Article: The doggie will see you now

The Sunday Times (UK)
Saturday, May 06, 2006

The doggie will see you now

By Charlotte Hunt-Grubbe

Rescuing people from avalanches is old hat. Now, man's best friend can 
diagnose cancer, explain our phobias - even warn us of terrorist attacks.

American security officials were in a state of high anxiety on Saturday, 
September 24 last year. Air sensors placed above Washington, DC, to alert 
the US government to biological terrorist attacks had detected a potentially 
dangerous bacterium. They urgently needed to contact accident and emergency 
departments to see if unusually high numbers of people were coming in with 
symptoms of a pneumonia-like disease.

Officials scrambled for the phones, but they just couldn't collate the 
information from the hundreds of hospitals in and around the American 
capital fast enough. The situation seemed bleak. But help was about to come 
from an unlikely source. A thousand kilometres away, in a laboratory at 
Purdue University in the leafy town of West Lafayette, Indiana, the vet and 
epidemiologist Dr Larry Glickman received a call from the Department of 
Homeland Security. He checked through his computer database of thousands of 
up-to-the-minute medical records. Nothing looked suspicious, and within 
minutes he had given the authorities the all-clear: it was a false alarm, 
not a bio-terror threat. But how exactly did he do it?

Glickman enjoys teasing the engineers and physicists on his campus who are 
hard at work designing new ways to detect deadly organisms in the 
environment. "They are developing the most sophisticated chemical-detection 
technology in the world, but mine beats theirs hands down. It detects 
bio-terrorist attacks, emerging infectious diseases, cancer, drugs, 
explosives and even helps us find disease-causing genes. I tell them it's 
called the D.O.G. - most of them don't understand what I mean."

We have long known that dogs can save lives as bomb detectors and in 
search-and-rescue teams. But Glickman is one of a growing number of experts 
who believe that these animals will prove invaluable to medical science. 
Starting with a hunch in the early 1980s, Glickman convinced the government 
to fund research into a particular type of lung cancer - a mesothelioma - in 
dogs. He was convinced there was a link between asbestos fibres in the 
owners' homes and dogs suffering from this cancer. "It's often 40 years 
before a person gets a mesothelioma following exposure to asbestos fibres," 
he explains. "But owing to their shorter life span, dogs usually develop it 
at seven to nine years of age." The research revealed that there was indeed 
a connection.

Glickman had hit on a novel way of keeping tabs on all kinds of 
environmental hazards around us. His training as a vet had taught him that 
dogs are susceptible to many of the same toxins as humans - and develop 
symptoms sooner - and what could be more useful than having these companions 
that share our lives alert us to hazardous things we can't see? Just a quick 
root around a flower bed is enough for dogs to pick up bacteria and 
chemicals - most of which they carry around quite happily without becoming 
ill - but when they start showing symptoms of a disease, this can warn 
owners to take precautions against what might be about to infect them.

"There are many situations where the dog is a sentinel to let us know about 
environmental hazards," says Glickman. He points out that dogs have been 
shown to be affected by Lyme disease  - a bacterial infection carried by 
ticks - two months before it can reach a level that poses a threat to 
humans. If vets see dogs with infected ticks, they can log positive cases by 
computer and help Glickman predict whether tick activity in specific areas 
is increasing. This could direct health departments to spray infected areas 
long before Lyme disease becomes a threat to humans. But the value of this 
system goes even further. If vets pick a few ticks off these dogs and send 
them to Glickman's lab, he can test to see what other organisms the ticks 
are carrying, such as tularaemia - the pneumonia-like disease that concerned 
US security officials last year. "Who knows how many diseases we could pick 
up in animals before they spread to humans?"

In 2004, Glickman realised there might be a wider application for his 
findings. If he could feed medical records from a chain of pet hospitals 
across 44 US states into a computer, he would be presented with a real-time 
map showing which diseases American dogs were carrying around in their 
blood. By logging into the system, he could see if there was a pattern of 
suspicious symptoms in any given area.

It was exactly what Homeland Security was looking for when the red alert 
went up over Washington last year. "There were 17 pet hospitals within 50 
miles," recalls Glickman. "We just logged into our system and looked for 
flu-like symptoms. And when we saw that there was no unusual activity, we 
gave the all-clear."

The database is now also proving invaluable for monitoring the safety of 
drugs used on humans and animals. Two years ago, the US Food and Drug 
Administration (FDA) withdrew a drug that had been routinely used to 
vaccinate about 250,000 dogs each year against heartworm - a mosquito-borne 
parasite that causes large worms to develop in the heart. The FDA had 
suspected the drug was causing adverse reactions. But it was much more than 
the health of dogs that was at issue: the World Health Organization had been 
using the same drug on humans in parts of sub-Saharan Africa, to see if it 
could prevent a common cause of blindness - onchocerciasis, which affects 
18m people. By analysing medical data from 6.8m dogs, Glickman showed that 
the drug was not a danger, and that the human trials could continue.

If dogs can alert humans to diseases, and to chemicals that are causing us 
harm, can they also help us find disease even closer to home - in our own 
genomes?

Lancelot is a striking six-year-old briard-beagle mix, resembling a 
curly-haired labrador. Standing up against his research-kennel gate in a 
garden off a quiet Philadelphia street, with black nose and shaggy paws 
pressed against the netting, his dark button eyes follow his playmates as 
they scamper over the grass. But five years earlier, Lancelot led a very 
different life. "He was totally blind at birth," says Dr Gustavo Aguirre, 
professor of medical genetics and ophthalmology at the University of 
Pennsylvania, who is researching inherited diseases of the eye. Lancelot was 
suffering from a canine form of a human genetic disease, Leber's congenital 
amaurosis (LCA), a cause of near-to-total blindness in children. In 2001, 
Aguirre and his colleagues at Cornell University and the Universities of 
Pennsylvania and Florida developed a way to restore Lancelot's sight. By 
injecting a genetically engineered virus into his eye, they delivered a 
healthy copy of the gene to his retina. The outcome? "A very happy dog with 
an extremely privileged life," Aguirre says.

Later this year, a group at the University of Pennsylvania will prepare to 
treat visually impaired humans with defects in the same gene that caused 
Lancelot's blindness. As the genetic instructions in dog and human cells are 
so similar, scientists hope the same therapy will work for humans. "It 
should be a very exciting year," says Aguirre. "I have my fingers crossed."

The domestic dog is an incredibly diverse animal: no other species of mammal 
on Earth has so many body shapes and sizes. The UK Kennel Club recognises 
204 breeds, from hefty, web-footed Portuguese water dogs to handbag-sized 
puffball pomeranians. But in creating such a remarkable diversity, breeders 
have inflicted our "best friends" with a variety of complaints. The majority 
of the breeds that we know now developed very quickly over the past 300 
years (from just 20 breeds in 1750, the number rose to 76 in 1905), and this 
was achieved through very rigid inbreeding by kennel clubs. Unfortunately, 
this inbreeding has weakened the natural mechanism for overriding genetic 
diseases, and has unmasked genes that would otherwise have remained hidden. 
The animals now carry a heavy burden of faulty genes: cancer, blindness, 
deafness and even psychiatric conditions can appear in dogs at similar 
levels as in humans.

So is there anything we can do to help our favourite breeds? If we could 
find more faulty genes - such as the one causing Lancelot's disease - could 
we remove them from the breed altogether? One potential solution comes from 
the very cause of the diseases: breed structure. As dogs are often bred with 
close relations, breeds share a narrowed pool of genes, which makes it 
easier for scientists to locate genes that are causing disease. But how can 
scientists read a dog's DNA to find these rogue genes?

Tasha, a brown-and-white boxer bitch, poses serenely for the camera. It is 
late 2005, and the biological journal Nature has just published the results 
of a two-year project to map her complete genetic code - the first genomic 
map of the domestic dog. Unbeknown to Tasha, she is a disease hunter's gold 
mine. Now that her genetic sequence has been unlocked, instead of having to 
trawl genomes for genes that may be causing disease, scientists can develop 
state-of-the-art tools to scan the DNA of sick and healthy dogs and identify 
any differences. If one breed suffers more than another from bone cancer, 
for example, researchers can compare the genomes of affected and unaffected 
dogs of that breed until they find the exact point where they differ.


Since the human and canine genomes are "about 75% similar", according to the 
Harvard geneticist Kerstin Lindblad-Toh, dogs are susceptible to many of the 
big killers we suffer from today, like cancer, heart disease and diabetes - 
and many of them are found in just one, two or a few breeds. "There are 
approximately 250 to 300 known genetic diseases in dogs, and it is likely 
that most of those are similar in humans," says Lindblad-Toh, who led the 
200-strong team that spent over two years sequencing and analysing the 
canine genetic code. If geneticists such as Lindblad-Toh find a disease gene 
in dogs and a good candidate for a drug to treat it, then dogs could help us 
in other ways - by speeding up treatment trials: since the animals have 
shorter life spans, you would receive conclusive answers to drug trials much 
more quickly in dogs than in humans. "I urge pet owners to contact 
researchers if their dog is suffering from a disease that plagues their 
breed," says Lindblad-Toh. "The more blood samples we have, the more disease 
genes we can look for. This is a chance for owners to do something for the 
future of their dogs - and because of the similarity in genes between humans 
and dogs, it will impact on human health too."

As well as sharing physical diseases such as cancer and epilepsy, dogs also 
share many of our emotional complaints. Irrational fears and other 
psychological conditions are prominent in certain breeds, and dogs are now 
being used to find the genetic basis of a number of disorders.

Dogs can even be obsessive-compulsive, it seems. "It is beyond 
heartbreaking," says Alice Moon-Fanelli, clinical assistant professor at 
Tufts University in Boston, Massachusetts. "The first time I saw it, I just 
knew I had to do something to stop it proliferating through the gene pool." 
Moon-Fanelli is describing her first encounter, 12 years ago, with an 
English bull terrier who was an obsessive tail-chaser. Witnessing this 
behaviour changed the focus of her scientific career. "We see separation 
anxiety, aggression, various phobias - like dogs terrified of 
thunderstorms - and even some that we liken to post-traumatic stress 
syndrome. I believe dogs and humans share all the same emotional types of 
things."

Many hard-wired canine instincts such as hunting, grooming and sex can be 
expressed in an abnormal fashion in dogs (as in humans), and these problems 
often crop up in certain breeds and run in family lines. In large, active 
breeds such as doberman pinschers, obsessive licking can lead to ulceration 
and, in extreme cases, to amputation. Other disorders include compulsive 
shadow-chasing and fly-snapping - in which the animals snap at imaginary 
insects. In the case of tail-chasing, Moon-Fanelli believes that the animals 
are displaying a form of displaced predatory behaviour. Serious, or 
"clinical", levels of tail-chasing are far more intense than the playful 
bouts of spinning seen in many pet dogs. In these clinical cases, the dog 
spins rapidly with no apparent focus, snapping and grabbing at the tail for 
up to 80% of its waking time, every day. The condition causes many dogs to 
suffer tail fractures, weight loss and exhaustion.

Moon-Fanelli hopes to use the sequenced canine genome to identify the actual 
genes involved in tail-chasing. This research may help owners to identify 
animals that are genetically predisposed to developing these problems in the 
future, so preventive measures can be put in place before the onset of 
full-blown illness.

Using a previously identified gene in humans for an obsessive-compulsive 
disorder (OCD), the team will establish whether the dog has the same gene 
and whether it correlates with affected animals. "For the past 12 years, I 
have been collecting blood, isolating DNA and developing a description of 
the disease for the English bull terriers that compulsively tail-chase," 
says Moon-Fanelli. "And finally the geneticists are ready."

To a dog, the world is about smells. A scent-marked gate post is packed with 
information about other dogs - decodable only by nose. Sniffer dogs are 
already employed all over the world by crime-fighting agencies - so why not 
in the research clinic?

In the summer of 2004, a brown cocker spaniel jumps up against a lab bench 
at Amersham hospital in Buckinghamshire, his long nose snuffling over rubber 
gloves and empty containers. "Get down, Tangle!" commands the clinical 
scientist Dr Carolyn Willis, deftly shifting urine samples onto a shelf, out 
of the reach of Tangle's nose. "They get so excited," she laughs. 
"Especially Tangle - he's always trying to get his nose into things."

Willis places seven Petri dishes in a line against the wall, and watches as 
a member of her training team guides Tangle past the dishes. The cocker 
spaniel sniffs the air above the plastic containers, then suddenly stops and 
drops to the ground, his nose inches away from one of the dishes. 
Remarkably, he has identified that the patient who gave this sample has 
bladder cancer.

The fact that dogs can detect cancer has long been mooted. In 1989 the 
medical journal The Lancet published a letter about a border collie/ 
doberman cross, who drove his owner to visit the doctor by frantically 
licking and biting at a mole on her thigh. The animal had detected a 
malignant skin melanoma, an early stage of skin cancer. Similar reports 
trickled in to the letter's authors over the next 12 years, and scientists 
began to take note. In September 2004 the British Medical Journal published 
the results of the Amersham study: the first real scientific evidence that 
dogs can detect cancer.

As Carolyn Willis is preparing her findings for a review by experts, 5,000 
miles away at the Pine Street Foundation in San Anselmo, north of San 
Francisco, the research director Michael McCulloch is standing holding a 
white polypropylene tube. "The fibres inside this hold the breath condensate 
of a woman with lung cancer," he explains, placing the tube in a perforated 
plastic container among four others, spaced evenly on the floor. A dog 
trainer enters the room, and Isabel, a silky Portuguese water dog, is 
released. The bitch sniffs the air over each tube, then stops suddenly by 
the third and sits down. Her trainer checks the number. Isabel is right: the 
correct sample is No 3.

Similarly intrigued by the Lancet reports of canine melanoma detection, 
McCulloch and his team embarked on a study of their own, to see if dogs 
could detect cancers much deeper inside the body, in the breast and lung. 
Aware that cancer patients have been shown to have traces of chemicals such 
as alkanes and benzene derivatives on their breath, he wondered if he could 
train dogs to detect these chemicals. It took professional trainers only 
three weeks before dogs detected breast cancer from breath samples at an 
accuracy rate of 88%, and lung cancer at an astonishing 99%. One of the 
samples accurately detected was from a woman with lung cancer who had died 
five years earlier. "We are very excited about the results," says Nicholas 
Broffman, the Pine Street Foundation's executive director. "The possibility 
of detecting lung and breast cancer early means treatment outcomes are 
better, and so is quality of life."

Does this mean that in the future we will be seeing "doctor dogs" in 
clinics? Sadly not. "Our aim remains to work with scientists and engineers 
to develop instruments to emulate dogs' olfactory capabilities," says 
Carolyn Willis. "That said, it is fully acknowledged by all concerned that 
the dog's nose is more sensitive than even the most sensitive mass 
spectrometer, and so the task will be difficult."

What gives dogs this "super-nose", and just how good is it? Stormy, a 
black-and-white schnauzer, is here in a laboratory at Florida State 
University to demonstrate. Stormy pads confidently behind his trainer, 
pushing his muzzle systematically into white Teflon boxes grouped in a 
semicircle around the room. He samples the air in each container for a few 
seconds, pausing to turn round and check the third and fourth again before 
making his choice.

These five specially designed boxes are the brainchild of the 
husband-and-wife researchers Dr Jim and Dianne Walker. After programming a 
computer to control the air flow through each box, the scientists have 
released minute quantities of a compound, which smells like bananas, into 
just one. They know exactly how much compound is in the air that Stormy is 
sampling, and when he sits down, it is a signal that he can detect it.

Scientists have strived for many years to find an accurate way to determine 
the sensitivity of a dog's nose, and the results from the Florida study are 
astonishing. When compared with results for human sensitivity to the same 
chemical, dogs' noses were found to be 10,000 to 100,000 times more 
powerful. "The dogs were detecting levels of the compound at one to two 
parts per trillion," Jim Walker beams. "The equivalent of one second in 317 
centuries."

That smell is a dominant canine sense will come as no surprise to the 
millions of people who battle to hold their dog's lead as the animal 
scrabbles in the gutter. "I've never heard of anyone having to motivate a 
dog to do an olfactory exploration," says Walker. In fact, according to top 
evolutionary biologists, it may well have been smell that brought man and 
beast together in the first place. If wolves were attracted to huge piles of 
refuse - faeces, leftover food - that appeared as humans formed permanent 
settlements more than 10,000 years ago, then less ferocious animals could 
have become used to humans nearby and even relied on them as a source of 
food.

Today, dogs are proving to be better companions that we ever could have 
imagined. More sophisticated than machines, and in tune with our every move, 
dogs can help us in ways that no technology can match - unlocking the 
secrets of our genes, protecting us from infectious diseases and even 
alerting us to cancer. But where will this extraordinary symbiotic 
relationship take us in the future? Are there still untapped uses for man's 
best friend?

"Who knows?" the Indiana epidemiologist Dr Larry Glickman muses. "When I 
asked for funding 20 years ago for dogs detecting cancer, people told me I 
was crazy. It is so important to be futuristic and to keep an open mind. Who 
knew 10 years ago that we would have an interest in bio-terrorism?"


http://www.timesonline.co.uk/article/0,,2099-2155556,00.html