This story describes the findings from two related reports on the early detection of brain plaque as a harbinger of Alzheimer’s disease. For both studies, researchers in the Netherlands analyzed data from the brain scans of 9,500 volunteers from five countries and confirmed previous findings suggesting that the presence of amyloid, a type of protein, plays a role in the development of Alzheimer’s disease (AD). Researchers also concluded from the data that scanning for amyloid can help predict who will develop AD 20 or 30 years before any symptoms appear.
The story’s bottom-line message about this research is accurate in that it warns against widespread screening for Alzheimer’s before we have an effective treatment. The story also includes three different independent perspectives on the research, which is a strong point in the coverage and — interestingly — a source of potential concern. In several instances, these experts are given free rein to discuss their professional theories when interpreting the data. But there’s not enough acknowledgment that their comments represent informed speculation rather than evidence-based fact. The reliance on conjecture is a strong reminder how much is still to be uncovered about this disease before safe and effective treatments can advance.
These studies bring together a large amount of data on amyloid scans and amyloid markers in spinal fluid, including those without dementia and with others forms of dementia besides Alzheimer’s disease, so it imparts more information on the specificity of the tests for Alzheimer’s dementia. The identification of amyloid early in those who will likely develop AD is important because a number of drugs are being tested to target amyloid, and it’s important to select these patients to maximize the chances these drugs will work. Similarly, if a drug is found to be beneficial, it will be important to identify people with an amyloid load who will mostly like respond to the drug.
One of the quoted researchers noted that the scans are expensive and not covered by insurance. The latter means they are not ready for clinical use to any large degree, so we’ll give the story a pass. We’d add that if either of these amyloid screening methods — PET scans or cerebral spinal fluid testing — becomes adopted as a preventative practice, it could involve many millions of people and be very costly to an already overburdened public health system.
The story quantifies only two specific findings from the studies: 30 percent of the patients with mild cognitive impairment (often a precursor to Alzheimer’s) that were screened in one study had no amyloid, and 88 percent of volunteers with any type of Alzheimer’s or other dementia (frontotemporal, vascular or dementia with Lewy bodies) had amyloid in their brains. That’s not much in the way of numbers. And the headline may set unrealistic expectations with its description of amyloid “predicting” Alzheimer’s. If we’re going to talk about “predicting” Alzheimer’s, then it’s not unreasonable to expect a specific accounting of how often the test is right and how often it’s wrong — which is something the story doesn’t really get into.
As the story later explains (to its credit), the benefits at this point are more in the context of research — in identifying people who are candidates for clinical trials (which really isn’t a new contribution for this study). These methods are not meant to identify people at risk of Alzheimer’s in the general public. The authors of the journal article in dementia patients conclude it has “potential clinical utility.”
Potential harms were expressed through the quote from Mount Sinai Hospital’s Dr. Samuel Gandy, who said giving preventative medication to people at a young, healthy age — as early as age 30 — would be exposing them to possible drug side effects for decades.
The article also notes that some experts cautioned against early screening of pre-symptomatic people because no drug exists to prevent or treat Alzheimer’s. This is important to point out, since it would be very distressing for most people to learn that they have incipient Alzheimer’s and that they have no current way to treat it.
What’s notable in this article is the inclusion of numerous bits of interesting conjecture provided by the sources consulted. But this speculation is presented with a bit too much certainty, and there’s no acknowledgment that we don’t know whether amyloid does in fact cause Alzheimer’s or is merely a consequence of the disease or a minor contributor. It also wouldn’t have required much more effort or many more words to expand the horizon to other, competing, yet unproven theories — including the hypothesis that tau “tangles” are also implicated in the disease.
For example, the story says: “Experts say previous trials of anti-amyloid drugs on people with dementia failed because their brains were already too damaged or because some patients, not screened for amyloid, may not have had Alzheimer’s.” This is really going out on a limb. First, there could be a number of possible reasons — besides the participants being “too far gone” — why the anti-amyloid drugs failed in these studies (e.g. maybe amyloid isn’t the key problem). Second, without re-screening the original volunteers in the drug trials, it’s difficult to know how many patients may or may not have had Alzheimer’s.
And in regard to the study that found 88 percent of the Alzheimer’s volunteers had amyloid in their brains, one of the study authors guessed that the test could have been “wrong” in those who showed no appearance of amyloid, or they had another disease which made amyloid “less prominent.” It is a good read in that we get a peek at experts articulating their theories, but a look at the bigger picture, and few more details on the quantifiable outcomes, would have made the article stronger.
We saw no indication of fear-mongering. Indeed, there was no mention of the prevalence, incidence or societal burden of Alzheimer’s disease which some readers could find helpful for putting the size of the problem into context. However, we don’t see this as a critical omission in the article.
This is a strong point of the story. It includes three independent experts in addition to study authors and the author of an editorial about the research.
There was mention of the two different screening methods used in these studies — PET and cerebral spinal fluid test — but no discussion of which method was more precise or if they yielded similar accuracy. In addition, it would have been helpful to know the accuracy of clinical diagnosis of AD — it’s given in one of the papers, for comparison.
Dr. Gandy’s quote implied scans were available, although expensive and not covered by many insurance plans. We consider this a satisfactory recognition that the scanning technology is available now.
These are meta-analyses, useful in that they pulled together a lot of data, but the findings aren’t truly novel, although the story makes them seem so. We’ve known for many years that amyloid begins to be deposited decades before symptoms begin, that educated people have more of it, that not everyone with mild cognitive impairment goes on to develop AD, and that not everyone diagnosed with AD has amyloid. The story doesn’t acknowledge this background explicitly. One association found that was different from other recent studies was that those with subjective cognitive complaints didn’t differ in amyloid load from individuals with normal cognition. As the study authors said though, studies differ on this point and more work needs to be done.
Because the article contains comments from sources who appeared to have no involvement with the research, we are confident this article didn’t rely solely on a press release.