A recent article in the New Yorker, “Did the World Just Gain Hundreds of Millions of Alzheimer’s Patients?” is an extensive look into several aspects of recent Alzheimer’s disease research, covering questions such as arguments among scientists as to the definition of a disease, whether a condition which may not yet be a disease should be treated, and how good the evidence needs to be to justify that treatment. And more. I am not going to review all the details covered in that article because it is very long and complicated and if you want it, you can read it. I will summarize some parts of it in order to address the issues that I find of greatest concern.
Most people today know what Alzheimer’s disease (AD) is, at least that it is the most common cause of dementia, occurring most often in old people. The article looks at the history of the disease, and notes that when Alois Alzheimer described it over a century ago he called it pre-senile dementia. I knew this because I am a physician and am old and when I was in medical school (only half a century ago), it still was. Friends of my family had Alzheimer’s, and it was very sad because they were relatively young, even in their 40s. Senile dementia, occurring when you were old, was then considered a normal variation – old people lost their memories and eventually their minds. Eventually, it was figured out that it was the same condition and, like many other diseases that are most common in old people, some people got it younger.
Alzheimer and later scientists discovered several abnormal changes in the brains of people who died of the disease. The most important for understanding this article is the presence of deposits of a substance called amyloid. It is not important for most of us to know what it is, just that it is abnormal and occupies space in the brain, presumably where normally functioning brain tissue used to be, so was assumed to be the cause of many of the changes in Alzheimer’s.
Much more recently, some drug companies have discovered substances that can reduce the amount of amyloid in the brain, and this, along with blood tests that might identify AD disease without expensive imaging (or autopsy!) have excited a lot of people, scientists included. The logic is that the blood test can tell you if you have AD (or may be likely to get AD; this is one of the important issues) and then the drug can shrink the amount of amyloid in your brain and make you, at least a little, better. This would be important, because the existing drugs approved for treatment of AD do not make people very much better most of the time.
But there are some concerns, and these are the source of disagreements among AD experts. Among them: is amyloid the cause of AD or is it a result of AD? Would reducing it would make people clinically better (that is, have less dementia)? There are, after all, other pathological abnormalities. And some people with AD don’t have that much amyloid, and some people with amyloid don’t have AD. Indeed, what is disease? What should be treated? Those who advocate for treatment say it is great that the drugs can be used before people are very symptomatic, to prevent AD. Isn’t that the best thing, prevention? Think about treating high cholesterol to prevent heart disease.
Of course, there is the question of whether the drugs actually work, but this is also depends on your definition of “work”. Does “work” mean that they reduce the amount of amyloid, or does it mean that the mental impairment of the person is less, that is, they are less demented (this is what is meant by clinical effect)? Or, maybe, does it not so much improve the actual clinical situation, but slow the progression of the disease? This would also be good, but is harder to measure given that the disease progresses at different rates in different people and that clinical illness is not clearly correlated with the amount of amyloid in the brain. The difference between using drugs to decrease amyloid in AD and using statins to reduce cholesterol is that there is strong evidence both that statins reduce cholesterol and that this reduction leads to a significant reduction in heart disease. Not true for amyloid drugs.
Then there is the fact that there would be many billions of dollars to be made by the drug companies treating those hundred million (or so) more people who do not have clinical AD but have a positive blood test, evidence of the existence of amyloid, and possibly a genetic predisposition (the article spends a lot of time looking at a particular person who has the risk genes and some amyloid but despite being in his late 70s has no clinical signs of AD). And there is the other fact that a very high proportion of those scientists advocating for treatment are employed by or have received large research grants from the companies that make these drugs, while few of those who oppose it have received such funding. And, finally, there is the risk of “side effects” (that is, effects, but not the ones we want) that can be serious – like brain hemorrhage, no minor thing!
The question of “what is a disease?” -- as opposed to a laboratory abnormality -- is, to me, critical. Sometimes treating a laboratory value is convenient and effective -- when there is clear evidence linking it to improved outcomes in people’s clinical condition, as with cholesterol. Sometimes it is equivocal, benefiting some people but not others. In these cases, it is important to identify who is likely to benefit. And sometimes it is of no benefit, or even causes significant net harm. “What is a disease?” has important social as well as physical implications. In the 1980s, a blood test that could identify people with HIV was developed years before there was any effective treatment. So, the result of testing positive was that you could lose your job, housing, insurance, but could not be treated.
I have a lot of questions about the slippery slope of defining, diagnosing, and treating diseases. One touchstone for me is “does treatment create a significant net benefit in the actual (clinical) condition of the person?” Just treating lab values, or imaging studies, or even possible unquantified risk, is not, to me a good thing. There is – and will be, in AD and other conditions -- disagreement about the definitions and the cutoff values for diagnosis. (Note, for example, that there is now a condition called “pre-diabetes” defined by blood sugars higher than “normal” but not high enough to be diabetes. Before this term was invented, it used to be called “not diabetes”!)
There are two things that I am sure of, though. One is that the likelihood of a big corporation making lots of money should not be the criterion for treating people. The other, sadly, is that it often is.




