When peer review fails, people get hurt

This is one of a five-part series from Cosmos Weekly, investigating the peer review process.

After hanging up the phone to her mother, Rachael Lonergan phoned the police. “He’s out there somewhere,” she told them.

By the time she’d made it to her parents’ place, and before the police turned up, her father’s car was safely parked in the driveway and he was inside the house.

Her dad, David, explains Lonergan, was a very intelligent senior business executive – widely read and educated – so “to see him go from that, to a wreck, not even questioning where he was or why” was incredibly difficult for the family. They’d arranged for David’s driver’s licence to be revoked, but still had to hide the car keys to combat his tendency to pick them up as if he was going to drive somewhere.

“It was horrific at the time,” Lonergan recalls. “Mum called to tell me they’d had a big argument. He’d grabbed the keys off the sideboard and disappeared in the car, unlicenced, with advanced dementia.”

When the police arrived later, they were surprisingly calm: “Oh yeah, we get these calls all the time; this happens every day.”

This incident was one of many challenges in the final years of David’s decade-long battle with Alzheimer’s disease – although the signs were there long before the official diagnosis was made, via the detection of plaques seen on a brain scan.

“Dad, Mum and I went to a seminar for dementia, where they talked through the different types of dementia: what they mean, their impact on behaviour and what to expect,” says Lonergan. “It was very relevant and informative. And at the end of all that, Dad looked at me and said, ‘It was really interesting, Rachael, but I wish you’d told me sooner that you have dementia.’ He just completely dissociated from it and never had any self-insight.”

Alzheimer’s is a “neurodegenerative disease that results in a constellation of symptoms that we call dementia”, explains Dr Lila Landowski, lecturer and neuroscientist at the University of Tasmania (UTAS) and Director of the Australian Society for Medical Research. “When a person has dementia, nerve cells in various parts of their brain become disconnected and die, affecting how well those regions can do their job. And because the brain controls everything we do, dementia has the ability to affect every facet of our physical function.

“As the disease spreads to other parts of the brain, you lose the ability to perform basic functions, like speaking or swallowing. You might choke or aspirate your food and develop pneumonia,” says Landowski. “It’s a common cause of death in Alzheimer’s.”

And death from dementia is surprisingly common.

In 2020, according to statistics from the Australian Institute of Health and Welfare, dementia was the leading cause of death in women and the second leading cause of death in men in Australia – second only to coronary heart disease.

The Alzheimer’s puzzle

Although there are many other diseases that cause dementia (including Parkinson’s and Lewy Body diseases and vascular dementia – often caused by stroke), Alzheimer’s has long been associated with the formation of sticky plaques comprised of a waste product called amyloid beta peptide (Aβ).

“Along with finding amyloid beta in plaques in the brain tissue of deceased dementia patients, research has also found links between dementia and genetic mutations affecting amyloid beta production, so it was natural to connect the dots and assume that amyloid beta led to these plaques and to Alzheimer’s disease,” says Landowski.

However, what really stumped researchers was the lack of consistency between the number of plaques observed and the severity of dementia symptoms.

“Researchers couldn’t understand how some people had lots of plaques and no symptoms, yet others had very few plaques and absolutely flagrant symptoms,” explains Landowski.

Likewise, it was a mystery as to why drugs that completely cleared the brain of amyloid beta did not correspond to patient recovery from the disease or an amelioration of symptoms.

Then, in 2006, Nature published research led by an upcoming neuroscientist, Dr Sylvain Lesné, from the University of Minnesota in the USA, working within the lab of Professor Karen Ashe, announcing the discovery of a toxic oligomer (a special type of amyloid) known as Aβ*56 (“amyloid-beta star 56”) that appeared to directly cause dementia-like symptoms (specifically memory impairment) when purified and injected into mice.

This supports the popular, but still controversial, theory suggesting amyloid plaques were directly responsible for causing Alzheimer’s disease, and also an equally controversial new medication, Simufilam, currently in Phase Three trials in the US and Western Australia.

“Researchers couldn’t understand how some people had lots of plaques and no symptoms, yet others had very few plaques and absolutely flagrant symptoms.”

Dr Lila Landowski

This “missing link” between dementia symptoms and amyloid beta was met with great enthusiasm. It sparked hope among researchers and pharmaceutical companies alike – not to mention those experiencing the debilitating impacts of Alzheimer’s disease in themselves or, as in Rachael Lonergan’s case, their loved ones.

Funding began to flow.

“I’ve had conversations with family members around the genetic links for Alzheimer’s, and I’ve said to them, ‘Don’t worry, because by the time we’re that age, they will have cures, because they’re nearly there’,” Lonergan says, thinking back.

Unfortunately, after 20 years, millions of dollars invested in research and pharmaceutical development, evidence is emerging that all is not well with the Aβ*56 discovery.

Following the lead of Dr Matthew Schrag, a neuroscientist and physician at Vanderbilt University, Science launched a six-month investigation into the work of Lesné, a key contributor to the 2006 paper on Aβ*56, which had become a heavily-cited study in research of Alzheimer’s disease.

The key findings? Hold onto your hat…

As Charles Piller reports in Science, Schrag found “suspect images in dozens of papers” and reported his suspicions – which he dubs ‘red flags’ – to the relevant journal publishers. Several of these questionable red flags are images of Western blots (a technique used to separate and distinguish proteins in a sample) from the 2006 paper showing evidence of alleged image tampering.  At Science’s request, Shrag’s findings were reviewed by two well-regarded Alzheimer’s researchers, who agreed with the analysis, “casting doubt on hundreds of images, including more than 70 in Lesné’s papers”.

The 2006 Lesné paper now has a Nature editor’s note as a header, which states: “The editors of Nature have been alerted to concerns regarding some of the figures in this paper. Nature is investigating these concerns, and a further editorial response will follow as soon as possible. In the meantime, readers are advised to use caution when using results reported therein.”

As the situation continues to unfold, it’s becoming increasingly apparent why only a small number of other research groups claim to have found signs of the Aβ*56 oligomer, and why study after study based on the premise of this “missing link” have failed: it may not exist at all.

Aside from the alleged manipulation of images in several of the papers – investigations into which are still ongoing, and which comprise a substantial, oft-cited portion of the body of discovery work in the area – the actual method used to produce the ground-breaking results has limitations, according to Landowski.

“These little amyloid beta peptides are extremely sticky,” she says. “In the hours or days taken to produce and purify the peptides and prepare them for injection into the mouse, it’s very likely to have aggregated and turned into something else, that is no longer the Aβ peptide.”

Landowski is understandably frustrated. “We have a funding crisis in Australia at the moment. People obviously want to fund big ideas, backing those that are the most well-known and have the most evidence behind them.” Funnelling money towards fabricated research could “ultimately mean that other worthy ideas – both those to do with Alzheimer’s and others not related to the disease – miss out.”

And, particularly for those participating in clinical trials for medicines based on this research (including simufilam) the cost is not just financial.

“These drugs aren’t without side effects,” says Landowski. “The antibody immunotherapies that clear out amyloid beta can cause fluid accumulation in the brain, strokes, death. The risk is low, but people have died from trialling these therapies.”

Isn’t this what peer review is meant for?

Arguably, the fact that Aβ*56 may not exist should have been caught years ago at the peer review stage. After all, isn’t this how the scientific method is supposed to work: hypothesis, test, analyse, conclude, review and then publish?

If someone is reporting something that isn’t there, using potentially manipulated images and a research process that is scarcely able to be repeated by other experts in the field, then surely it should be picked up before it makes it to publication in prestigious journals like Nature?

To pick up the discrepancies in the 2006 paper, says Landowski, a reviewer would need days or weeks. There simply isn’t time for this in the current research ecosystem, and especially not for free (a normal expectation of academics performing a peer review). “It’s beyond the scope of the reviewer’s expertise and well beyond what you’re actually expected to do: assess the science based on its claims, its veracity and the quality of the data.” she says.

“You assume that the person publishing isn’t fabricating the data.”

Unfortunately, it just isn’t that simple. Disentangling the problems of peer review from the scientific method is only the first step to understanding how the system can fail.

Amyloid beta plaque has underpinned vast swathes of research over the past four decades, so it’s little surprise the scandal has attracted attention globally. Amyloid beta and drugs targeting brain plaques as treatments for Alzheimer’s disease are still being developed and tested.

For Rachael Lonergan, it’s much more personal.

“Due to the family’s history, one of my family members was tested for dementia,” she says. “They had a very expensive brain scan to look for plaques, which wasn’t at all reimbursable, but we thought it was worthwhile. We were relieved to find out there were no plaques, so we thought she was safe – she didn’t have [dementia]. But of course, that’s all in question now.”

“How did nobody, through that period of time, go: ‘hang on a minute, this isn’t making sense’?”

Rachael Lonergan

Now in her early 50s, Lonergan is within two decades of the age her father started showing symptoms of dementia. After watching her father’s steady decline over the 10 years up to his death in 2019, she’s anxious.

“And I’m thinking, I mean, this started in his late 60s. So it’s not that far away, so that concerns me.”

And she’s angry.

“I’ve been saying they will have cures because they’re nearly there. And now we now know that that’s not true. The idea that it’s gone on so long, that it wasn’t detected, is unbelievable to me.”

Global media coverage of this case is keeping the issue from being swept under the rug. Retractions relating to the research will be publicised and well-known – which is essential in ensuring the problems aren’t forgotten or accidentally perpetuated by unwitting researchers.

Despite the latest Alzheimer’s allegations, Lonergan, a triple-negative breast cancer survivor and creator of the CanDo app (which aims to support and coordinate people going through cancer treatment and their supporters), remains fully in favour of the scientific method.

“I feel angry because, through my journey and through what I do with the app, I have met incredible scientists and incredible researchers that I would absolutely trust and support,” she says. “As a layperson who operates in the periphery of this world, I’d like to believe researchers have integrity and are held accountable. The idea that there are people in that world that are causing that world to lose this integrity makes me really angry.”

Despite being a self-professed ‘layperson’ rather than a specialist academic, Lonergan – who has two vastly different experiences of the real-life implications of the academic process – asks the questions on many people’s lips: “Obviously, this raises serious questions about peer review and the scrutiny of data, and how many studies were done to get to this point where they were making these assertions about these plaques?”

“And how did nobody, through that period of time, go: ‘hang on a minute, this isn’t making sense’?”

---

This story is part of a five-part series from Cosmos Weekly on peer review. Read the other four:

• Is it time to review, the review?
• Putting science under the microscope reveals not all is well
• Peer review: where does it lie in the scientific process?
• Scientific fraud, poor research and honest mistakes leads to thousands of retractions

Next week: Clare Kenyon interviews Elisabeth Bik and Ivan Oransky on Monday. Subscribe to our free daily newsletter to be the first to see it.