select search filters
roundups & rapid reactions
factsheets & briefing notes
before the headlines
Fiona fox's blog

expert reaction to study looking at further evidence for the transmission of amyloid beta pathology

Research published in Nature demonstrates a potential association between batches of human growth hormone contaminated with amyloid-β protein and the onset of Alzheimer’s disease.


An NHS Blood and Transplant spokesperson, said:

“The research paper authors say there is no suggestion Alzheimer’s disease is transmissible by blood transfusion.

“There is no evidence for a change to the age guidelines on blood donation.  Any evidence for a change would be a matter for policy making bodies.”


Dr Tibor Hortobágyi, Reader in Old Age Psychiatry, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, said:

“In 2015 Jaunmuktane and colleagues demonstrated amyloid-beta plaques and cerebral amyloid angiopathy (CAA) in some people who received cadaveric human growth hormone (c-hGH) in the 1980s and who developed the transmissible prion disease CJD.  It occurred at a young age (i.e. not typical of sporadic Alzheimer’s disease) and without any genetic predisposition to early-onset Alzheimer’s disease.  This supported the suggestion that amyloid-beta pathology could possibly be transmitted through a prion-like mechanism.  Subsequent studies provided further support by demonstrating amyloid-beta deposition following certain neurosurgical procedures (in particular dura grafting, where brain material was taken from a dead human donor) and old-style c-hGH inoculation (which is not used any more) in young patients.

“However, there has been no direct proof that the implicated c-hGH batches do contain amyloid-beta as a potential source and possible initiating factor of amyloid-beta pathology in the recipients.  The current report by Purro and colleagues has now proven this.  Furthermore, in mice genetically modified to be capable of expressing amyloid-beta pathology – one of the characteristic features of Alzheimer’s disease – CAA has been shown to occur after inoculation of amyloid-beta from contaminated c-hGH batches from the 1980s.  The study reports that the development of amyloid-beta was more significant when the mouse brains were injected with homogenized brain samples from dead Alzheimer’s disease patients, than when they were injected with the contaminated growth hormone samples, which may have been because there was higher amyloid-beta content in the diseased brains than in the contaminated c-hGH samples.  This shows a quantitative effect and that there could be a threshold below which perhaps no seeding occurs, which is reassuring.  A breed of mouse was used that is designed to naturally develop Alzheimer’s disease precursor proteins, because the normal mouse does not have the capacity to develop human Alzheimer’s disease-like pathology – it needs to be genetically modified with capability to express human amyloid precursor protein (APP) which is the prerequisite to developing Alzheimer’s disease.  Mice brains were injected directly, rather than having the material injected into their bodies – this is not comparable to medical procedures going on in people today, or to administering c-hGH in the past, but was done because the c-hGH batches were few and precious and with intracerebral injection much lower concentrations were necessary.

“This is excellent research which provides further support that amyloid-beta propagation within the brain shares similarities with prion diseases, and this avenue of research is promising to help find better treatments for this relentlessly progressive and incurable dementia.  In CJD, pathological prion protein is the pathogenic protein, whereas in Alzheimer’s disease amyloid-beta and modified (hyperphosphorylated) tau are the two hallmark pathological proteins.  The chemical structure of prions versus amyloid-beta and tau is quite different, and the two diseases, CJD and Alzheimer’s disease, are also clinically different.

“Although pathological tau protein was detected in the c-hGH batches, there is no evidence so far that it is transmissible in humans or indeed in the animals looked at in this study.  Therefore, in this study it is amyloid-beta pathology (mainly vascular, i.e. CAA) and not the full spectrum of Alzheimer’s disease, that has been detected in these mice.  In people, CAA is usually part of Alzheimer’s disease pathology but is not always, and CAA can occur without Alzheimer’s disease developing.  There is a suggestion from other work that that CAA is related to removal of amyloid-beta from the brain parenchyma and therefore could actually be a ‘good sign’.

“So, these mice did not develop Alzheimer’s disease because there was no tau and there were no characteristic Alzheimer’s disease-type neuritic plaques.  Behavioural tests regarding memory have not been performed, because that was beyond the scope of the study, which focusses on transmissibility of amyloid-beta pathology.  The key message of the paper is that when c-hGH contaminated with amyloid-beta was injected directly into their brains, amyloid-beta was transmissible and caused CAA in the mouse model, and so there is indirect evidence that c-hGH recipients who developed amyloid-beta pathology could have received seeds from these amyloid-beta-containing c-hGH batches in the 1980s.

“It is important to stress that there is no indication that Alzheimer’s disease is a contagious disease or is transmissible via blood transfusion.  Because this c-hGH has not been used since 1985, new cases of transmission via this route can be ruled out.  If people reading about this did not received the old-style c-hGH in the 1980s or earlier, they cannot acquire amyloid-beta pathology in the way demonstrated in mice in this study.  This study did not look at surgical instruments at all, so we can’t make any conclusions about that and any suggested implications about that are speculation beyond the scope of this work.”


The Royal College of Surgeons and the Society of British Neurological Surgeons, said:

“We welcome the study by Purro et al investigating the potential link between amyloid beta contamination of human cadaveric pituitary derived hormone and cerebral amyloid angiopathy.  Research which addresses issues of public safety is to be encouraged and supported.

“However, care must be taken in extrapolating data from genetically modified mice with preparations injected directly into the brain, to the situation in human patients who previously had human growth hormone injected into the blood.  Given the public interest in research of this sort and the understandable concerns that may arise, we all have an obligation to report the interpretation of research findings in a balanced and responsible fashion.  On current evidence, the risks of developing one of these degenerative diseases as a consequence of neurosurgery seems substantially smaller than the everyday risks of the procedures themselves and the diseases they are used to treat.

“The findings reported are interesting but speculative and may have little bearing to the situation in humans.  In order to corroborate the hypothesis presented in this paper, more representative studies are required that directly address the pathology in human patients.  We therefore welcome larger studies of a multidisciplinary nature to include pathologists, physicians, surgeons, epidemiologists, statisticians, instrument manufacturers, charities representing patients and other patient public involvement.  These studies will be challenging given the substantial incubation times.

“In the meantime, NICE is currently working in collaboration with NHS England and the SBNS to update its guidance to take account of the evolving understanding of these diseases and their prevention.”


Prof Roger Morris, Professor of Molecular Neurobiology, King’s College London, said:

Are the findings of this paper of public interest?

“The public interest issue to which the authors draw our attention is that of appropriate sterilization of neurosurgical instruments that are re-used for multiple patients.  Procedures were originally based on destroying DNA/RNA-based infectious mechanisms, which were then found not to be sufficient for destroying the protein-based infection of Creutzfeldt-Jacob Disease (CJD).  During the decade 1997-2006, substantial research went into developing methods that eliminated transfer of CJD via surgical instruments.  Practice has adapted accordingly.  Whatever protein-based infection associated with Alzheimer’s Disease (AD) is relatively minor and I seriously doubt if the results of this paper will change practice beyond that already adapted for CJD.  However, I am not a clinician.

Are these findings of scientific interest?

“Yes, although not as expected.

“The authors wanted to test whether the CJD-contaminated preparations of human Growth Hormone (hGH), injected from 1958-85 into children deemed to be of small stature and so lacking appropriate levels of hGH, had detectable levels of proteins thought to play causative roles in Alzheimer’s Disease (AD).  They duly found readily detectable levels of the two main amyloidogenic peptides generated from Alzheimer’s Precursor Protein, Aß40 and Aß42, and Tau protein, in CJD-contaminated batches.

“These authors have previously reported (Nature 525 (2015) 247-250) that children injected with these contaminated preparations of hGH, who died as young adults from CJD, also showed some signs of Alzheimer’s disease.  In the current paper, the authors tested whether homogenate of human Alzheimer’s brain, and then the batches of hGH contaminated with Alzheimer’s amyloidogenic proteins, induced Alzheimer’s Disease in a new, improved mouse model of Alzheimer’s.  These transgenic mice carry a humanized version of Alzheimer’s Precursor Protein with two disease-promoting mutations, leading to the mice developing a wide range of symptoms including a multitude of Aß42-amyloid aggregating around neurons, in late middle age (around 24 months old in a mouse).

“Neither the homogenates of human Alzheimer brain, nor the contaminated hGH preparations, caused any accumulation of Aß42 amyloid around neurons at 120 or 240 days post-injection (before the transgenic Alzheimer’s kicked in).  This could not be due to a failure of the inoculum to have access to neurons – the injection procedure used with CJD-infected brain homogenate transmits the disease readily and reproducibly to brain neurons, as shown repeatedly in the authors’ lab.  The Aß42 amyloid in these preparations failed, in the time allowed, to induce the Aß42 being constantly produced by the neurons to aggregate into amyloid deposits around neurons.

“However, both the homogenate of Alzheimer’s brain, and the contaminated hGH preparation, did induce amyloid deposits of Aß42 around the blood vessels surrounding, and within, the brain.  Such deposits are found in Alzheimer’s (they are called cerebral Aβ-amyloid angiopathy), and are predicted to weaken blood vessels and predispose to stroke.  Their presence could not be due to persistence within the brain of amyloid from the initial injection, which was all cleared by two days post-injection.  The authors have a prime facie case that their inocula, of homogenized Alzheimer brain and contaminated hGH, caused the cells of blood vessels to start producing amyloid plaques, a disease of the blood vessels.  It therefore adds, in a new mouse model of Alzheimer’s Disease, to a series of demonstrations that Aß42 amyloid is ‘transmissible’, in the sense that it can promote the formation of additional amyloid, and this could be a mechanism by which amyloid spreads in blood vessels.

“However, despite positively inducting amyloid production by the blood vessels, their innocula have failed to induce it in neurons.  We know that Alzheimer’s is a complex disease; this paper suggests it is more complex than we thought.  In science, we form hypotheses and test them, and congratulate ourselves when experiments confirm our hypotheses.  This is all good steady stuff.  But when an experiment shows the unexpected, then our hypothesis is inadequate, if not downright wrong.  That, in my view, is the exciting aspect of this paper.”


Prof Bart De Strooper, Director of the UK Dementia Research Institute, UCL, said:

What do we know and what’s new? 

“The new work by our colleague Professor John Collinge and others is very interesting and adds to the ongoing discussion of similarities between the pathogenic mechanisms in prion diseases and other neurodegenerative disorders, in particular Alzheimer’s (AD) and Parkinson’s disease (PD).  AD and PD are characterised by abnormally folded proteins that spread through the brain and cause dementia.

“The new publication shows that some growth hormone extracts prepared from human brain and used in the past to treat children with growth deficiency contained abnormal folded proteins related to AD, called ‘amyloid seeds’.  The work shows that injecting human growth hormone extracts containing such seeds into mice that were genetically modified to develop AD symptoms, accelerates and aggravates their amyloid pathology.  The results are reminiscent of studies showing that injection of prions induces prion-like pathology in mice.

Are there implications for public health?

“It is important to recognise that the experimental design described in the new work very much favours the chance that such amyloid transmission will be observed.  Extracts from pooled brains were directly injected in the brains of mice that were genetically modified to develop amyloid plaques.  The fact that no seeding was observed after injecting the same material into wild type (normal) mice means we should be very cautious about extrapolating the conclusions to real life conditions.  Transmissibility of amyloid is clearly very low, and will therefore occur only under exceptional conditions in human beings.  The treatment of patients with human brain extracts is obviously one such exceptional condition and was ended more than 30 years ago precisely to avoid this problem.

“Other theoretical possibilities for human to human contamination include brain surgery and blood transfusion.  It is possible that neurosurgical instruments could be contaminated with amyloid seeds – though there is no evidence of this from this study – and it is of course appropriate to continue to be vigilant in order to avoid cross contamination.  However there is absolutely no reason to postpone or decline brain surgery based on the current evidence.

“The second route of possible transmission is via blood transfusion.  This has been a concern in the field for some time, and several studies using mice that were similarly genetically ‘primed’ to develop AD-like symptoms have shown that this route of transmission is theoretically possible.  This prompted a large retrospective study in Sweden and Denmark to assess the risk of blood-based transmission AD and PD.  That study, published in 2016, examined health records of almost 1.5 million people involved in blood transfusion and found no evidence for increased rates of AD or PD in patients that received blood from donors who went on to develop those diseases.  Those results provided real-world evidence that any such risk is extremely small.  Nevertheless, it is worth monitoring these risks.”


Prof John Hardy FMedSci, Professor of Neuroscience, UCL, said:

“The work by Purro and colleagues is of both scientific and direct medical interest.  The transmission of prion diseases by direct inoculation is well established and this work, together with that of Jucker and colleagues, shows that this could be a method of transmission of amyloid beta pathology in humans.  However, one has to remember that the ways of transmission of prion disease such as CJD were by either direct innoculation of brain tissue derived extracts or by grafting dura mater (the skin that encapsulates the brain) – that is, by very direct contact with contaminated tissue.  In these circumstances about 200 young people were infected with CJD and died even though, among the tissue donors, prion disease would have been exceptionally rare.  In contrast, as far as we know, none of this cohort died of Alzheimer’s disease, though as the last paper from this group described, some showed early stages of pathology.  We might actually expect that the tissue donors may have included many with Alzheimer’s pathology among their number because Alzheimer’s disease is extremely common – so the implication is that although the process of transmission of amyloid pathology may be formally similar to that of prion diseases in these historic cases, the effective infectivity is far lower.

“It is worth reiterating that there has been no epidemiological evidence suggesting significant numbers of individuals have caught Alzheimer’s disease through this route.  Clearly, however caution is sensible.  Pituitary growth hormone injections are a thing of the past, but other neurosurgical procedures which involve persons with dementia should be carefully monitored for transmissibility issues as should blood transfusions from the elderly (60+) and from persons with a family history of early onset dementia.”


Prof Diane Hanger, Professor of Neuroscience, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, said:

“It is important to note that this study did not directly examine Alzheimer’s disease, blood transfusion risk or decontamination of surgical instruments.  In this study, Purro et al. report that amyloid beta peptide was detected in batches of human cadaveric growth hormone (hGH) that had been prepared using a specific method and stored for several decades.  Intra-cerebral injection of two of these batches of hGH into transgenic mice resulted in the development of cerebral amyloid angiopathy (CAA).  Iatrogenic Creutzfeldt-Jakob disease (iCJD) has been associated with the development of cerebral amyloid angiopathy (CAA), which is present in the majority of iCJD cases with amyloid pathology.

“This work supports the view that components of certain hGH preparations may have the potential to seed amyloid pathology and result in CAA in an experimental model.  Whether development of CAA from this material is possible after peripheral administration, as was the case for hGH, is not known.  It is also conceivable that the very prolonged storage period of the hGH batches used in this study could affect the ability of this material to seed amyloid deposition.

“Importantly, the relevance of these findings to the development or transmission of Alzheimer’s disease is unclear since tau pathology was not examined in this study and hence the results should be interpreted with caution.  In addition, a recent study found no evidence for the transmission of neurodegenerative diseases through blood transfusions.”


Dr David Reynolds, Chief Scientific Officer, Alzheimer’s Research UK, said:

“Previous research suggests that amyloid can spread through a person’s brain in a similar way to the prion protein responsible for Creutzfeldt-Jakob disease (CJD).  This new work in mice strengthens the case already proposed by this team that amyloid may have been passed between humans through historical contaminated growth hormone injections.  The new research does not, however, provide evidence that the amyloid build-up initiated by these injections causes Alzheimer’s disease itself, an important question that remains to be answered.

“Although scientifically interesting, this research focused on a small set of samples from people who had a very specific neurosurgical procedure, last carried out in the mid-1980s.  Although the findings might sound concerning, strict guidelines surrounding the sterilisation and use of surgical equipment have already been introduced since the discovery of prion protein contamination and CJD.

“Importantly, there remains no evidence of Alzheimer’s disease being transmitted through blood transfusions and further work to understand the mechanics of amyloid transmission is ongoing.”


Dr James Pickett, Head of Research, Alzheimer’s Society, said:

“This is a scientifically robust study, which shows a small number of people that had this rare growth hormone replacement procedure decades ago likely had the amyloid protein transferred to them.  We’ve known for a long time that amyloid protein is involved in Alzheimer’s disease, but it is just one component.  Although researchers found that some of the people who received this procedure had changes in their brain related to the amyloid protein, they didn’t have Alzheimer’s disease itself.  The procedure in this study was phased out over 35 years ago, and more modern approaches do not have the same risk of exposure.

“There remains absolutely no evidence that Alzheimer’s disease is contagious.  There are no examples of Alzheimer’s being transmitted from person to person via any current surgical procedures and there is good evidence to show that blood transfusions don’t increase your chances of developing Alzheimer’s disease.

“Alzheimer’s Society has been funding research for 40 years to understand the different components that cause Alzheimer’s and dementia, including amyloid protein, to help us understand this devastating condition better, and one day find a cure.”


Prof Tara Spires-Jones, UK Dementia Research Institute Programme Lead, and Deputy Director of the Centre for Discovery Brain Sciences, University of Edinburgh, said:

“There is no reason based on this paper to worry about transmission of Alzheimer’s disease between people.

“Several points argue strongly against the conclusion that Alzheimer’s disease may be accidentally transmitted by medical procedures.  First, in this study the mice had injections of growth hormone samples derived from human brain material directly into their brains, which is not done in humans.  Second, even with direct injection into the brain, mice only developed a very small amount of amyloid pathology along blood vessels, which is not Alzheimer’s disease.  Amyloid pathology is one of the types of changes seen in Alzheimer’s brains but on its own is not sufficient to cause disease.  Third, this type of growth hormone derived from post-mortem human brain has not been used since the 1980s, so this type of transmission will not occur in future.  Finally, there is no evidence from the wider literature for Alzheimer’s disease being transmissible in humans despite many years of study.”


Prof David Brown, Professor of Biochemistry, University of Bath, said:

“It has been well established that certain proteins termed amyloids can trigger their own aggregation into clumps that cannot be broken down by the body.  In the early 90s (i.e. 25 years ago) it was shown that aggregates of the protein beta-amyloid, when injected into a primate’s brain, would induce the appearance of large numbers of further beta-amyloid aggregates1.  It should be noted that this did not cause Alzheimer’s disease.

“The current manuscript from the group of John Collinge does little more than reproduce the finding from this much older work.  The main difference is that the tissue used to cause this artificial transmission of beta-amyloid aggregates is tissue from human cadavers.  The paper does not provide any evidence whatsoever that Alzheimer’s disease can be transmitted.

“The other main issue addressed by this work is the use of potentially ‘infected’ human tissues or surgical instruments.  The horrendous potential to transmit diseases with such material came to light with the discovery that prion diseases such as CJD could be transmitted quite easily from such tissues.  Bans and procedures were put in place many years ago to prevent this from occurring.  These procedures are still in place and while this publication emphasises the importance these procedures have, there is little reason to assume more stringent procedures are needed.”


‘Transmission of amyloid-β protein pathology from cadaveric pituitary growth hormone’ by Silvia A. Purro et al. was published in Nature at 16:00 UK time on Thursday 13 December 2018.


Declared interests

Dr Tibor Hortobágyi: “I have nothing to declare.”

Prof Roger Morris: “No interests to declare.”

Prof Bart De Strooper: “I am director of the Dementia Research institute.  I have also appointments at the UCL, the KULeuven and VIB, all research institutes with interest in neurodegeneration.  Finally I have received grant support from pharmaceutical companies and have been consultant for several companies and biotech.  I do not have major conflict of interests with the current study.”

Prof John Hardy: “No c.o.i.”

Prof Diane Hanger: “No interests to declare.”

Dr David Reynolds: “No conflicts of interest.”

Dr James Pickett: “No declarations of interest.”

Prof Tara Spires-Jones: “I have no conflicts of interest regarding this paper or comment.”

Prof David Brown: “I have no conflicting interests.”

None others received.

in this section

filter RoundUps by year

search by tag