Penn researchers criticize Alzheimher's drug

Friday, the FDA approved Amyvid, a potentially groundbreaking drug in Alzheimer’s research developed by Penn faculty members. However, some of their colleagues have come out against it. Spearheading the debate, College junior and biological basis of behavior major Mateen Moghbel drafted a scientific editorial against Amyvid last summer.

Specificity

When he read the clinical data for the drug, Moghbel was not that impressed.

“The compound seemed very simple,” he said, which made him skeptical that the tracer could work. In general, the more specific the tracer, the bigger and more complex the molecule should be.

In the editorial, Moghbel questioned whether the Amyvid PET scans accurately represent the amount of amyloid in people’s brains.

He explained, the tracer’s specificity — its ability to target only the amyloid plaque — is crucial because the particles are in such low concentration in the brain, even for patients with abnormally high amounts of plaque.

For instance, the editorial points out the tracer has been observed in the brain’s white matter, which typically contains very little amyloid.

University of Pittsburgh professor William Klunk, a pioneer in amyloid imaging, recognized this was an issue. “That’s always been one of the problems of these tracers.” But, he added, “We’ve dealt with this by only measuring the retention in the grey matter,” and not the white.

The editorial also argued the tracer should not have been found in the frontal lobe either. Amyloid had previously only been found in a small area in the frontal lobe, which did not correspond with Amyvid’s imaging results, Professor of Radiology Abass Alavi, one of the editorial’s co-authors, said.

Klunk contended, on the contrary, scientific literature shows the frontal lobe contains a lot of amyloid plaque.

The editorial also pointed to the issue of spatial resolution. For Moghbel, “amyloid plaques are smaller than the spatial resolution of PET scans,” so the particles might be too small to be picked up. He said, moreover, the plaque is not concentrated enough in the scanned areas, which compounds the problem.

Klunk countered that the scans are not meant to measure the plaques individually, but collectively. “You can’t hear a single cricket,” he said, “but you can hear a whole yardful of them from a yard away.”

Radiology professor Hank Kung also dismissed the criticism, which he said “could be applied to any imaging study.”

But Alavi responded, the particular problem with amyloid imaging was plaque particles are both small in volume and feebly concentrated.

This is why Bengt Långström, a Swedish professor who co-wrote the editorial and pioneered the field of amyloid imaging with Klunk, has even deeper doubts about Amyvid. “These compounds bind to junk proteins [and not amyloid plaque] in the brain that have nothing to do with Alzheimer’s Disease,” he said in an email.

Getting FDA Approval

Nevertheless the FDA, after several years of negotiation, has now decided to approve Amyvid.

For a drug to enter the market, it has to go through a three-phase FDA approval process. The first phase determines whether the drug is safe for patients, which was not an issue in the case of Amyvid, because the tracer is only taken in small doses. The next two phases determine the drug’s efficacy.

In 2008, the FDA asked Avid to prove the tracer’s efficacy. The company had to show the tracer could successfully dye amyloid plaque. And the only way to verify results was to open up the brain and compare the amount of plaque found inside to what the scans mapped.

So the team sought patients approaching the end of their lives to acquire both live PET scans and eventually, post-mortem biopsies of their brains.

“We went to hospices and nursing homes and we said, ‘we’re looking for patients who have a six-month life expectancy,’” assistant Radiology professor Daniel Skovronsky said.

Klunk raised ethical concerns, saying it is unheard of to ask for autopsy validation for a drug.. Skovronsky could have easily touched a raw nerve by conducting a study requiring participants to be on their deathbed.

But families were happy to participate. Skovronsky recalled that once the study was over, he went to see the families to thank them. “And at every single meeting, all they wanted was to thank me.”

That method propelled Amyvid ahead of all of its competitors. It allowed Skovronsky’s lab to compare Amyvid scans with brain biopsies, available within a couple of years instead of decades.

In May 2010, Avid received positive results, but the FDA still had some concerns. In March 2011, it declined to approve the drug, mainly because trial scan readings proved too inconsistent.

The FDA found there were too many discrepancies between the interpretations of Amyvid scans in trial readings conducted by three highly trained doctors.

According to the Wall Street Journal, for instance, “one patient whose autopsy yielded a positive result for Alzheimer’s got a score of four from one doctor but just a one from the other two doctors.”

Despite these issues, the FDA was satisfied when Avid devised a new training program to streamline the interpretation of the scans. The new programs include training modules and in-person training sessions. All in all, Skovronsky “found the FDA to be quite collaborative.”

But while Amyvid successfully addressed the technical concerns of the FDA, critics still have reservations about the basic assumption underlying the tracer’s purpose.

Please note All comments are eligible for publication in The Daily Pennsylvanian.