The cause of Alzheimer’s disease has troubled the science world’s best detectives. Michael Regnier asks: can such a mystery really be solved if we gather enough clues?
The King of Crete demanded that, to atone for the death of his son, the city of Athens send him seven young men and seven young women every seventh year. When they arrived in Crete, they were sent into the Labyrinth, at the centre of which lived the Minotaur, a monstrous creature with the body of a man but the head of a bull. If the Minotaur didn’t kill the young Athenians, they were condemned to wander the tortuous maze, losing first their way and eventually their lives.
This human tribute had been paid twice when Theseus, a prince of Athens, took his place among the 14 youths. In most retellings of the myth, it is taken for granted that the hero will defeat the Minotaur in combat. But the physical fight is not the point of this story – it’s how Theseus overcame the mental challenge of navigating the Labyrinth. As well as his sword, he took with him a ball of thread, a clew, to line his path. No matter how deep into the maze he had to go, the thread would bring him out safely.
Theseus’s clew is the root of the modern English word clue, and clues are the best way to solve a mystery. They are signposts showing that a puzzle, no matter how convoluted it seems, has a straightforward solution. Real life is messier than myth or fiction, but we still tend to believe that our problems can be reduced to something manageable, if we only gather enough clues.
© Tom Gauld/Heart
“Auguste D–” was the name on the blue cardboard case file. Auguste Deter, admitted to the Hospital for the Mentally Ill and Epileptics in Frankfurt on 25 November 1901. Senior physician Dr Alois Alzheimer examined her the next day and over three days after that.
“What is your name?”
“What is your husband’s name?”
“Auguste, I think.”
“Ah, my husband.”
Aged 51, Deter had an unusually severe and progressive dementia. Her first symptom had been intense jealousy towards her husband. Soon her memory had begun to fail, she was often disoriented and would hide things in her flat; sometimes she felt someone wanted to kill her and she would start screaming. She died within five years.
By then, Alzheimer had taken a job in Munich but, unable to forget Deter, he arranged for her brain to be sent to him after her death. He wanted to get inside her head, literally, and see what 32 pages of medical notes had failed to show: the cause of her mental degeneration. Alzheimer cut thin slices from her brain and examined them under the microscope, looking for clues.
Your brain is a dark, damp maze. The 1.5 kg or so of soft brain tissue that is folded and pleated in your skull holds within it another, more complex labyrinth of nerve cells reaching out to each other in the darkness, making electrical and chemical connections, forming pathways and circuits that somehow give rise to consciousness and cognition, the memories and thoughts that define you. In Deter’s brain, that labyrinth had become a trap. Connections were lost, cells missing, memories wiped, intellect destroyed. As its order and organisation broke down, her brain made desperate attempts to reroute signals, wandering the same loop of thought over and over or generating bizarre, frightening delusions. The interweaving neural paths that once allowed her to interact with the world eventually became impassable, overgrown with knotty plaques and tangles – these were the clues Alzheimer found.
In Deter’s otherwise healthy nerve cells, many of the fibres along their branch-like extensions were unusually thick and impregnable – the ‘tangles’. At the same time, lying between her cells were small, round clumps, or ‘plaques’. It seemed obvious that these plaques and tangles – the twin hallmarks of what is now called Alzheimer’s disease – made the difference between a healthy brain and a diseased one.
© Tom Gauld/Heart
Detective fiction was invented in 1841 by Edgar Allan Poe. Inspired by the burgeoning sciences of the early 19th century, the American writer created Monsieur C Auguste Dupin, a Parisian sleuth. Dupin was supremely intelligent, observant and rational. He was also reclusive and boastful, but capable of astounding leaps of deduction that cut through the most mystifying crimes. Literary commentators of the day said Poe had moved the focus of storytelling from the heart to the head: romance and drama were sacrificed for logic and reasoning; thoughts counted more than actions.
Dupin’s first case was The Murders in the Rue Morgue: two women had been brutally murdered in a fourth-storey room from which there was apparently no possible exit – the windows were fastened from the inside, the chimneys too narrow for even a cat to climb, and a crowd of neighbours had blocked the stairs to the locked door, the key on the inside, as they listened to the horrific shrieking within. One woman’s head was almost completely severed from her body; the other’s body was crammed head-down up a chimney. Insoluble to all of Paris, not least the police, the mystery becomes transparent to Dupin thanks to the clews (Poe uses this spelling throughout) he finds at the scene of the crime. The killer is indeed an inhuman brute – it turns out to be an escaped orang-utan.
As the fictional construct of an author solving constructed crimes, Monsieur Dupin was infallible. Real life would prove more challenging.
“Alzheimer’s disease is a construct; all diseases are constructs,” says Simon Lovestone, Professor of Old Age Psychiatry at the Institute of Psychiatry, King’s College London. “They’re our best attempt to explain the world.”
Alois Alzheimer’s descriptions of Deter added a new construct to the medical textbooks: a rare form of dementia that attacked memory, self-awareness and rational intellect, was accompanied by plaques and tangles in the brain, and developed much earlier than ‘senile dementia’, the gradual loss of mental faculties that was all but accepted as a natural part of old age.
This definition came under threat in the 1930s, as it was discovered that some apparently healthy old people also had plaques and tangles in their brains when they died. If Alzheimer’s clues were red herrings, not specific to the dementia at all, it would leave nothing else on which to start building an understanding of the disease.
From the 1960s, scientists took a more rigorous approach. Rather than noting merely the presence or absence of plaques and tangles in people’s brains, they began to measure how many there were and how these quantities related to performance in standardised tests of cognitive ability. In general, more plaques and tangles were present in cases of more severe dementia, suggesting they were indeed linked. Where people had plaques and tangles but no dementia, it may have been that they died before symptoms appeared, or that our brains could compensate up until a certain point. Overall, researchers were now satisfied that Alzheimer’s disease was real.
In fact, Alzheimer’s disease was redefined in the 1970s to include anyone with dementia who was then found to have plaques and tangles in their brain when they died. This turns out to include many people who would previously have been diagnosed with ‘senile dementia’ as well as those who fit Alzheimer’s original description. This new construct became by far the most common form of dementia – according to the charity Alzheimer’s Research UK, about half a million people in the UK have Alzheimer’s disease, accounting for more than 60 per cent of all people with dementia.
“To me as a scientist, Alzheimer’s disease is a process in the brain which is manifested by plaques, tangles and neuronal loss,” says Lovestone. “It is defined by progressive loss of cognition and an ability to function in the daily world.
“People lose their memories and then they lose their ability to do complex tasks and then increasingly simple tasks, and this results in them needing care. And it is frequently accompanied by behavioural and psychological symptoms, which can be everything from sleep disturbance to frank psychotic features.
“To me as a clinician, Alzheimer’s disease is a patient before me with a set of characteristic symptoms, who I’m currently unable to help very much.”
The body of Mary Rogers had been found floating in the Hudson river, off the shore of New Jersey, in July 1841. A tobacconist’s clerk, she was referred to in the press as the “beautiful cigar girl”. The ensuing investigation gave the American journalists ample opportunity to come up with a slew of sensational theories. Some papers claimed Rogers’s former employer was the culprit, some that she was the victim of a gang rape and murder; others had more far-fetched ideas.
Edgar Allan Poe thought he knew better. The following year, he translated the salient details of the case to Paris so that Monsieur Dupin could tackle the analogous Mystery of Marie Rogêt, serialised in Snowden’s Ladies’ Companion. Before he could publish the final instalment and reveal the culprit, however, real life intervened: a woman confessed on her deathbed that Rogers had not been murdered but had actually died during a botched abortion on her premises.
No one was ever convicted in the Rogers case, largely because no one could ever be sure what had happened. In Poe’s story, Dupin never commits to a solution. Poe left his ending inconclusive, instead discussing the nature of probability and error when choosing between competing theories.
The 1970s redefinition of Alzheimer’s disease attracted more attention, more money and more researchers. Everyone had their own pet theory. John Hardy, a scientist who has come as close as anyone to finding a convincing explanation of Alzheimer’s disease, says it was the land of anything goes: “People said it was aluminium, ‘slow viruses’, oxidative damage – not that any of these are necessarily 100 per cent wrong but it was just an anarchic ideas mush.”
Out of the mush, however, came a simple theory that led to the first set of treatments for Alzheimer’s disease. Scientists had discovered that we need a neurotransmitter called acetylcholine to form memories. Since memory is the first thing to go in Alzheimer’s disease, and nerve cells that used a lot of acetylcholine seemed especially vulnerable, researchers were convinced this could be the basis of a treatment. They designed drugs such as Aricept to bolster acetylcholine and help maintain the function of those specific nerve cells.
It would have been a relatively simple solution to Alzheimer’s disease – if it had worked. Unfortunately, the benefit to patients is short-lived before the disease regains the upper hand and mental degeneration continues. And yet there is still no better treatment available today than these imperfect drugs.
© Tom Gauld/Heart
Poe returned to the elegant simplicity of a constructed crime for Dupin’s third and final case. In it, the detective must locate The Purloined Letter, which has been brazenly stolen from the Queen of France. The thief, a politician referred to only as Minister D–, turns out to be Dupin’s mirror image, his arch-rival: both detective and crook renowned for their genius and distinguished in mathematics and poetry. These similarities are deliberate, Poe’s point being that a detective has to be able to understand and think like his quarry to catch him.
By the time Dupin is asked to help, the police are on the Minister’s trail, surreptitiously searching his home nearly every night for three months. They have used microscopes to examine the walls, floors, tables and even the rungs of the chairs for signs that the letter had been secreted within. Only Dupin understands that Minister D– would have anticipated such thoroughness and avoided any attempt to conceal the letter: therefore, unlikely as it seems, he must have not hidden it at all.
Donning a pair of green spectacles as a disguise, Dupin calls on the Minister. He spots a soiled and crumpled letter lying carelessly in a cheap plasterboard card-rack. Despite not matching the police’s description, this, he deduces, is obviously the stolen object. He is, of course, proved right when he later steals back the letter, which had simply been folded over on itself.
© Tom Gauld/Heart
Plaques and tangles had not been forgotten amid the attempts to explain Alzheimer’s disease. Scientists scrutinised their structures with ever more powerful microscopes, and then identified what they were made of. Tangles consist of tau, a protein that usually helps to maintain the structure of nerve cells. In Alzheimer’s disease, tau moves to the wrong part of the cell, folds over on itself and forms helical filaments, which then build up as tangles. Another protein that can fold back on itself is amyloid beta. It is made from a longer protein called amyloid precursor protein, or APP, but no one knows whether it has a proper job in the body. In Alzheimer’s disease, amyloid beta misfolds and forms long fibres that accumulate and become plaques.
Having gone hand-in-hand for decades, plaques and tangles were divided now it was known they were made of different proteins. Looking for a simple explanation of the disease, scientists required one or the other to be the cause – for two proteins to be equally involved would be improbably complicated. Inevitably, people took sides: it was amyloid versus tau.
The amyloid camp seized the initiative. John Hardy and colleagues made a breakthrough in the genetics of Alzheimer’s disease and developed the ‘amyloid cascade hypothesis’. This quickly became the dominant explanation of the disease and has directed most research efforts ever since. It is so well established it even took centre stage in the plot of a primetime British TV detective drama in 2013.
Hardy laughs when he remembers that episode of Lewis, in which a murderous Alzheimer’s researcher is apprehended by the eponymous detective. “He found a body in the attic of an Oxford college. It turned out to be the body of a PhD student who had been murdered by their very glamorous woman supervisor. The student had come up with the amyloid hypothesis and the supervisor had stolen it from her.”
That’s not quite how it happened in real life. In the late 1980s, Hardy was racing many other scientists around the world to find an ‘Alzheimer’s gene’. They were all studying families with lots of cases of aggressive dementia, the kind that takes hold when someone is in their 40s or 50s rather than over 65. Because the disease runs in these families, there was likely to be a consistent genetic factor that caused Alzheimer’s disease in those who inherited it in each generation.
“It’s much easier to go down a river than to go up,” Hardy explains. “Instead of looking at the brain and trying to work out what had happened backwards, one could find the gene and work out what had happened going forwards.”
The race was hard-fought, with false turns and retraced steps. Eventually, in 1991, Hardy’s team won, identifying the first genetic mutations to be associated with Alzheimer’s disease – and they were all in the amyloid gene. Soon, other research groups found different mutations in the amyloid gene that were associated with the disease, as well as in two other genes that help to control amyloid beta. It was compelling evidence that a fault in amyloid beta kick-starts Alzheimer’s disease.
In the amyloid cascade hypothesis, which Hardy and colleagues formalised the next year, everything followed in sequence: the formation of plaques, the tangles, nerve cells failing and dying, and progressive dementia the end result. Its linear order made it an appealing explanation.
“The good thing about the amyloid hypothesis is that it forced people to think about the disease in a simple way,” says Hardy. “Suddenly everyone had a framework on which to pin their experiments.
“It might have been wrong – well not wrong, but not completely correct – but at least there was a framework.”
The framework was embraced by the pharmaceutical industry, who began developing drugs to head off the cascade by blocking the production or accumulation of amyloid beta. It would have been a relatively simple solution to Alzheimer’s disease – if it had worked. In the last 15 years or so, drug after drug designed to work on amyloid beta has failed to provide enough benefit to enough patients to be considered effective. Of the dozens of drugs that have been tested in that time, some have been withdrawn because of detrimental side-effects; others just did not show enough of an effect slowing patients’ mental decline or helping them get by in everyday life.
The failure of these drugs does not necessarily mean the amyloid hypothesis is wrong – Hardy and other scientists point to a number of flaws in the way many of these drugs were tested. On the other hand, it may mean the hypothesis needs tweaking: perhaps amyloid beta starts the cascade, which is then sustained by something else. But the drug trials have not given a clear answer, to Hardy’s obvious frustration: “The summary of all those trials is we really don’t know. It’s like reading tea leaves.”
However, one of the most recent drugs to be tested, solanezumab, has shown a hint of an effect in people with mild Alzheimer’s disease – enough to be re-tested just in people in the early stages of the illness. If the effect is confirmed, this drug could potentially be used to treat people when symptoms first appear, or even as a preventive treatment for people with evidence of plaques but no symptoms. Anything positive to come out of these ‘failed’ trials would be something to work with: “If any of these drugs starts to work, even if it’s a small effect, then we will crack Alzheimer’s,” says Hardy.
“I used to try and fix my motorbike and if you can’t get the bike to start at all, then you’re fucked. But once you get it to kick over a few times and it just needs tuning, that’s a lot easier.”
While no one disputed the genetic discoveries in Alzheimer’s disease, the amyloid hypothesis did not heal the divisions between the amyloid and tau camps – if anything, it made them wider and more acrimonious.
Michel Goedert was in the team that identified tau as the major constituent of tangles in 1988; he has focused on tau ever since. When amyloid beta turned out to be the common link between the genetic mutations, he recognised its significance as the first step on the path to disease, but he did not like some of the interpretations of tau’s role that followed.
“It led many people to say that the tangles were something of no consequence. Worse than a by-product: an innocuous by-product,” he says. “I didn’t buy it but the pressure was so strong, you couldn’t ignore it. So there was this period of seven years and there was – well, not just these battles, but it seemed to be clear who had won the battle.”
Tau research became deeply unfashionable. Those who persevered in the 1990s say it was extremely difficult to get funding or to get their results published in the more prestigious scientific journals.
Vindication came in 1998, when a group of scientists – including Hardy, who says he has now published far more on tau than amyloid beta – discovered that a rare form of dementia was caused by mutations in the tau gene. This meant that misfolded tau was enough to damage nerve cells and it was unlikely, therefore, to be just an innocent bystander in Alzheimer’s disease.
The spread of tangles in the brain is much more closely related than plaques to the progress of dementia in Alzheimer’s disease. When it was discovered that tau could cause damage, the amyloid hypothesis was updated to accommodate the idea that it might be tau, not amyloid beta, that actually kills nerve cells and causes dementia. “You could call the APP [amyloid] dysfunction the initiator of the whole thing but the tau dysfunction is the executioner,” says Goedert.
There are some who go further, however, saying that tau is the real driving force behind Alzheimer’s disease. According to them, the reason the tau gene has not been linked to the disease is that tangles are a natural phenomenon of ageing in the brain. For most of us, this only becomes a problem in old age when the tangles have blocked lots of pathways, similar to the way our muscles become weak in old age. In Alzheimer’s disease, some other factor – maybe faulty or excessive amyloid beta – causes problems to occur earlier and progress faster.
Either way, people with Alzheimer’s disease might well benefit from treatments that stop tau. There are such drugs in the pipeline, but the tau field lags behind amyloid.
Only one tau drug has reached the final stages of testing and it has attracted controversy, not least because of the man responsible for its development.
In 1907, the same year that Alzheimer published his first report of the case of Auguste Deter, a French novelist called Gaston Leroux published The Mystery of the Yellow Room. It owed much to Poe’s detective stories, telling the tale of a crime attempted in a room from which there was apparently no possible exit. Leroux’s innovation was to create rival detectives, setting a brilliant young investigative journalist against a grizzled master sleuth hired by the police to help with their inquiries.
The twist is that one of Leroux’s investigators turns out to be the culprit. This blurring of the line between detective and crook heightened what Poe had instinctively recognised when he made detective Dupin and criminal Minister D– so similar, sharing brilliant intelligence, mathematical ingenuity and poetic imagination. Almost every fictional detective since has shared this ambiguity: to get results, sometimes you have to cross the line; you cannot always play strictly by the rules.
© Tom Gauld/Heart
Claude Wischik was born in France and raised in Australia. He studied mathematics and philosophy, dabbled in theatre and film, then went back to university to do a medical degree. Having decided to specialise in psychiatry, he came to the UK – the Medical Research Council Laboratory of Molecular Biology (LMB) in Cambridge – to get a PhD.
He joined a small team trying to identify the protein in Alzheimer’s tangles. His first task was to purify the samples so that they contained just the tangle filaments. Wischik struggled to make headway, so a colleague recommended trying a compound called alcian blue on the samples. To his horror, far from purifying them, it made the tangles disappear.
In the end, they managed to purify the tangles. By this time, Michel Goedert had arrived and between them they proved that Alzheimer’s tangles were made of tau. But within a few years, Wischik was off the team. He disagreed with the others about what made tau accumulate in the brain during Alzheimer’s disease. Already battling against the dominance of amyloid beta, the tau camp could not support more divisions.
“I was really the black sheep in the LMB,” says Wischik, “which is why I had to get out.” He completed his psychiatry training and got a job as director of the Cambridge Brain Bank Laboratory, through which he was able to continue his research on Alzheimer’s disease and tau.
He had not forgotten the compound that dissolved tangles. He had even found others that did the same, or did it better. The best was a dye called methylene blue, and he was sure it could be the basis of a drug to undo tau tangles: “I decided, in this amyloid versus tau war, the only way to win is to win,” he says. “The only way to win is to develop a treatment that works.”
Years later, Goedert saw Wischik on TV. The BBC was reporting from the 2008 International Conference on Alzheimer’s Disease in Chicago, where Wischik had just made a big announcement. Goedert remembers the interview: “He said what he’d discovered was the most important discovery since Alzheimer’s description of the disease.”
Serendipity had struck. The father of Wischik’s son’s classmate happened to be a surgeon and venture capitalist. He offered Wischik the financial opportunity he needed. Together they set up a company, TauRx Therapeutics, raising the money to turn methylene blue into a fully fledged tangle-busting drug.
In 2008, they announced results of a trial of 321 people in which their drug had, over two years, slowed the mental decline of people with Alzheimer’s disease by 90 per cent compared with those taking a placebo. By 2012, TauRx had raised $200 million to undertake the last stage of drug testing – a clinical trial involving many hundreds of patients across 22 countries.
Wischik has clearly persuaded private investors to back his idea. He has persuaded the regulators at the European Medicines Agency and the Food and Drug Administration in the USA that it is not a dangerous gamble for patients to try the drug. But he has yet to persuade the Alzheimer’s research community.
“I thought that when we announced the data that this would just be embraced,” he says, “whereas I got really criticised.” What worries other researchers is that the new trial is going ahead before the full data from the previous trial have been peer reviewed and published in the scientific literature for all to scrutinise. Until then, they do not have much more than his word that it works: “All I can say is, he should publish his work,” says Goedert. “Then we could tell.”
Wischik is focused more on future results than on writing up past data. The next trial of the drug, now called LMTX, is due to report in mid-2015, when we will discover whether or not it works better than the amyloid-based drugs that, so far, have all fallen at this final hurdle.
“We will publish the wretched data,” Wischik says. “I’ve got ten papers I need to write, as well as business plans and investor relations and patents and going around begging for money, running the team – and I get yelled at by my wife for not writing papers. I will, but I just need a bit of space to do it in and that space doesn’t seem to arise.”
© Tom Gauld/Heart
With just three short stories, Edgar Allan Poe created detective fiction, a genre now so ubiquitous it seems odd that anyone ever had to invent it. Monsieur Dupin, the archetypal sleuth, has been copied, transformed and reinvented innumerable times since.
One of Poe’s admirers was the Argentinian writer Jorge Luis Borges, who in 1941 – precisely 100 years after The Murders in the Rue Morgue – began publishing his own trio of detective stories: The Garden of Forking Paths, Death and the Compass, and Ibn-Hakam al-Bokhari, Murdered in His Labyrinth. These were no simple homage: Borges folded the genre in on itself, trapping his detectives in figurative and literal mazes as each case becomes increasingly convoluted. Success and failure are redefined as his protagonists become consumed by cases that often fail to resolve, even when the solution is revealed.
Borges’s constructs are more elaborate than Poe’s. They are self-conscious warnings against assuming that every puzzle has a clew to follow to comprehension. And as one of Borges’s characters says, “There’s no need to build a labyrinth when the whole world is one.”
Those who have lost a relative or friend to Alzheimer’s disease often say it is as if the person dies twice – the mind first and then the body, unaware of who they are, what they are or where they are. Faced with a disease that traps people in their own degenerative brain, our inclination might be to reconstruct that disease in all its complexity in the hope of understanding it. But scientists will tell you that what you really need to do is construct a thread that will guide you out of the labyrinth.
“Yes, the world’s complicated – we get that,” says psychiatrist Simon Lovestone. “But if one wants to understand the world, the approach that has been most useful in human history has been to reduce it to tractable elements.” In other words, build a simpler construct from the evidence. The theory based on acetylcholine and memory was one attempt, as is the amyloid hypothesis, as is the idea that it is all about tau.
If Wischik’s tau drug proves successful, he will claim victory in the “amyloid versus tau war”. At last, there would be something to work with, to fine-tune and make more effective at treating Alzheimer’s disease. But it would not prove his tau-only explanation of the disease: an Alzheimer’s drug that dissolves tangles in the brain would be just as effective if tau’s destructive function is initiated by amyloid beta. If his LMTX drug works, perhaps none of that matters; if it fails, we will still not know how the tangles get there, the extent to which amyloid beta is involved or whether the amyloid cascade hypothesis is true or not.
And the amyloid hypothesis itself may not be as simple as it first promised: “If I had a critique of the field,” says Lovestone, “it is that we haven’t actually taken seriously the fact that there is a cascade and that we actually need to deconvolute that cascade.” While the amyloid hypothesis presents a short series of steps that lead from a fault to Alzheimer’s disease, we still do not understand the potentially complicated links that connect each step.
With amyloid-based treatments stalling in trials, researchers are delving deeper into the gaps in the amyloid hypothesis, adding to the clues we have but threatening its original simplicity. Some scientists are still trying to prove that amyloid beta can be directly toxic to nerve cells, while others are working out the role of the immune system, which causes inflammation in the brains of people with Alzheimer’s disease. John Hardy says discovering the normal function of amyloid beta is the missing piece of the jigsaw.
Everyone knows that ageing is the biggest risk factor for Alzheimer’s disease and unravelling that link is unlikely to be straightforward either. Meanwhile, there are reports that dementia is linked to diabetes, hygiene, copper, alcohol, obesity, gum disease and more. With theories multiplying yet again, there is a danger that the research is creating a labyrinth of its own, confusing understanding and even blocking the scientists looking for a solution.
Hardy has written that if Auguste Deter were alive today, “her sad prognosis would be much the same as in 1906”. Alzheimer’s disease may well be a construct, but with no author to decide where the answer lies, we are essentially lost. We may be deep in the labyrinth or just round the corner from the exit. The only course is to follow every possible thread, hoping each time that this will be the clew that leads us out.