A now-seminal study published in 2006 provided evidence that the toxic accumulation of a protein called beta-amyloid in the brain was tied to the development of Alzheimer’s disease. Recently, an assistant professor from Vanderbilt University suggested that some of the images in this study were manipulated by the authors. What does all of this mean?
In 2006, a group of researchers from the University of Minnesota published a dementia-related study titled in the journal NatureTrusted Source called “A specific amyloid-β [beta-amyloid] protein assembly in the brain impairs memory.” The study provides evidence supporting a specific protein clump in the brain, known as beta-amyloidTrusted Source, as a cause for Alzheimer’s disease.
The study used a mouse model to show how these protein clumps — also known as amyloid plaquesTrusted Source — could cause dementia.
Because of its findings, this study became very influential in Alzheimer’s disease research. To date, it has been cited in over 2,200 scientific papersTrusted Source and accessed more than 34,000 times.
Now an article recently published in Science reports that an assistant professor of neurology at Vanderbilt University believes some of the images were manipulated in the 2006 Nature study, calling into question the validity of its findings.
What is image manipulation in peer-reviewed articles?
Image manipulationTrusted Source is the process of changing a photograph. According to Dr. Elisabeth Bik, microbiome and science integrity consultant at Harbers-Bik LLC, photographic images can be easily digitally altered, for example, when we remove wrinkles or a mole from someone’s face in a photographic portrait.
“In scientific photography, you are not supposed to alter the image beyond some light general contrast adjustments applied to the whole image,” Dr. Bik explained for Medical News Today.
“Most journals nowadays explicitly forbid making any digital alterations. But if a researcher does an experiment, and the results are not as clean or if the results are completely different from what they expected, it is tempting and easy to digitally remove a stain or scratch in the background, or to add or remove some cells or change the thickness of a protein band. It is much faster to do some photoshopping than to redo the experiment.”
– Dr. Elisabeth Bik
This is certainly not the first time the images in a study have come into question. A study from 2016 — on which Dr. Bik was a co-author — found that 3.8% of scientific papers published in 40 journals between 1995 to 2014 had potentially problematic images, with at least half suggestive of deliberate manipulation.
To help fight the issue of image manipulation in scientific papers, in 2021, eight journal publishers outlined a three-tier approachTrusted Source to help editors flag potentially problematic imagery.
What might this mean for dementia research?
MNT reached out to both Dr. Matthew Schrag, assistant professor of neurology and director of the Cerebral Amyloid Angiopathy Clinic at Vanderbilt University, who has made the allegations against the 2006 Nature study, and to the lead author of the study, Dr. Sylvain Lesné, associate professor in the department of neuroscience at the University of Minnesota. Neither of them responded to our questions.
A University of Minnesota public relations representative stated that the university is aware that questions have arisen regarding certain images used in peer-reviewed research publications authored by members of the University faculty, and that they were following due processes to review the questions any claims have raised.
Because of the influence that the 2006 Nature study has had on Alzheimer’s disease research, Dr. Bik said that if more research proves image manipulation, it would come as a blow to certain lines of enquiry.
“The 2006 Nature paper by Lesné et al. has been influential, and has led lots of researchers to pursue the same hypothesis and to replicate the study,” she pointed out.
“The AB*56 [beta-amyloid] work has also not yet directly led to any clinical trials. But it has encouraged several other lines of research that pursue slightly different angles, which have been tested in clinical trials. Yet, no experimental drug has been proven to be effective against Alzheimer’s,” added Dr. Bik.
“It is fair to say that the 2006 Nature study has led to a lot of wasted research money and effort, and raised a lot of false hope in patients,” Dr. Bik remarked. “There are other, alternative hypotheses to the beta-amyloid story, and perhaps there now will be more money to test those alternative ideas.”
Dr. Grace Stutzmann, professor and discipline chair of neuroscience and director of the Center for Neurodegenerative Disease and Therapeutics at Rosaline Franklin University of Medicine and Science, however, told MNT that even if the alleged image alterations in the 2006 Nature study were intentional, she did not think this would undermine all the research conducted in the field thus far.
“This case involves a specific single arrangement of beta-amyloid from a single lab that is in question, and there are many other amyloid variants that have been studied across multiple labs that were replicated,” she explained. “The [Alzheimer’s disease] field, in general, is much larger than just amyloid, so in reality, it’s a proverbial needle in a haystack.”
According to Dr. Sara Imarisio, head of research at Alzheimer’s Research UK, if these allegations of image manipulation are true, then, in the wake of the study, research groups may have planned experiments based on an erroneous premise, diverting valuable researcher time that could have been better spent elsewhere.
“But the findings from the paper were very specific and, contrary to some reports, have not significantly affected the progress or direction of research into Alzheimer’s,” she added. “Even for research groups who work in this particular area, findings that can’t be reproduced will be identified as controversial and lose credibility, while genuine findings will come to predominate and guide the direction of future studies.”
Dr. Maria C. Carrillo, chief science officer at Alzheimer’s Association, states that, as we continue to move forward, it is important to note that this investigation is related only to a small segment of Alzheimer’s and dementia research, and does not reflect the full body or science in the field.
“As such, this should not influence the field’s accelerating pursuit of the initial causes and other contributors to Alzheimer’s disease and other dementias,” she adds.
The Alzheimer Society of Canada also released an official statement regarding this matter, saying the allegations are a “serious concern and require deeper inquiry. Scientific integrity is crucial, and any potential diversion of money or time is a cause for concern.”
What can journals do to prevent future misconduct?
Dr. Charles Glabe, professor of molecular biology and biochemistry at the University of California Davis, stated that science depends on trust and the understanding that fabricators will ultimately be caught.
“Image duplication and copying were caught by software tools that compare bands on a gel pixel by pixel,” he noted. “This is good and well, but now that fabricators know that copying bands is easily caught, they will just run a different gel and use that one instead of publishing the same band twice.”
And Dr. John Hardy, professor in the Department of Neurodegenerative Diseases & Reta Lila Weston Laboratories at the UCL Queen Square Institute of Neurology, told MNT that it is very difficult to prevent fraud.
“One thing which has changed and was important in this case was image recognition software, which can catch stuff that previously people had got away with,” he said. “This has meant quite a lot of ‘old fraud’ has now been caught — like DNA testing of crime scenes.”
Moving forward, Dr. Bik said that to help scientific journals check for potential image manipulation there should be increased scrutiny by scientific publishers.
“Scientific publishers should spend some money and effort on quality control of submitted articles,” she advised. “They make a lot of profit but seem to not do enough screening manuscripts for signs of concern or fraud.”
“They should hire experts in statistics, ethics, and image forensics to screen such papers, and not rely on unpaid peer reviewers, who might not know how to look for misconduct.”
“Journals and institutions should also penalize researchers proven to have committed misconduct, and be much quicker to retract papers,” Dr. Bik added. “Some of these concerns about the Lesné papers were already raised years ago. Journals and institutions are too slow and too hesitant to tackle these problems, and that will have to change.”