Amyloid Imaging: Now Covered By CMS, With Major Caveats

On the heels of the final decision from CMS regarding coverage for amyloid imaging in late September, Molecular Imaging Insight talked with amyloid PET imaging expert Peter Herscovitch, MD, SNMMI 2013-2014 president elect in an exclusive interview to discuss the potential impact of the new climate of coverage in the near to distant future. The CMS coverage decision now allots one amyloid PET scan per patient in approved clinical trials, but it is not yet clear how these studies will proceed. He offers some key points of conversation for stakeholders in dementia imaging.

MI: Do you have any idea how long it could take CMS to cultivate the data needed to bring amyloid imaging up for another review?

PH: I obviously don’t know what the CMS thinks. But if you look at their decision summary, what they have will take a substantial period of time. There is a lot that has to be done to get a CMS study to the point where it might actually change their opinion. They say you need a clinical research study, which they have to approve. If it is a clinical research study, it has to be done using the standard rules for doing research in humans. That means writing the protocol, having it approved by an [institutional review board (IRB)], in one or probably more than one academic center. Then they need to enroll the patients, follow them, collect the data, and analyze the data, which will hopefully show that amyloid PET scanning the brain improves patient health outcomes to the satisfaction of CMS.

In some cases, they even recommend an autopsy to be done. Alzheimer’s, as  you know, is a slowly progressive disease. It can take many years to demonstrate a change in patient health outcomes. It can easily take four to five or more years to collect this type of data.

The final (CMS) decision memo, unlike the draft memo, which was put out this summer, says the objectives of studies need to look at short-term outcomes related to changes in management as well as long-term outcomes. Those long-term outcome studies that take years and years are problematic. It is possible, for example, to have a study looking at short-term outcomes that would perhaps ask this question: Does the use of amyloid PET scanning of the brain help physicians make a better choice with regard to medication? There are medications that are indicated for patients with Alzheimer’s disease, like acetylcholinesterase inhibitors. These should not be used, for example, in people with frontotemporal dementia.

A study could be designed to show that if, in addition to the standard work-up and careful diagnostic evaluation by a dementia expert, there’s also the addition of an amyloid scan to the mix and more accurate medication choices are made, which is something that is a patient outcome benefit. That would be one way to go. Or, for example, it may be possible to show that using an amyloid scan earlier on in the work-up with patients may obviate the use down the road of addition CT and MRI scans and perhaps lumbar puncture. So you are reducing the use of other costly tests or tests that are burdensome for Alzheimer’s patients and their families.

Could you have one scan for short-term outcomes and another for long-term outcomes to see longitudinal changes in a patient?

That probably wouldn’t be necessary. As it turns out, amyloid is deposited in the brain several years before the patient develops any symptoms of cognitive decline. Once a patient has Alzheimer’s disease, there is a fairly substantial amyloid burden in the brain and it doesn’t get much worse over time. If a patient is being evaluated today for cognitive decline and memory impairment by a dementia expert and Alzheimer’s disease is a high diagnostic possibility, and then you do an amyloid scan and the patient does not have amyloid burden in the brain—the patient does not have Alzheimer’s disease. That doesn’t mean he or she won’t get worse down the road, perhaps because of another form of neurodegenerative dementia, but it won’t be Alzheimer’s disease. 

As a spokesperson for SNMMI, could you say what the potential adverse impact could be for rural healthcare and institutions that might not gain access to CED-appointed clinical trials?

Yes, t here are adverse effects. SNMMI, as well as the Alzheimer’s Association, feel there are clinical scenarios where individual patients with cognitive decline would benefit from an amyloid scan. These were clearly spelled out in the appropriate use criteria developed by a joint task force of SNMMI and the Alzheimer’s Association, which looked at the available evidence as well as expert consensus opinion and identified scenarios where patients would benefit. But if the scans are not available to patients and by and large they won’t be if they’re not covered by Medicare because they are quite expensive, then we feel the care of patients with cognitive decline and possible Alzheimer’s disease won’t be as good as it would otherwise be. Amyloid scans can increase the diagnostic confidence of dementia experts for this group of patients and inform their treatment plans. There also is a strong value for patients’ families and caregivers—essentially the value of knowing. Having greater certainty of diagnosis by dementia experts can help these families make more informed decisions about patients, including lifestyle decisions, continuing or retiring from a position and doing things you always wanted to do with your family or your life.

What kind of registries will there be—will they look like the National Oncologic PET Registry (NOPR)?

It depends on the type of trial. If it is a trial meant to show short-term outcomes related to changes in management, then it could be a NOPR-type study. It could be that a management plan is developed, the dementia expert shows the result of the scan and changes the management plan. Down the road the physician would say, yes, this patient is or is not on a certain medication or did or did not get an extra CT scan because of information from the amyloid scan. That would be a NOPR-registry type study. On the other hand, the long-term studies are much more difficult. The PET scan part of it is relative minor if patients have to be included from a wide variety of referral bases, say, urban and rural, general practice and academic centers and then they have to be followed for years to see if they benefitted from the scan. That is a very intensive clinical research protocol that involves a lot more than doing the scan and then acting on it.

CMS suggested in the final decision memo that coverage with evidence development in clinical trials would encourage more dementia imaging and treatment studies. Do you think that is true?

I think the CMS decision will actually discourage not only radiopharmaceutical companies, but the field of nuclear medicine and molecular imaging as a whole. It’s discouraging not just to nuclear medicine physicians, but also basic scientists who are trying to developing new amyloid agents.

Would the game change if a viable treatment becomes available?

Absolutely. As you probably know, there are numerous clinical trials for anti-Alzheimer’s therapies. Some of them target amyloid. It is very likely that amyloid scans will be used to confirm Alzheimer’s disease and exclude other diagnoses. It also is being used to enrich patient populations.

If such a medication were to be approved, the potential expense of the drug and any adverse side effects would prove the utility of amyloid imaging and restart the clock. As knowledge of Alzheimer’s disease accumulates, the odds of a viable treatment continue to grow.

I am very optimistic, but I don’t have any inside knowledge into which therapies are going to be the most promising or which will ultimately work.