Is 'miracle' Alzheimer's cure too good to be true?
Walter, a 79-year-old former police officer, has Alzheimer's disease, the most common form of dementia. The disease gradually chipped away at his brain, rendering the simplest of functions impossible. "My father became unable to walk, unable to maintain basic hygiene," Jim recalls. "He had withdrawn into himself and was leading an empty, non-interactive, almost painful existence."
Yet three years on the clock appears to have been turned back on Walter's disease in a way that few experts would have thought possible. Now he can walk without a stick and manages to call his son three times a day. "In fact, the phone calls have started to bug me," Jim jokes.
Walter made his seemingly miraculous recovery after receiving a controversial new treatment designed by physician Edward Tobinick, director of a private clinic in Los Angeles called the Institute of Neurological Research (INR) who also holds an unrelated position at the University of California. Under Tobinick's supervision, Walter and 50 or so other Alzheimer's patients receive weekly or fortnightly injections of an anti-inflammatory drug called etanercept, normally used to treat rheumatoid arthritis.
It doesn't work for everybody, but when it does the results are striking. Earlier this year, Tobinick published a case study on Lawrence, an 81-year-old man with severe memory problems who, 2 hours after his first injection, was "back where he was before," according to his wife Rosa (Journal of Neuroinflammation, DOI: 10.1186/1742-2094-5-2). Last month Tobinick published his latest results showing rapid improvement in verbal skills even in people with severe Alzheimer's (BMC Neurology, vol 8, p 27). Videos showing patients before and after treatment, available on the INR website, tell similar tales, as do families' own videos posted on YouTube.com. These "miracle awakenings" have created a storm of media interest and speculation about a possible breakthrough for a disease that already blights the lives of 30 million people and their families worldwide and is expected to increase dramatically (see diagram).
Among Alzheimer's experts Tobinick's treatment has been greeted with a mixture of excitement and scepticism. "It seems strange that years' worth of damage to the brain by the pathology of Alzheimer's disease can be reversed in a matter of minutes," says Gordon Wilcock, a clinical gerontologist at the University of Oxford. "It seems implausible. But I've been in medicine long enough to know that things that seem implausible sometimes actually happen to work."
Clive Holmes, a biological psychiatrist at the University of Southampton in the UK, is even more cautious. "With Alzheimer's, the damage to the brain happens gradually over a period of years," he says. "The best thing you could normally hope for is to keep things where they are. The underlying structure - of dead brain cells - will still be there."
Part of the reason for this scepticism is that treatment with etanercept flies in the face of the conventional explanation of Alzheimer's. This holds that the symptoms of the disease are caused by clumps of misfolded protein in the brain called beta-amyloid plaques. Once they appear the plaques trigger a cascade of irreversible events: formation of tangled fibres within neurons; inflammation; the destruction of neurons and shrinkage of the brain. Etanercept, however, has no effect on beta-amyloid plaques.
There is, though, an increasingly credible alternative model of the disease in which inflammation is the primary causal agent. This is regulated by signalling proteins called cytokines, which recruit and activate immune cells. One of the key cytokines is tumour necrosis factor alpha (TNF-alpha). That, says Tobinick, is why his treatment works. Neutralising TNF-alpha is exactly what etanercept does, mopping up the molecule by binding to it. "Etanercept is very specific," he says.
There is also evidence that TNF-alpha works together with beta amyloid in the brains of Alzheimer's patients to hinder the formation of new connections between neurons - a process central to normal memory and learning. Roger Anwyl at the University of Dublin, Ireland, has found that in mice engineered to lack a TNF-alpha receptor, beta-amyloid plaques did not prevent the formation of new connections (European Journal of Neuroscience, vol 22, p 2827).
Tobinick got the idea to try the drug on Alzheimer's patients after he used it to treat people with severe back pain. "I noticed that it was having cognitive effects," he says. "These people didn't have Alzheimer's but their mood and their thinking would rapidly improve. You could see a striking difference in their faces - more animated, more aware and more attentive. I felt it could not be attributed to loss of back pain alone."
Tobinick decided to try the drug on people with Alzheimer's, and set up a six-month pilot study in which 15 patients were given weekly doses (Medscape General Medicine, vol 8, p 25). The results seemed impressive, with cognitive function improving in all participants. However, the study lacked a crucial element: a placebo control.
The placebo effect is known to be particularly strong in Alzheimer's. When given a placebo, patients' scores improve on the simple tests used to assess cognitive function, such as drawing the numbers on a clock face and counting from 1 to 10. "It is shocking how dramatically better patients do on the cognitive tests when they've been given only a placebo," says Ben Barres, a neurobiologist at Stanford University, California. "Almost every study shows this effect. Without careful controls, you can really be fooled."
Tobinick acknowledges the need to do a placebo-controlled trial but insists his results are meaningful. "In long-term Alzheimer's studies, the placebo effect declines over six months. But in our study we documented continued improvement," he says.
Despite these early apparent successes, his peers remain to be convinced. By posting videos on the INR website and actively courting media attention, many feel that Tobinick has put the cart before the horse. Critics suggest that even if the treatment fails in controlled trials, the publicity might still attract a stream of hopeful patients to his private clinic for a treatment that costs between $10,000 and $40,000 per year. He also holds patents on the treatment and owns shares in Amgen, the company that makes etanercept under the trade name Enbrel.
Tobinick rejects these criticisms as unfair. "It is extremely difficult and expensive for an individual physician to conduct a placebo-controlled trial without industry or academic support - and no one was willing to help." The only practical option was a pilot study, Tobinick says. "Our clinical results were highly statistically significant. Traditionally, a successful proof-of-concept study is followed by a placebo-controlled study, financed by the drug manufacturer."
This hasn't happened, though. Amgen has publicly distanced itself from Tobinick's treatment, stating on the company website that "there is insufficient and unsubstantiated scientific data to support the use of Enbrel as a means of treating Alzheimer's disease." The company also finds "the rapidity of response... to be implausible" and states that without placebo-controlled data, it is not possible to evaluate the treatment.
Tobinick is disappointed with the company's attitude. "Isn't the raison d'etre of pharmaceutical companies to find new treatments for important diseases with great unmet medical need?" he says. "How can they fail to investigate when a clue has emerged?"
Ten other physicians, trained by Tobinick, are now offering the treatment. Del Charbonier from Warren, Michigan, is one of them. He uses it on four patients, including his father. "It isn't a cure," he says. "The patients get better and do quite well - up until day six, when things start to deteriorate and they need another shot. My dad still forgets things. But it is definitely the best treatment that I can give to any of my patients. The other drugs just don't work: they claim to halt the decline, but they never claim to make patients better."
“It is definitely the best treatment I can give to any of my patients. The other drugs just don't work”
Most existing drugs and experimental treatments for Alzheimer's focus on trying to break up and remove the amyloid plaques. So far with little success, according to Susanne Sorenson, head of research at the London-based Alzheimer's Society. "The companies keep saying they have drugs in trials but you never see a full set of data or a drug coming up for licence. At the end of the day, they haven't produced the goods so the logic is that the drugs must have failed."
In the latest setback, a vaccine designed to break up beta amyloid successfully removed plaques but did not slow the progression of the disease (The Lancet, vol 372, p 216). However, another plaque-attacking drug, PBT2, was recently reported to improve brain function (The Lancet Neurology, DOI: 10.1016/S1474-4422(08)70167-4).
So are these various attempts at treatment focusing on the wrong thing? There is some evidence that they might be. In 2001 a group at the University of Cambridge autopsied the brains of 109 elderly people who had died with no symptoms of dementia. One-third of them had plaque content similar to people with Alzheimer's (The Lancet, vol 357, p 169). "It is an important and intriguing finding," says Holmes. "By definition, you have to have plaques to have Alzheimer's disease, but on their own they are not sufficient to cause the disease. The plaques may be a red herring."
Even if blocking TNF-alpha could in principle alleviate Alzheimer's symptoms, it's not clear how a drug like etanercept could get into the brain in the first place. The main difficulty is that brain tissue is surrounded by tightly packed cells which make up the blood-brain barrier. This highly selective filtration system protects the brain from circulating pathogens and toxins, and keeps out large molecules, including drugs like etanercept.
Tobinick believes that he has found a way around this problem. He injects the etanercept into the neck, into a network of hundreds of tiny veins running around and through the spinal cord. Then he tilts the patient's head towards the floor at an angle of about 30 degrees for 5 minutes.
The rationale is that the spinal veins join up with those in the brain, and while most veins in the body have valves to ensure that blood only flows in one direction, these veins do not. That means blood can flow both into and out of the brain, he says.
In as-yet-unpublished research, Patrick McGeer at the University of British Columbia in Canada, working with Tobinick, found that etanercept could be carried to the brain in this way. After injecting etanercept with a radioactive marker into rats' tails and tilting the animals downwards, PET scans showed that the drug rapidly reached the fluid filled cavities, or ventricles, in the centre of the brain. It's not clear whether it has crossed the blood-brain barrier at this point, however, and the study found no evidence that the drug moved from the cerebrospinal fluid in the ventricles into the deeper brain tissue where it would need to go to treat Alzheimer's.
Still, if etanercept could get this far in humans, it would be pretty close to where it might need to be. Tobinick points out that some brain areas involved in Alzheimer's, such as the hippocampus, are located near one of the brain's ventricles. It's also known that molecules reaching the cerebrospinal fluid can penetrate deeper into the brain. "Once in the cerebrospinal fluid, even large molecules can gain access," says Conrad Johanson of Brown Medical School in Providence, Rhode Island, who has reviewed this area. Vitamin C reaches the neurons in this way rather than crossing the blood-brain barrier, says Johanson.
For now, there are no clear answers. Tobinick plans to continue using his treatment and says he hopes that clinical trials will be done to answer many of the outstanding questions. McGeer points out that a controlled trial would not need to be large or long-term in the first instance, because it would be aiming to replicate results that take place within minutes or hours.
"At the moment, clinicians are plodding along prescribing drugs that do not deal with the underlying pathology," McGeer says. "Confirmation is urgently required to determine if, for the first time, there is a treatment that really works."
While Tobinick waits for funding to carry out a placebo-controlled clinical trial, Jim Skotchdopole is already convinced that the treatment works. "We've got our father back," he says. "He is alive again."
Mental Health - Discover the latest research in our continuously updated special report.
Daniel Elkan is a freelance writer based in London
From issue 2668 of New Scientist magazine, 06 August 2008, page 32-35