Group Calls for Retraction of Paper on the Risks Associated with Testosterone Therapy

An ad-hoc group of physicians and researchers is calling for the retraction of a much-cited study that helped spur an ongoing federal investigation into the possible risks of testosterone therapy.

The study, which was initially published by the Journal of the American Medical Association in November and corrected twice thereafter, examined records of men with low testosterone levels (<300 ng/dL) who underwent coronary angiography in the Veterans Affairs system between 2005 and 2011.

Of the 7,486 patients who received no testosterone therapy, 681 died, 420 suffered myocardial infarction (MI), and 486 had strokes, according to data at JAMA’s website, which has been updated with all corrections. Of the 1,223 patients who received testosterone therapy, 67 died, 23 had MIs, and 33 had strokes.

“At 3 years after coronary angiography, the Kaplan-Meier estimated cumulative percentages with events were 19.9% in the no testosterone therapy group compared to 25.7% in the testosterone therapy group, with an absolute risk difference of 5.8% (95% CI, −1.4% to 13.1%),” the study read.

Even before the first of the corrections, the study’s methodology attracted some comment.

For example, the men who were prescribed testosterone were healthier than those who were not, wrote Anne R. Cappola, MD, ScM, in an accompanying JAMA editorial.

However, Cappola continued, “the men receiving testosterone had a higher risk of the composite adverse outcome, which runs counter to what would be anticipated if there was residual confounding from unmeasured advantageous health variables.”

Other concerns revolved around numbers, and several of the originally published numbers turned out to be wrong. The first correction to the article came out in January, which added fuel to the fire. The second correction came out in March, which added more.

In a letter to JAMA’s editor dated March 25, a group of more than a dozen researchers headed by Abraham Morgentaler, an associate clinical professor of urology at Harvard Medical School, called for a full retraction.

“In response to the criticism that the authors had improperly excluded 1,132 men from analysis who had received a testosterone prescription after experiencing an event (MI or stroke), the authors indicated they had revisited this value presented in the text and Figure 1, and had made ‘an incorrect notation’ regarding this value. They now assert the numbers of men excluded for this reason was 128, not 1,132. This is an 89% error rate, involving >1,000 individuals,” the letter authors wrote.

Morgantaler et al. also noted, “The number of men now excluded due to missing coronary anatomy increased from 397 to 1,301, a 69% error rate, involving 904 individuals. Astonishingly, 100 women were now identified among the original group of 1,132 individuals, meaning that one out of 11 ‘men’ in the study were actually women.”

The signers of the letter, some of whom have financial ties to testosterone therapy makers, assert that such errors undermine any possible credibility the article may have and they request a retraction.

In an article on the Wall Street Journal website, Morgantaler (who receives funding from Abbvie, which markets AndroGel) said, “When you look at the big picture, what matters is they can’t get the numbers right. And so there’s nothing that’s believable in this paper… They looked at one part of the data and what they reported is so different. It’s gross data mismanagement. When the data are so off, it’s impossible to believe anything that was originally published in the paper. And now we have ads from lawyers on television looking for men who had heart attacks or strokes while on testosterone. The article that set the stage for this concern and prompted the FDA to issue a safety alert [that is] based on data that is no longer reliable.”

JAMA and the authors of the study, on the other hand, stand by the findings and note that even after changes to the data, testosterone use was associated with more adverse events.

In the Wall Street Journal article, co-author Michael Ho, MD, said “In the original publication, there was misclassification of the reasons why patients were excluded. This was corrected in [our] response letter that was provided to and published in JAMA. But this did not change the results of the paper. Changing the reasons for study exclusion does not alter the study results. Misperception about that could suppress research that raises important questions about patient safety.”