A new set of misleading and false allegations in The Stanford Daily

In February of last year, The Stanford Daily posted a story containing serious allegations regarding a paper I published in the journal Nature in 2009, while at Genentech. The story was based on unsubstantiated claims from anonymous sources that were later found to be false – as I had asserted all along – by both Genentech and the Stanford Board Special Committee’s Scientific Panel. As noted in other media outlets, the errors in this earlier reporting have yet to be corrected.

The Daily has now posted another article on that paper that again contains inaccuracies and relies on anonymous sources without documentary evidence. Given this erroneous reporting, I wish to set the record straight.

The new article has many problems, chief among them being the implication that the 2009 paper’s findings are incorrect. In fact, the majority of the paper’s findings are correct – as I have discussed previously (e.g., here) – and they have stood the test of time. This includes the fact that a protein called APP, implicated in Alzheimer’s disease, binds to a protein called DR6, and that together they regulate neural degeneration during brain development. We later discovered that a few findings in the paper were incorrect and needed to be revised, including the specifics of how the two proteins bind, which we did in follow-on papers in 2012 and 2014. Those revised findings have also been fully validated.

Other issues in the piece include the following:

• The article implies that I was aware that some of the results presented in Nature 2009 were not solid prior to the paper’s publication. This is false. The 2009 paper proposed a specific model for how the APP protein interacts with DR6, involving the N-APP portion of APP. This was based on several lines of evidence from my lab, together with a key direct binding result from other scientists at Genentech who were not in my lab and who examined binding of N-APP made in mammalian cells to DR6 using proteins also generated outside my lab. Their binding data were included in the paper and the scientists who performed the experiments, as well as the scientist who generated the proteins used in those experiments, were acknowledged in the paper. At the time of publication, I was not aware of any binding experiments, from those scientists or others, showing results inconsistent with those we presented in the paper. I firmly believed the data were all correct and accurately presented, as discussed in the Scientific Panel’s report. Contrary to The Daily’s reporting, the findings in Genentech’s April 2023 report do not state otherwise.
• The article misrepresents my comments during a presentation to Genentech investors around the time of Nature 2009’s publication. First, I believed the information I was presenting about the findings in the 2009 paper was entirely accurate. In addition, a transcript of the presentation shows that I was appropriately measured. After introducing both the dominant theory of Alzheimer’s that involves amyloid (also known as Ab) as well as our new theory (the “N-APP DR6” theory), I stated: “[S]tepping back a little … we believe that we’re well-positioned, whatever the relative contributions of these two mechanisms, by virtue of having a best-in-class antibody on the Aß theory and by leading on the N-APP DR6 theory.” I referred to the new findings in Nature 2009 as a “theory,” and simply noted that the company was well-positioned, “whatever the relative contributions of these two mechanisms” to Alzheimer’s disease – an appropriate statement about an early-stage theory.
• The article states that it is “unclear exactly what led [me] to retract the paper entirely rather than correcting only certain parts.”  However, the article then goes on to quote from our retraction notice, which discusses in detail that the catalyst for the retraction at this time was the discovery during the past year of additional anomalies involving a few of the figures in the paper. While the new issues are minor and do not appear to affect conclusions presented in those figures, the original data are no longer available for some of them. Given that, and since our subsequent research showed that certain specific claims in the 2009 paper were not correct and we reported corrected versions of those findings in our 2012 and 2014 papers, we concluded that the scientific community was best served by retracting Nature 2009 at this time.
• The article questions the approach I took to correcting the original paper a decade ago by claiming that “Genentech confirmed” this past April that I was “urged” to retract the 2009 paper in 2012. This misrepresents what Genentech said. Genentech’s April report actually states that after my departure from the company, “one senior leader” urged that the paper should be “retracted or corrected” (emphasis added). I did, in fact, correct the paper. Precisely because most of the findings in Nature 2009 were valid, my approach to correcting the paper a decade ago was through publishing follow-on papers, an approach that was consistent with normal scientific practice and conformed to how senior scientists in my field dealt with similar situations in prior decades.

One final point – The article states that “the Stanford investigation found that [I] failed to correct the scientific record on various occasions when falsification was brought to [my] attention across three different labs and two decades”. This is false. I was not aware of any falsification in some of my papers until it was discovered in the past year (late 2022-23) as part of the Panel’s review. Prior to that there was no falsification that was brought to my attention that I could have corrected.  What had been brought to my attention in prior years were other types of issues that were not thought to involve data falsification (e.g., some errors in figure assembly, and “beautification” of the edges of some images outside the data-containing regions). I intend to address this issue at greater length in a later post.

In summary, it is disappointing to see further erroneous reporting from The Daily.

Marc Tessier-Lavigne