Debate over Alzheimer's tool: The PET scan's effectiveness in diagnosing the disease is unclear, some experts say. By Jamie Talan, Staff Writer
At 55, Matthew Dubiner was given a diagnosis that prompted him to quit his job and sell his house — and wait for the inevitable forgetting of all that he had learned and loved. Dubiner, now 62 and living in Great Neck, arranged to see a doctor because he had been experiencing blocks of forgetfulness. He passed the battery of cognitive tests, which Dr. John Halperin, professor and chairman of neurology at North Shore University Hospital in Manhasset, found "perplexing" given his memory complaints. Then Dubiner underwent a Positron Emission Tomography, or PET, scan.
Halperin agreed with the conclusions by Dr. David Eidelberg, who administered the test: "clear evidence of frontal lobe dysfunction," with a tentative diagnosis of either Alzheimer's or another fatal form of dementia, Pick's disease. Dubiner became a regular at weekly meetings of the Long Island Alzheimer's Foundation for middle-aged people diagnosed in the first stages. But, unlike his comrades, Dubiner's memory never worsened. "I used to say that if this is Alzheimer's, let's get on with it," said Dubiner who, upon hearing the diagnosis, retired from 33 years of teaching sixth-graders in Oceanside.
Today his doctors are saying those memory lapses may not have been a symptom of Alzheimer's — that they may have been triggered by a seizure disorder, one that can be treated with medication. They have not said that they were wrong about the initial diagnosis — simply that they are still not sure. In fact, Dubiner has been told to continue the Alzheimer's medicines prescribed seven years ago. "I wish this were a perfect science, but it isn't," Halperin said. "Short of a brain biopsy, there is nothing absolute in Alzheimer's."
The PET scan's value in diagnosing Alzheimer's is now the topic of debate. On April 5, the National Institute on Aging will co-host a meeting with the Centers for Medicare and Medicaid Services to consider whether the costly technique — each scan costs around $2,000 — should qualify for reimbursement, said Neil Buckholtz, chief of NIA's Dementias of Aging branch. "We want to get a handle on what works and what doesn't."
Some doctors say that it should remain in the research lab until more proof of its worth is in. Others insist it has proven itself a worthy diagnostic tool. "Are these scans ready for primetime? No, they are not," said Mony de Long, director of Brain Health at NYU School of Medicine. He said more tests on "large groups of people" are necessary before determining if the technology is feasible for diagnosing Alzheimer's. And even then, he said, the results would be subjective.
"You can't take any individual scan and say that this person has Alzheimer's," said Dr. Ronald Petersen, a researcher at the Mayo Clinic in Rochester, Minn. "We have an ethical and moral obligation not to cause undue worry or even a misdiagnosis. The technology is evolving, but we're not there yet."
But Dr. Gary Small, director of the UCLA Center on Brain Aging, said he has tested the scan's diagnostic ability — and found it has 95 percent accuracy in assessing who had Alzheimer's among 284 middle-aged and elderly adults with memory problems. The patients were followed for nine years.
Historically, Alzheimer's has been a disease diagnosed by exclusion. A person's ability to remember, organize, concentrate and perform thinking tasks can be impaired by a number of medical problems, including stroke, brain trauma, epilepsy and depression. A diagnosis was always prefaced with "probable," and certainty came only at autopsy.
The 1973 introduction of the CAT scan provided the first window into the living brain. Then MRIs arrived, offering a view of structural abnormalities. But part of the difficulty in deciphering such scans is that individual brains vary in performance.
Janet Walsh found out that the scans aren't flawless. At 49, Walsh, of Port Washington, has a family history of Alzheimer's and a genetic makeup that puts her at risk for late-onset disease. Almost 10 years ago, she joined a study at New York University and had her first PET and MRI scans. The results were normal. Two years ago, she had a second PET scan at North Shore's brain imaging program, and the results were not so reassuring: "Parietal and frontal glucose hypometabolism consistent with a neurodegenerative disorder such as Alzheimer's disease." "I was shocked," said Walsh. She underwent a third PET scan at NYU, which found no evidence of any neurodegenerative disease, de Long said.
Why the difference? NYU's de Long says it's a matter of subjective interpretation. North Shore's Eidelberg agrees. "We are making assumptions about what a brain is supposed to look like, and what an Alzheimer's brain looks like," Eidelberg said. "At this point, there has not been sufficient information provided [by scans] to assure a specific diagnosis in individuals with minimal or mild symptoms of the illness." Eidelberg added: "We don't do scans in a vacuum." In other words, his "impression," as it is labeled on the report, is routinely sent to the patient's doctor. In Walsh's case, it was faxed directly to her.
Walsh, who started the Long Island Alzheimer's Foundation and owns Memory Concepts, a Port Washington company that develops memory tools for Alzheimer's patients, will continue in the NYU study that is following people with a family history of the disease. Advances in understanding Alzheimer's have led to a push to find others like Dubiner and Walsh — people who may be at risk or whose short-term memory loss is worsening, typically the first sign of illness.
"There is still no test for Alzheimer's — no PET scan, no MRI, no blood test," said Dr. John Morris, a professor of neurology and director of the Alzheimer's Disease Research Center at Washington University in St. Louis. "It gets down to the physician making the diagnosis. And it isn't always right. And in the earliest stages, it's even harder to be right."
Mayo's Petersen coined the term "mild cognitive impairment," or MCI, to explain the memory and thinking impairments that he believes are an early indication. People with such symptoms "are more normal than not," he says. Volunteers in his studies undergo neuropsychological tests every year. So far, he said, 10 to 12 percent of those with mild cognitive impairment progress to Alzheimer's each year.
"People are trying to get better at diagnosing the disease earlier," said John Hardy, an Alzheimer's researcher who heads the intramural research program at the National Institute on Aging. "As you try to do that, you are likely to make the wrong diagnosis more often. That is inevitable." But he said accuracy has improved since he entered the field in 1979. "Now, good clinicians can be accurate 90 percent of the time," he said. "It's better than my mechanic can do in diagnosing a problem."
Many experts say the technology will have to improve if it's going to play an effective diagnostic role. The most recent advance involves a substance being developed at the University of Pittsburgh that can tag amyloid, the sticky substance that builds up into plaques surrounding brain cells. The most widely accepted theory of Alzheimer's is that amyloid build-up is the biological trigger for cell death.
Led by Dr. William Klunk, the Pittsburgh team tested the amyloid targeting substance, injecting it into the bloodstream of 16 people newly diagnosed with Alzheimer's and nine non-Alzheimer's volunteers. The tagging substance crosses into the brain and allows the team to identify the diseased brains. The study was published this month in the Annals of Neurology.
Such a compound could one day be used to diagnose the disease years before symptoms develop, said Dr. Norman Relkin, director of the memory disorders program at Weill Cornell Medical Center. He believes that early diagnosis will be more important as the therapies improve: The earlier the biological process is stopped, the better chance at staving off symptoms.
The national Alzheimer's Association has endorsed PET scanning only when doctors have done a complete work-up and are still confused about the diagnosis, said Marilyn Albert, a professor of neurology at Johns Hopkins University School of Medicine in Baltimore and a leading scientist in the field. At this point, "no measure alone is accurate enough," she said. As for Dubiner, he is getting on with his life. These days, he volunteers at Walsh's memory-training company in Port Washington and tutors high schoolers for the SATs. "This whole thing has been wrong. No one should go around thinking they have Alzheimer's if they don't," he said. "I've wasted a lot of precious time."
Source: News Day (www.newsday.com)