered to be unreliable.

According to an investigation and analysis of the data conducted by Shiga University of Medical Science (report received 25 November 2013), 10.1% of the albumin-to-creatinine ratio (ACR) data reported in the article did not match original patient records. In the valsartan and amlodipine groups, respectively, 12.5% and 8.1% of the ACR data reported in the article differed from patient records. Close examination of these differences showed that ACR values reported for the valsartan group were smaller, and those for the amlodipine group were larger, than values reported in patient records; these differences “worked to the advantage of valsartan.” The investigative panel of Shiga University of Medical Science concluded that the above-cited article is “inappropriate for [a] scientific paper.”

It should be noted that an erratum to this article recently appeared in Diabetes Care (2013;36:4172. DOI: 10.2337/dc13-er12). As confirmed by the corresponding author, the affiliations for two of the listed SMART investigators were incorrectly listed as Shiga University of Medical Science and Osaka City University. The correct affiliation for both investigators, as stated in the erratum, was Novartis Pharma K.K. The erratum was published in the December 2013 issue of Diabetes Care, which posted online 21 November 2013.

That paper, whose results were mentioned in last fall’s stories about the investigation, has been cited 33 times, according to Thomson Scientific’s Web of Knowledge.

And here’s the notice for “Impact of Renin-Angiotensin System Inhibition on Microalbuminuria in Type 2 Diabetes: A Post Hoc Analysis of the Shiga Microalbuminuria Reduction Trial (SMART)” from Hypertension Research, which refers to the Diabetes Care retraction:

The authors have indicated to the journal that this paper should be withdrawn in response to the interim report on the Shiga Microalbuminuria Reduction Trial (SMART), and in consequence of the retraction of the main report from the same trial. After careful consideration, Hypertension Research editorial committee formally retracts this paper with agreement of the authors.

The Hypertension Research paper has been cited eight times.