A provocative critique of Banner's 4/4 RCT

[Julie G]

As you'll recall, Banner Alzheimer's Institute is spearheading a randomized clinical trial to preventatively treat ApoE4 homozygotes with two anti-amyloid therapies. Given Harvard's recent finding that amyloid appears to be a protective response to infection and bacterial pathogens, I actually think this critique is actually a bit on the mild side. Your thoughts?

A critique of the GENERATION RCT: What are we willing to accept for preventing Alzheimer's disease?
http://www.sciencedirect.com/science/ar ... 2216300540

On Sept 28, 2015, the GENERATION randomised clinical trial (RCT) was submitted to the ClinicalTrials.gov registry (NCT02565511). The study will test the effectiveness versus placebo of two drugs targeting amyloid (CAD106 and CNP520) on cognition, global clinical status, and underlying pathology in individuals at risk for the onset of clinical symptoms of Alzheimer's disease. The study is funded by Novartis and AMGEN, in collaboration with the Banner Alzheimer's Institute (Phoenix, AZ, USA), the US National Institute on Aging, and the Alzheimer's Association.

We are aware that the registry only provides limited information about GENERATION, but some aspects of the study methodology seem worrisome. Participants will be eligible for randomisation if they are aged 60–75 years, present with a Mini-Mental State Examination score of 24 or higher, and are homozygous for the APOE ɛ4 genotype. It is noteworthy that the presence of brain amyloid will not be considered an eligibility criterion. In other words, the investigators are planning to treat healthy individuals with anti-amyloid drugs only because these participants present with a specific genetic profile. The genotyping of the APOE ɛ4 allele (with all its limitations) is dangerously translated into an exclusive diagnostic biomarker for judging whether a person might be a candidate for a pharmaceutical intervention, regardless of any clinical manifestations (ie, cognitive impairment) or the targeted pathophysiological abnormality (ie, brain amyloid deposition).

In the medical literature, a growing number of statements request early interventions against Alzheimer's disease, with the false assumption that early diagnosis might increase the effect size of available treatments.1 and 2 This approach is arguable because predicting the boundaries between so-called normality and abnormality might simply lead to overdiagnosis and overtreatment.3

We realise that the GENERATION trialists are acting in the context of a research project without necessarily implying a direct implementation of their future findings in clinical practice. Nevertheless, too many times a research concept has been automatically translated into a clinical paradigm. For example, the actual use of the mild cognitive impairment concept in clinics might be questionable, because this concept was originally framed only for research purposes and this occurrence is still surrounded by multiple ambiguities (eg, spontaneous reversibility of this disorder to normality).4

In conclusion, we are worried that GENERATION might pave the dangerous way that assumes that everything can be done in the name of prevention, even treating individuals just on the basis of their genetic profile. We believe that this approach is ethically arguable and hope that more careful arguments will be taken into consideration during the implementation of the study.

We declare no competing interests. The opinions expressed in this Correspondence are those of the authors and do not necessarily reflect the views of their institutions.

[Verax]

JulieG, thank you for this posting. I followed the link to Sciencedirect and encountered a paywall but I saw the link to the DOI and went to the full Lancet Neurology correspondence with the references omitted above. https://doi.org/10.1016/S1474-4422(16)30054-0
"What are we willing to accept for preventing Alzheimer's disease?"
Matteo Cesari, Nicola Vanacore, Marco Canevelli
Published: June 2016 https://www.thelancet.com/journals/lane ... 0/fulltext

On the right of that published letter is a link to a reply from the Study
https://www.thelancet.com/journals/lane ... 2/fulltext
"What are we willing to accept for preventing Alzheimer's disease? – Investigators' reply"
Pierre N Tariot Pierre N Tariot, Jessica B Langbaum, Eric M Reiman on behalf of the Alzheimer's Prevention Initiative
Published: June 2016

I'm not going to copy either letter as you can read for yourself--and if you consider joining a study you definitely should read both closely. All I can say is that in my personal opinion the reply was satisfactory. Since the letter called into question clinical trials to prevent AD, we should review the whole idea in 2011 of "Alzheimer’s Prevention Initiative: A Plan to Accelerate the Evaluation of Presymptomatic Treatments"
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3343739/

[NF52]

I'm not going to copy either letter as you can read for yourself--and if you consider joining a study you definitely should read both closely.

Hi Verax,
The last part of your URL slipped through the url brackets, so here's a corrected link. I agree that it's beneficial to people considering participation in a clinical trial to read as much as possible. Thanks for contributing to the discussion!

What are we willing to accept for preventing Alzheimer's disease? – Investigators' reply

[Harrison]

Interesting critique and response. It is interesting that a positive amyloid scan is not considered an eligibility criteria, and the investigators indicate that is a plus since E4/4's that are amyloid negative are at risk. However, they are only recruiting E4/4s that are 60 and older. Based on the work of many, particularly Clifford Jack and colleagues, it is unlikely that someone in that age range would be amyloid negative but still at risk. That suggests most participants would already have amyloid and would not be treated in a "pure" prevention mode, and the people without amyloid would likely not accumulate it. With the previous failures of the Merck BACE inhibitor, it is unlikely to work in the Generation Trial. Most Alzheimer's researchers don't hold out much hope for the active immunization strategy either. Despite this pessimism, I admire what the Generation researchers are trying to accomplish by trying to treat early in an at-risk population. I think further refinement of identifying subjects early with polygenic risk scores, blood biomarkers, etc will be the best way to proceed. A treatment for people with established Alzheimer's disease is the Holy Grail, but I worry that there is a point where the damage is too extensive to reverse, at least based on how little we actually know about the course of the disease.

[Verax]

... It is interesting that a positive amyloid scan is not considered an eligibility criteria,...

Well, there are two Generation Program studies. Study 2 has among criteria the followig:
"Carrier of at least one APOE4 gene if Heterozygotes, elevated brain amyloid (as measured by CSF Abeta or amyloid PET imaging)."
https://www.clinicaltrials.gov/ct2/show/NCT03131453
So I suppose you are talking about the Generation Study 1
https://www.clinicaltrials.gov/ct2/show/NCT02565511 which does not screen for amyloid but does follow with PET scans.

However, they are only recruiting E4/4s that are 60 and older. Based on the work of many, particularly Clifford Jack and colleagues, it is unlikely that someone in that age range would be amyloid negative but still at risk. That suggests most participants would already have amyloid and would not be treated in a "pure" prevention mode, and the people without amyloid would likely not accumulate it.

The Generation studies use two drugs, one a BACE1 inhibitor to try to prevent build-up, and in another study an amyloid antibody to try to remove it.

No matter about the PET scan for these subjects, once in the study they will be followed and biomarkers refined and shared with other researchers to standardize and verify even if the drugs fail. Maybe controls evade LOAD even with E4, why? It takes a drug company to spend money on such expensive investigations (PET scans about $3,000) and it is urgent to prevent. They aren't going to spend the money unless there is a chance at a drug. Thus they got agreement from CMS to speed up the process, both for biomarkers such as PET in IDEAS, and drugs if any hope of positive effect. I can't speak for the study, but I asked about the age limits and I was told for Study 2 that I would be eligible up to my 76th birthday, as they are looking for patients asymptomatic but who might develop LOAD in the next 5 years. I don't think they are excessively particular about enrollment nor the agreed criteria wrong, but possibly they are having difficuty getting volunteers because of misunderstandings such as what you seem to have, as you just repeat the original letter and don't reply to the investigators' reply. There is a Catch-22 you point out here but maybe you should go back and read what CMS and the industry has agreed on the path to prevention (see above links as well). This is a most important issue for all of us here who pray for prevention of the coming catastrophe of AD in the world. Sorry for my hasty scribbling, but if you don't agree, how do you believe we should spend our time and money?
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3343739/

[NF52]

It is interesting that a positive amyloid scan is not considered an eligibility criteria... However, they are only recruiting E4/4s that are 60 and older. That suggests most participants would already have amyloid and would not be treated in a "pure" prevention mode, and the people without amyloid would likely not accumulate it. With the previous failures of the Merck BACE inhibitor, it is unlikely to work in the Generation Trial... Despite this pessimism, I admire what the Generation researchers are trying to accomplish by trying to treat early in an at-risk population. I think further refinement of identifying subjects early with polygenic risk scores, blood biomarkers, etc will be the best way to proceed...

Hi Harrison,
I don't know that there is either a possibility or a need for a "pure" prevention mode. I've been involved in drop-out prevention with kindergarten students based on research that early habits of parental support for school attendance have strong associations with finishing high school. Even those efforts are targeted to at-risk populations: those kindergarten students who are absent more than 5-10% of the time without a medical reason. Prevention always has to strike a balance between going far enough back to intervene before pathology is irreversible and not so far back that it's impossible to track the benefits.
Here's a somewhat unguarded comment from one of the Generation Study leaders, quoted in an August 2016 National Institute on Aging news article. As someone who has a 62% chance of being on the BACE-1 inhibitor CNP520 and 38% chance of being on placebo, I admit to an "ouch" reaction that I am in a "sweet spot" with my high genetic risk.

"We have to find that sweet spot between starting early enough before Alzheimer’s pathology develops and yet late enough so that there’s enough decline over the next few years to see if the treatments actually work,” said Eric Reiman, M.D., executive director of BAI and one of API’s leaders. “Our goal is to help individuals at the highest imminent risk in a way that helps everyone at risk for Alzheimer’s, and to do so as soon as possible,” he added.

People at genetic risk for Alzheimer's disease to test prevention drugs
https://www.nia.nih.gov/news/people-gen ... tion-drugs

For myself, I frame the key questions of the Generations 1 and 2 studies like this: Given a highly select group people at high genetic, age-related and often familial risk of a clinical diagnosis of MCI or AD in the next 5-15 years:

• 1) Will a drug that inhibits the production of brain amyloid, with reductions of up to 85%, be safe over 5-8 years (including a 3-year voluntary extension)
  2) Will those taking the drug(s) show a corresponding gain of health-years compared to placebo controls matched as baseline on cognitive tests of auditory and visual immediate and delayed memory, vocabulary, numeracy, episodic memory, visual perceptual and gross motor skills, soft sign neurological testing, daily living skills
  3) Will PET scans of amyloid, MRI scans of brain volume and microbleeds, and developing blood biomarkers substantiate that stable or reduced pathology has been achieved?
  4) Will blood tests taken every 13 weeks over the trial period allow for low-cost, effective pre-clinical biomarkers to allow research to be done at less cost, with less invasive techniques and for population-based studies to follow at-risk people in the 45-60 year old range?

I am an optimist, because the fact that other BACE-1 drugs have failed, including the Janssen drug JNJ-54861911 in the EARLY trial last week, does not mean that CNP520 is doomed or CAD106 (an immunotherapy arm of Generations 1). Failure in any medical or scientific research rarely means that the entire approach is doomed; usually it leads to improvements through analysis of critical attributes. I do appreciate your support for what the participants and researchers are trying to do.

[Kenny4/4]

As you'll recall, Banner Alzheimer's Institute is spearheading a randomized clinical trial to preventatively treat ApoE4 homozygotes with two anti-amyloid therapies. Given Harvard's recent finding that amyloid appears to be a protective response to infection and bacterial pathogens, I actually think this critique is actually a bit on the mild side. Your thoughts?

A critique of the GENERATION RCT: What are we willing to accept for preventing Alzheimer's disease?
http://www.sciencedirect.com/science/ar ... 2216300540

On Sept 28, 2015, the GENERATION randomised clinical trial (RCT) was submitted to the ClinicalTrials.gov registry (NCT02565511). The study will test the effectiveness versus placebo of two drugs targeting amyloid (CAD106 and CNP520) on cognition, global clinical status, and underlying pathology in individuals at risk for the onset of clinical symptoms of Alzheimer's disease. The study is funded by Novartis and AMGEN, in collaboration with the Banner Alzheimer's Institute (Phoenix, AZ, USA), the US National Institute on Aging, and the Alzheimer's Association.

We are aware that the registry only provides limited information about GENERATION, but some aspects of the study methodology seem worrisome. Participants will be eligible for randomisation if they are aged 60–75 years, present with a Mini-Mental State Examination score of 24 or higher, and are homozygous for the APOE ɛ4 genotype. It is noteworthy that the presence of brain amyloid will not be considered an eligibility criterion. In other words, the investigators are planning to treat healthy individuals with anti-amyloid drugs only because these participants present with a specific genetic profile. The genotyping of the APOE ɛ4 allele (with all its limitations) is dangerously translated into an exclusive diagnostic biomarker for judging whether a person might be a candidate for a pharmaceutical intervention, regardless of any clinical manifestations (ie, cognitive impairment) or the targeted pathophysiological abnormality (ie, brain amyloid deposition).

In the medical literature, a growing number of statements request early interventions against Alzheimer's disease, with the false assumption that early diagnosis might increase the effect size of available treatments.1 and 2 This approach is arguable because predicting the boundaries between so-called normality and abnormality might simply lead to overdiagnosis and overtreatment.3

We realise that the GENERATION trialists are acting in the context of a research project without necessarily implying a direct implementation of their future findings in clinical practice. Nevertheless, too many times a research concept has been automatically translated into a clinical paradigm. For example, the actual use of the mild cognitive impairment concept in clinics might be questionable, because this concept was originally framed only for research purposes and this occurrence is still surrounded by multiple ambiguities (eg, spontaneous reversibility of this disorder to normality).4

In conclusion, we are worried that GENERATION might pave the dangerous way that assumes that everything can be done in the name of prevention, even treating individuals just on the basis of their genetic profile. We believe that this approach is ethically arguable and hope that more careful arguments will be taken into consideration during the implementation of the study.

We declare no competing interests. The opinions expressed in this Correspondence are those of the authors and do not necessarily reflect the views of their institutions.

My thoughts are -please tell me more about how an anti amyloid RCT is a bad thing and how Amyloid is a good thing?
I know Amyloid has anti bacterial properties and maybe anti viral but isn’t this the stuff that clogs the neurons and such?
My theory is our amyloid gives us 4’s an immunological advantage but dooms us with Alzheimer’s in the end.

I know this: My mother who has Alzheimer’s gets much worse with Alzheimer’s when she gets a UTI. She recovers some but is always a step down in thinking after. My guess is her apoE 4 gene ramped up amyloid to help fight the infection but in the process the amyloid clogged her neurons.

Please explain clinically why anti amyloids are bad as I may be behind the times and ignorant of new info.
Thanks!

[Julie G]

My thoughts are -please tell me more about how an anti amyloid RCT is a bad thing and how Amyloid is a good thing?
I know Amyloid has anti bacterial properties and maybe anti viral but isn’t this the stuff that clogs the neurons and such?

That's a bit of an oversimplification, Kenny. First, the Banner trial isn't a "bad" thing. It's an unknown thing. Banner is putting specific anti-amyloid therapies to the test by using pre-symptomatic, at risk, folks; both 3/4s and 4/4s.

Till now, the amyloid hypothesis hasn't panned out. Reducing amyloid hasn't improved cognition (the opposite in some cases) nor reversed cognitive decline in symptomatic people. Supporters of the hypothesis have theorized that we haven't started early enough in the disease process. That's exactly what the Banner trial is now trying to do.

Amyloid appears to be a protective response to various insults. However, when too much is present it can accumulate to harmful amounts and eventually tau tangles which lead to AD. Researchers are trying to determine when, if ever, to intervene. We still have lots of unknowns. That said, we have a handful of folks from our community participating in this (and similar trials.) I applaud their courage and service towards the cause.

[Harrison]

So I suppose you are talking about the Generation Study 1
https://www.clinicaltrials.gov/ct2/show/NCT02565511 which does not screen for amyloid but does follow with PET scans.

Yes, I was referring to the generation 1 trial. They can later stratify for those who are amyloid negative at the first scan, but I just have concerns that the mix of amyloid negative and positive across two different drug with different mechanisms will be problematic. I have not studied the Generation 1 study in detail, but if a scan is not part of the enrollment process, you could imagine amyloid positive people will be assigned to the BACE inhibitor, which is unliklely to help, and amyloid negative people assigned to the active immunization, which will also not help.

I did read Tariot et al's response, and I understand where they are coming from. I don't think the Generation study is unethical or dangerous. I am just looking at this with the mindset that with the exception of aducanumab, no anti-amyloid treatment has shown promise in my opinion. Unfortunately, aducanumab likely won't be able to be administered to E4 carriers because of unacceptable rates of ARIA. If the Generation study fails, which it has a fairly highly likelihood of doing based on the general success rate of AD trials and the mechanisms tested, it will further decrease Pharma's appetite for for doing long, early prevention studies, which are clearly needed.

I do want to point out a sentence you may have missed in my comments:
"Despite this pessimism, I admire what the Generation researchers are trying to accomplish by trying to treat early in an at-risk population."

how do you believe we should spend our time and money?

As I mentioned in my first post, I think further refinement of identifying subjects early with polygenic risk scores, blood biomarkers, etc will be what is needed for successful trials. We don't have that yet, so I appreciate the Generation studies and hope we can learn from them. In addition, while I appreciate the long lead time that is needed for a study such as this, I am disappointed that an ambitious study like the Generation trial is stuck using a BACE inhibitor which is highly unlikely to have efficacy. I would be happy to be wrong.

[Verax]

CMS fast-tracked these BACE1 inhibitor trials part of the Alzheimer Prevention Initiative as they were ready to go for presymptomatic at-risks, and in conjunction with the trials they agreed to share information and standardize biomarkers such as the blood test you mention and PET scans. It was obvious that some evidence they were treating AD not something else was needed so presence of amyloid would ordinarily be required by CSF or PET, but it seems the antibodies haven't removed deposits, but maybe the inhibitors can, we have to do something before deposits. They realized this Catch-22 and attempted to work around it, maybe not to your satisfaction.

Look at how the new drug for migraine, Aimovig. also developed by Novartis/Amgen. Development of a drug to prevent LOAD will cost a couple of billion dollars. The injectable Aimovig costs $6,900 a year, less than the $10,000/y expected https://www.bloomberg.com/view/articles ... ing-answer Would you pay that much money to prevent LOAD if you knew it was 50% effective for an E4/E4? Maybe there is some generic in the same class that turns out to be cheaper, but lacks the RCT backing (like losartan instead of candesartan)?

But if the trials can give us data about the natural history of LOAD and develop better biomarkers it will possibly lead to prevention instead of cure, with lifestyle modifications emphasized, in other words more like an open-source reCODE. NIH is trying to spend a lot of money soon on the AllOfUs program (see https://www.joinallofus.org/en ). I don't think the conventional drug company development process and primary care system can deal with the current and pending massive crises of T2D, NASH, and AD. We need validated free apps and sensors for tracking, health coaches, artificial intelligence in the cloud, nutrigenetics personalized, precision medicine informed by good data in RCTs. But I don't think you or I will see it, we are too slow.