Brain scans could help detect Alzheimer's early, possibly to help doctors treat the ailment before permanent damage is done; and new agents could make the scans more accessible to patients, according to a spate of research shared Monday at the SNM's annual meeting in Salt Lake City.

Based on post-mortem studies, most researchers believe the build-up of protein fragments known as beta amyloid deposits in the brain leads to, or at least is deeply linked to Alzheimer's, an eventually fatal dementia, explained Chester A. Mathis, the director of the radiology department at the University of Pittsburgh.

"There seems to be a correlation between people with cognitive deficits and the amount of plaques and neurofibrillary tangles," Mathis told DOTmed News by phone. "But are they really the cause of the death? Or just markers, like gravestones, that just mark the death?"

Whether killers or victims of circumstantial evidence, the plaque deposits might still give clues to the early discovery, and thus treatment, of the disease, thanks in part to one of Mathis' discoveries.

Mathis, who received this year's Aebersold Award for advances in basic nuclear medicine science at SNM on Sunday, helped discover the 11C Pittsburgh Compound-B, or PiB.

PiB is a PET tracer that binds to beta amyloid deposits, and might help doctors detect the disease before irreversible dementia sets in.

In one study presented at the SNM meeting Monday, led by Dr. Christopher Rowe of Austin Hospital in Melbourne, Australia, researchers followed 188 patients-- some with Alzheimer's, some with mild cognitive impairment, and some healthy controls -- between around 20 months to three years to see if PET brain scans picking up plaque deposits could predict later developments of dementia.

Rowe and his team found those with a positive scan for beta amyloid build-up had a 66 percent chance of developing dementia. Only one person with a negative scan went on to develop dementia.

"She was an 85-year-old woman who was frail," Rowe told DOTmed News, and the dementia might not have been Alzheimer's related. "If you have mild cognitive impairment and a positive scan, you have a 13 times greater risk to develop to full-blown dementia in two years," he said.

However, while high PiB binding in those with cognitive impairment was associated with increased risk for Alzheimer's, the change in beta amyloid load wasn't associated with cognitive decline.

Also, showing high PiB binding is rather common as we age, so a positive scan doesn't necessarily mean you should be worried. In the study, only 14 percent of healthy subjects with high PiB levels developed mild cognitive impairment or Alzheimer's after a year and a half, and around 21 percent after three years.

"A positive scan with someone with completely normal memory, there's no indication to panic," Rowe said. "But if your memory is declining, and you've got a positive scan, there could be a serious problem."

If you live long enough, you'll get beta amyloid deposits in your brain, Rowe said. Nearly one-fifth of patients in the study between the ages of 60 and 70 had positive scans. For those between 70 and 80, nearly half had a positive scan. And nearly five out of six patients over 85 in the study tested positive.

"The ones that showed progression over a couple of years already had some subtle memory impairment that was still within the normal range, but starting to get within the [worse] end of the normal range," he said.

Rowe said this progression mirrors what is known about the prevalence of Alzheimer's disease. At 85, about one-quarter of the population is believed to have Alzheimer's, he said, compared to only one-tenth at age 75.

"Every five years the prevalence doubles," he said. "So basically once you're in your 90s, almost 50 percent of the population has Alzheimer's disease. [Our scans are] following the same trajectory, but 10 to 15 years earlier."

Currently, Rowe is taking part in a massive, decade-long follow-up study dubbed AIBL, the Australian Imaging Biomarkers and Lifestyle study. It has 1,000 participants, who are followed every 18 months, with 215 patients getting a repeat PET or MRI scan at 18 month follow ups.

"If further long-term follow up confirms what we believe, that beta amyloid is a very early predictor of Alzheimer's disease, it provides a means of detecting patients and administering therapies" before the disease strikes, Rowe said.

"A big caveat," he added: "once we have therapies."

Currently, while some treatments can help alleviate some of the symptoms of Alzheimer's, no approved therapy can cure or delay the progression of the disease.

"There have been several dozen [trials] that have failed," Mathis said.

One trial that Mathis worked on, published in April in Lancet Neurology, involved a drug that could decrease amyloid presence in the brain but without improving symptoms. But only 28 subjects were used, and a larger, follow-up "powered" study will be performed to see if it works.

Mathis suspects many of these trials fail because they're begun too late.

"All of those trials have been conducted in Alzheimer's patients in moderate to severe disease. It's like treating cancer after it's metastasized, or treating heart disease after it's advanced so far there's no recovery. The brain cells are dead -- you're not going to bring them back....That's why imaging is so important," he added, "to identify those people who potentially could benefit before they have the disease."

NEW MARKERS

But the problem with PiB, which has a 20 minute half-life, is that a medical center needs a multi-million dollar cyclotron and trained chemists on staff to use it, Mathis said. He estimates there are only 60 to 70 centers around the world where it could be used.

But companies are racing to find alternatives. In a separate study presented at SNM, researchers led by Osama Sabri from the University of Leipzig in Germany carried out tests with florbetaben, a fluorine-based PET imaging agent made by Bayer HealthCare. In the phase 2 trial, the researchers found it to be quite successful in picking out patients with Alzheimer's, distinguishing between ailing patients and healthy controls with 80 percent sensitivity, 90 specificity, according to the researcher's report.

Flourine-based compounds, like florbetaben, have a nearly two-hour half-life, and are accessible to about 90 percent of the population, Mathis said. In addition to florbetaben, Avid Radiopharmaceuticals is studying florbetapir, and GE Healthcare has its own compound, GE-067 (which Mathis is helping to develop).

But does it work as well as PiB? There are nearly 200 papers about PiB, but only a handful about the fluorine-based agents, Mathis said. "Are they good enough? I don't have an answer to that. Nobody does," he said.

Right now, companies that make the drugs are running phase 3 trials, comparing brain scans using the agents with autopsy analyses. Trials should wrap up within six months to two years, Mathis said.