Here’s a potentially interesting paper that’s recently appeared, saying that numerous brain regions in Alzheimer’s patients appear to show signs of fungal infection, as opposed to control samples. The authors claim detection through antibodies, and to have isolated fungal DNA as well, identifying several species.
The authors (from two groups in Madrid) have advanced this hypothesis before, and this paper seems to be an attempt to provide more evidence. I’m not enough of a histopathologist to say how strong that evidence is. The microscopic structures they identify as fungal cells do seem to stain fairly distinctly with antibodies, but (1) I don’t know the microanatomy of human neurons enough to say that that’s what these structures are, and (2) the antibodies themselves are polyclonal, and can potentially cross-react with other targets. (The authors, though, say that the observe no cross-reactivity with human proteins, only with other fungal species). They see evidence of fungal structures in both the neurons and in the cerebrovasculature as well.
Most of the figures in the paper are from one Alzheimer’s patient and one control, but the study did look at an additional ten patients in each category. The authors say that the same pattern repeated, with signs of fungal infection in all the AD samples and none of the controls. The structures seen in the histopathology slides vary, which could be completely believable, or could also be a versatile form of wishful thinking. And as mentioned, they were able to PCR out fungal DNA from these tissue samples, with different patients showing different mixtures of species present (and no one sample ever showing just one species). This is described as a nested PCR assay, which had to be set up carefully to detect very small amounts of fungal DNA, and again, I can go both ways on this: this sort of painstaking work may be exactly what’s needed to amplify such samples, or it could be a form of turning up the gain so high that you start reading out patterns in the noise.
There’s also a cause-and-effect question: if fungal infection does turn out to be associated with Alzheimer’s, does it mean that such infections give you Alzheimer’s, or that Alzheimer’s makes you more prone to the infections? This is a question that can be answered in the clinic, and I think it should be. There are a lot of antifungal drugs to choose from, and there are certainly a lot of Alzheimer’s patients out there. The authors reference two recent cases in the literature of cryptococcal meningitis “masquerading as Alzheimer’s disease”, and suggest that the improvement seen in these patients on treatment with antifungals might be a more widespread phenomenon. That would be quite a breakthrough, but I also wonder how many other people with AD have had systemic antifungal treatment over the years, presumably without such clinical outcomes. (On the other hand, the most common therapy for cryptococcal meningitis is amphotericin B, i.v., which is not something that you do lightly, so there may not be as many overlapping cases as you’d think(.
At any rate, this hypothesis goes into the “well worth testing” category, because the barriers to such testing aren’t high, and the potential returns are so great. It’s a long shot, but the bacterial origin of ulcers was a long shot, too, and that one was completely correct. I hope someone takes the authors up on it!