Beginning in 1958, roughly 30,000 people worldwide—mostly children—received injections of human growth hormone extracted from the pituitary glands of human cadavers to treat their short stature. The procedure was halted in 1985, when researchers found that a small percentage of recipients had received contaminated injections and were developing Creutzfeldt-Jakob disease (CJD), a fatal neurodegenerative condition caused by misfolded proteins called prions.

Now, a new study of the brains of eight deceased people who contracted CJD from such injections suggests that the injections may also have spread amyloid-β, the neuron-clogging protein that is a hallmark of Alzheimer’s disease. The study is the first evidence in humans that amyloid-β might be transmissible through medical procedures such as brain surgery—according to the researchers. Skeptics, however, note that the CJD prion itself often triggers unusual amyloid deposits; epidemiological studies, they say, find no connection between the injections and increased risk of developing Alzheimer’s disease.

Aside from CJD and the related mad cow disease, kuru is perhaps the most famous prion disease. Endemic to Papua New Guinea and now essentially eradicated, kuru is transmitted through the ritual consumption of human brain tissue at funerals. Increasingly, however, scientists are recognizing that a number of other neurodegenerative diseases, including Alzheimer’s, Huntington disease, and Parkinson’s disease, also involve aberrant proteins that act like “seeds” in the brain. They convert otherwise normal proteins into fibers that “break, form more seeds, break, and form more seeds,” says John Collinge, a neuropathologist at University College London and lead author of the new study.

Still unknown in Alzheimer’s is what role misfolded proteins such as amyloid-β and tau play in the disease, and whether they are transmissible through direct contact with or consumption of contaminated brain tissue. Although scientists have successfully induced amyloid-β transmission in rodents, these experiments relied on “massive” overexpression of the protein, says Samuel Gandy, a neuropathologist at the Icahn School of Medicine at Mount Sinai in New York City. “Exhaustive” attempts to reproduce such transmission in primates have failed, he says, leading many to doubt whether such propagation is possible.

In the current study, Collinge and colleagues examined the brain tissue of eight people, aged 36 to 51, who died of CJD roughly 30 to 40 years after they received the growth hormone injections. Four had a pattern of amyloid-β that pathologists consider moderate-to-severe in people with Alzheimer’s, though they lacked a second type of protein, tau, that is considered an important hallmark of the disease as well, the team reports online today in Nature. Two had milder, more patchy deposits; one was amyloid-free. “It’s a highly unusual finding,” Collinge says. “In that age group, you really don’t see this kind of pathology unless you have a genetic predisposition to Alzheimer’s,” which none of them did, he says.

Still, scientists have known since the 1990s that the prion protein that causes CJD can “cross-seed” amyloid-β, causing abnormal deposits to form, and vice versa, Gandy says. In such a small, observational study, it is impossible to determine whether CJD itself caused the amyloid-β seen in the deceased subjects’ brain tissue, or seeds of the protein were transmitted via injection, he argues. None of the subjects showed signs of tau, the other protein associated with Alzheimer’s disease, he and others point out.

To explore the possibility that CJD, and not amyloid-β seeds, was the culprit, Collinge and colleagues also examined the brains of 116 people with a range of prion diseases unrelated to the hormone injections. They found little to no β amyloid pathology in that group, suggesting that CJD alone was not responsible for the pathology, they say. That’s a “strong argument” in the group’s favor, says Claudio Soto, a neuroscientist at the University of Texas Health Science Center at Houston. Given that prions come in many different forms, however, it’s still possible that the β-amyloid deposits found in the brains of the injection recipients were indeed caused by CJD, whereas the controls remained plaque-free, he notes.

Next, Collinge’s team plans to test vials of archived growth hormone from the original treatments to see whether they can detect amyloid-β protein “seeds.” One obstacle, however, is that scientists don’t know precisely what constitutes such seeds on a molecular level, Collinge says.

Although provocative, the new study cannot answer the question of whether pathogenic amyloid-β “seeds” can be transmitted from person to person through contaminated surgical instruments or blood, Collinge and Soto agree. There is no epidemiological evidence to support that possibility, and any alarm over Alzheimer’s infectiousness is premature, they emphasize. Still, “that’s something that needs to be investigated,” Soto says.