What a piece of work is a man! How noble in reason, how infinite in faculty! In form and moving how express and admirable!
In action how like an angel, in apprehension how like a god!
William Shakespeare, Hamlet
EVERYBODY KNOWS DR Stefánsson,’ said the driver as the taxi lurched across the snow. A bitter Arctic wind caked the windows in frost and a dimly lit sky hung overhead, the midday sun barely above the horizon. I’d come to Reykjavík in Iceland: a small, flat rock in the North Atlantic, whose inhabitants–not many, but some–are virtually immune to Alzheimer’s. How was this possible? And what did it mean?
We pulled up at our destination. As I climbed out of the car, the driver added: ‘I haven’t given him my DNA yet. But I will soon!’ To Icelanders, I would learn, this kind of talk was pretty normal.
On an August day in 1996 a tall, Icelandic, exceedingly philosophical man named Kari Stefánsson had an idea. A neurologist and pathologist, he’d seen countless Alzheimer’s patients–both dead and alive–and was beginning to tire of the slow, incremental and, in his view, erroneous approach to the problem. Biologists, he thought, could contemplate theories until the end of time, but they still wouldn’t have a concrete lead for drug companies to test. His belief was that not enough attention had been paid to a simple yet unalterable truth: the brain is hard-wired by genetics. Differences in the sequence of DNA’s four-letter code was the cardinal difference between Matthew and the wheat used at the Last Supper. It was the Holy Grail, Stefánsson insisted. And so, after twenty years toiling at American universities, he decided to return to Iceland with the singular purpose of eliminating common diseases by mining the genome of the Icelandic people.
It wasn’t nearly as ridiculous as it sounds. With record low levels of immigration since the Vikings settled Iceland 1,100 years ago, the island’s genetically homogenous population made it a unique natural laboratory. It was to Stefánsson what the Galapagos Islands were to Darwin. Unlike Darwin, however, Stefánsson was going to need a lot more than wits and a notebook. He wanted to collect and sequence the entire Icelandic genome, some 300,000 people. The cost would be enormous–certainly more than any money he could obtain publicly. And to make matters worse, it was illegal for an individual to create their own database on healthcare; many saw it as a disturbing, Orwellian prospect. So Stefánsson set up a private company, called DeCODE Genetics, and lobbied the Icelandic government to change the law.
He succeeded on both counts, and DeCODE immediately spread the word around Reykjavík and the wider community, asking anyone and everyone to give blood and/or saliva to help unlock the mysteries of human diseases. To ease the effort, the company sent out cheek swabs in the mail, telling people that a courier would come by to collect their sample, if they chose to give one. As an incentive, and to reach the remote villages outside the capital, the couriers were volunteers from the Icelandic Search and Rescue charity, which got a $20 donation for every sample it collected.
Not everyone was enamoured by Stefánsson’s plan. Some saw it as an infringement of their private, most personal information. As one Icelandic journalist put it: ‘It makes me very nervous… in Iceland everyone knows everyone and when you give your DNA sample, you are not just giving information about yourself.’1 Stefánsson couldn’t have disagreed more. The way he saw it, a healthcare system was only able to treat people by using the information amassed from previous generations. How was it fair, therefore, for anyone to take advantage of such a system and yet simultaneously refuse people the right to help improve that system for future generations?
He had a point. Critics had failed to appreciate that human samples were the lifeblood of medical advancements, not to mention optional and anonymous. Fortunately, many Icelanders did appreciate this. By 2004, 80,000 Icelanders had given samples; 120,000 by 2007, nearly half of Iceland’s population. The DNA sequencers could hardly keep up. To cope with the deluge, DeCODE installed colossal freezers containing gigantic car-manufacturing robots (one freezer contained half a million vials of blood) and supercomputers capable of holding 20 petabytes of data. To put that into context, it’s the equivalent of 10 billion floppy disks, or 10 trillion pages of text. But the usefulness of all that data paled in comparison to what turned out to be Iceland’s most precious resource: genealogy.
Genealogy is a national obsession in Iceland. Almost every Icelandic saga begins with a lengthy description of family trees.2 Here’s one example: ‘There was a man named Ulf, the son of Bjalfi and of Hallbera, the daughter of Ulf the Fearless. She was the sister of Hallbjorn Half-troll from Hrafnista, the father of Ketil Haeng…’ And another: ‘There was a man named Onund. He was the son of Ofeig Hobbler, whose father was Ivar Horse-cock. Onund’s sister Gudbjorg was the mother of Gudbrand Lump, whose daughter Asta was the mother of King Olaf the Holy. On his mother’s side…’ And on it goes. Icelanders have done this for centuries. Stefánsson himself can trace his ancestry back to the Viking poet Egil Skallagrimsson, who lived in ad 900.
This record-keeping has proved essential for the DeCODE project. Since DNA is inherited in chunks of code–vast stretches of ATCG, rather than being passed down in individual ‘letters’–many Icelanders’ genomes needn’t be sequenced. They could simply be inferred by combining family trees with clever computers.
When Stefánsson put this strategy into action, the discoveries came in thick and fast. New genes underpinning heart attack, autism, schizophrenia and many cancers were unearthed, as well as genes influencing smoking behaviour, skin pigmentation and even creativity. The discoveries made headlines coast to coast, and Stefánsson was enshrined in Time magazine’s top 100 people transforming the world. I remember the day I first heard about him. I was sitting in the lab, waiting for an experiment to finish, frustrated by how slow and inefficient academic research can often be. His success was bewitching. He was a maverick, a misfit, a rebellious pragmatist, unencumbered by politics and fully aware that curing big diseases requires big data and big capital. In 2012 the US pharmaceutical giant Amgen kept his dream alive by purchasing DeCODE for a little over $400 million. For Alzheimer’s research, this was all a prologue to another vital clue.
On 2 August 2012 Stefánsson published data showing that about 1 per cent of 1,795 Icelanders carry a genetic mutation shielding them from Alzheimer’s.3 Astonishingly, it was found in APP, the same gene underlying Carol Jennings early-onset. But where Carol’s mutation was due to a ‘T’ that should have been a ‘C’, the Icelanders’ mutation was a ‘T’ that should have been an ‘A’. This tiny genetic fluke had the effect of shifting beta-amyloid into reverse gear: while Carol’s brain became saturated with amyloid, the Icelanders’ brains produced half the usual amount. It was resounding support for John Hardy’s amyloid hypothesis, and an olive branch to pharmaceutical developers, now weary of constantly having their fingers burned by this lead.
But the mutation also hinted at some deeper, primordial truths–about ageing and why Alzheimer’s even existed.
‘I’ll give you an example of how strange memory is,’ Stefánsson offered, seated in his spacious office at DeCODE’s headquarters in the suburbs of Reykjavík, the imposing Hallgrímskirkja Church spire and craggy visage of Mount Esja visible from the window. ‘When I was seven years old, I went to an old movie theatre located exactly where this building is. The movie was The Scarlet Pimpernel. Then, about thirty years later, I was in Chicago taking my daughter and her friend to the movies. On the way back, the following verse suddenly popped into my head: “Is he here or is he there, the French are seeking everywhere, is he in heaven or is he in hell, the elusive Scarlet Pimpernel”. I haven’t seen that movie since I was seven… Where does this come from?’
Just shy of seven feet, Stefánsson is a Herculean man, with ice-blue eyes and thick white hair. His father was an author and radio journalist, disappointed by his son’s resolve to pursue science instead of writing. Stefánsson still remembers the summer night in 1968 when he and a classmate drank through the night, talking about life, purpose and the world of alternatives to choose from, before applying the very next day to medical school. Now, aged sixty-six, his childhood cinema recast as a genetics super-lab, he wakes every morning and comes into work feeling as if he’s ‘playing in the sandbox’.
He is an astute neurobiologist and a brilliant geneticist, but is one of the few people to stress how little we know about memory. ‘We haven’t the faintest idea how the brain generates memory. We don’t even have a useful definition of memory. And you’re going to write a book about a disease that assaults this function, but you cannot even define it! What the hell are you doing?’ Again, he had a point. I’d become so wrapped up in trying to understand Alzheimer’s that I’d swept the basic premise of what memory is under the rug.
‘What about LTP and the synaptic network idea?’ I countered. Wasn’t this at least somewhere in the ballpark?
‘What can I say?’ he acknowledged with a shrug of ambivalence. ‘It sounds reasonable. But my God is it magical.’
One of Stefánsson’s trademark characteristics is his ability to view disease at the population level. Where many see a cruel and utterly meaningless defeat, Stefánsson sees the tragic but inevitable cost of evolution.
Take schizophrenia: in March 2015 DeCODE used the DNA of 86,000 Icelanders–plus a further 35,000 people from the Netherlands and Sweden–to show that schizophrenia and creativity actually share genetic roots.4 The same genes underlying the disorder, it turns out, are also more common in painters, dancers, writers and musicians. So it isn’t that you develop schizophrenia and therefore think differently, Stefánsson explained; it’s much more likely that you think differently and therefore develop schizophrenia. Since the prevalence of schizophrenia is only 1 per cent worldwide, people with these genes have a roughly 10 per cent chance of developing the disorder. And this, he claims, is the small but striking price our species pays for the Mozarts, Shakespeares and Van Goghs of society.
A similar paradigm may exist for Alzheimer’s owing to another special property of the Icelandic mutation: it protects against memory loss and cognitive decline in normal ageing, too. By using a cognitive test given to residents of Icelandic nursing homes, DeCODE found that people who carry the mutation are nearly eight times more likely to reach eighty-five as mentally sharp as they were at their peak. For Stefánsson, this is irrefutable proof that Alzheimer’s is simply an accelerated form of ageing. ‘What’s often lost on people is that the brain is just an organ. And like everything else it’s perishable. I mean, you look at yourself in the mirror in the morning and you see over the years that you change: your skin changes, your hair changes, your muscles change. Your brain changes as well. It deteriorates.
‘I think Alzheimer’s is somehow an expression of this fact. I mean, is it a design flaw when a disease terminates a life when we are relatively old, or is it a masterpiece in the design? That depends on how you look at it. It depends on whether you look at it from the point of view of the individual, or the point of view of the species.’
‘Assuming that’s true, what evolutionary bill is Alzheimer’s footing?’ I asked. ‘Or is that the wrong question?’
‘This is exactly the right question,’ Stefánsson replied. ‘We’re born to have offspring and die. What our roles are beyond that is beyond me. I don’t know why we last this long. But there is some data to indicate that the Grandmother Effect is real.’
The Grandmother Effect is a fascinating theory. Formulated by anthropologists in the late 1990s, it argues that the reason grandmothers live many years after menopause is to help daughters with childcare. It was the Hadza hunter-gatherer tribe in Tanzania who first lent credence to the idea; researchers found that mothers in the tribe have more children if their own mothers helped raise them. (Grandfathers, by contrast, can stay fertile until they die and so the explanation for their longevity is that it’s either for them to continue mating, or that it’s some kind of genetic side effect of female longevity.)
It goes without saying that child-rearing demands grandmothers be mentally sound, certainly Alzheimer’s-free, and today evolutionary biologists are uncovering genes responsible for exactly that. One such gene, CD33, was found in our closest relative, the chimpanzee. Intriguingly, though, this gene only grants mind-protective status when it’s found in humans. And since chimps and other primates usually die once their fertility ends, it’s possible that genes like this have evolved solely in humans to make the Grandmother Effect a reality. As the distinguished Indian physician Ajit Varki told a journalist in 2015: ‘Grandmothers are so important, we’ve even evolved genes to protect their minds.’5
Still, Stefánsson is no fatalist about ageing and Alzheimer’s. Even if ageing is ‘paid for’ by Alzheimer’s, we have still evolved minds to eradicate the illness. Before getting to the knotty issue of treatment, however, I wanted to know what he thought about the influence of stress, diet, education and sleep–the topics still fresh in my mind.
He took a dim view. ‘I think there’s no solid evidence for any of these. There have to be environmental factors for Alzheimer’s, and whether it proceeds slowly or fast, but I honestly don’t know what they are.’
It wasn’t the answer I wanted to hear. Then again, much of what I’d learned was an answer I didn’t want to hear; it was almost becoming a sign of the truth. In any case, lifestyle measures were a low-hanging fruit. Iceland’s real gift was the message it sent to the pharmaceutical industry.
Around the time Big Pharma was testing amyloid immunisation in clinical trials, vast swathes of neuroscientists were devising a back-up plan. What if, they thought, instead of trying to rid the brain of amyloid, we prevented it from ever arising in the first place–like in the brains of Icelanders with the mutation?
Since the discovery that amyloid was merely the by-product of a normal protein–the amyloid precursor protein, APP–questions were asked about what the APP protein was actually doing in the brain. Beyond sitting at a neuron’s surface, with one end jutting inside the cell and the other out, no one really knew. Perhaps it was just another run-of-the-mill signalling molecule, a ‘sedan’ on the grand intercellular highway of brain chemistry. Whatever its purpose, it certainly had a routine: first an enzyme chopped off a large piece of it, which then allowed a second, smaller piece to be released from the neuron. Although I should say unleashed, for this small piece was beta-amyloid, the substance of plaques.
In 1999 five independent groups of scientists identified this chopping enzyme. They called it BACE (beta-site APP-cleaving enzyme).6 It turned out that Carol Jennings’s genetic mutation caused BACE to ramp up its activity, which had the effect of churning out beta-amyloid much faster than normal. It was akin to a faulty traffic light stuck on green–and too many sedans were getting through. So scientists looked into the possibility of blocking BACE and redressing the balance.
The results were not encouraging. From 2003 to 2011 a tide of animal tests unveiled serious side effects. Mice genetically engineered to lack the enzyme suffered blindness, seizures, spine abnormalities and memory problems to boot. Switching BACE off was clearly not ideal. What about chemically restraining it? In 2011 E-Lilly was among the first to try this tack. The ‘BACE-inhibitor compounds’, as they became known, were certainly better, but blindness remained a vexing fly in the ointment.
Still, progress of a kind.
So Lilly kept at it, tweaking and retweaking the recipe until the animals finally appeared normal. Compound LY2886721 was the coveted batch. It produced no side effects and yet, crucially, reduced beta-amyloid formation in the animals’ brains. A success! The company immediately moved to human trials. Upon giving forty-seven healthy volunteers daily doses of the drug for a fortnight, everything looked fine. Emboldened, Lilly funded a six-month Phase two trial in 130 cases of mild Alzheimer’s.
Here, you guessed it, is where things went wrong. An undisclosed number of patients showed signs of liver damage. Not wanting to take any chances, Lilly immediately terminated the trial. Merck, another US pharmaceutical company, picked up the baton. Their drug, dubbed MK-8931, made it through eighty-eight healthy volunteers with no side effects. And so they cautiously pressed on. Despite Merck’s headway, however, the message in the pharmaceutical industry was clear: invest elsewhere.
One could hardly blame them. Between 2000 and 2012, of the 244 Alzheimer’s drugs tested in 413 clinical trials, only one was approved (NamendaTM, a drug similar to the acetylcholinesterase inhibitors, and similarly insufficient). In total, the drug candidates racked up a lamentable 99.6 per cent failure rate–even higher than cancer, at 81 per cent.7 Our unsophisticated grasp of the disease, combined with the dizzying cost of drug development–it costs about $100 million per trial; over $2 billion all-in–made Alzheimer’s drugs, in the words of one pharmaceutical chemist, ‘almost perfectly set up for expensive failures’.8 The reluctance for renewed attempts almost seemed a fait accompli.
But then something magical happened. Further detective work at DeCODE revealed that the Icelanders’ protective gene caused their BACE enzyme to reduce its activity. In other words, the mutation was a natural BACE inhibitor. If that wasn’t proof this lead was worth pursuing, nothing was.
Wide-eyed and reinvigorated, Big Pharma returned to the table. And to share the risk, they teamed up: Lilly joined forces with AstraZeneca, the British–Swedish pharmaceutical giant, pledging a whopping $500 million to co-develop a new BACE inhibitor; Eisai, the Japanese company, struck a deal with the US company Biogen; and Swiss-led Novartis partnered with Amgen. The abundance of heavyweight competition marked a momentous victory for Alzheimer’s research.
Hungry for a release date, I rang every company.
‘We’re talking somewhere between five and ten years,’ said Sasha Kamb, Amgen’s Vice-President of Discovery Research. ‘DeCODE has proved that this idea should work, so I think the only question left is: when do we need to intervene and by how much?’
Ricardo Dolmetsch, Global Head of Neuroscience at Novartis, was even more optimistic. ‘Between three and eight years. I think Kari Stefánsson’s data provided the nail in the coffin that beta-amyloid is important and that if you inhibit BACE that would be a good thing.’
The most cautious estimate I got during these prying exchanges was from a representative for Eisai. ‘Seven to twelve years,’ she had said, not exactly bursting the bubble.
As to whether they will work, I kept something Stefánsson had told me fresh in my mind. ‘They’re going to be spectacular,’ he’d said.
Driving through the barren lava fields of the Reykjavík peninsula, en route to the airport, I mulled over Stefánsson’s special role in the abolition of Alzheimer’s. No doubt he was another William Summers, another outlier, only one with the formidable power of genetics on his side. When his contribution would actually help people like my grandfather, Arnold, Carol, Marie, Victoria, Li and Pam, I couldn’t say. Big Pharma’s projections seemed ambitious, but I vowed to remain optimistic.
Looking out of the window, at the flat expanse of all-consuming darkness, I suddenly glimpsed the fleeting glow of a streetlight puncturing the polar night. For a brief moment I could see the snow was melting.
Spring was coming.