Giulio Maria Pasinetti led this live discussion on 20 February 2002. Readers are invited to submit additional comments by using our Comments form at the bottom of the page.

See Related News Story: NO-Releasing NSAID Reduces β-Amyloid, Activates Microglia
See Recent Review


Live discussion held on 20 February 2002.

Participants: Paul Aisen, Douglas Galasko, Todd Golde, Amy Graves, Sue Griffin, Eddie Koo Alexei Koudinov, Kerry O'Banion, Giulio Pasinetti, Huntington Potter, Joe Rogers, Sarah Sagi, Yong Shen, Gabrielle Strobel, Kiminobu Sugaya.

Note: The transcript has been edited for clarity and accuracy.

Gabrielle Strobel: I am managing editor of the Alzforum. Hello everyone, and welcome to today's chat! Giulio, you have posted a summary. Maybe you could start by restating the subject of our discussion today? And remind us of the key most recent developments that have made NSAIDs more attractive again after disappointing results of earlier treatment studies.

Giulio Pasinetti: Our main idea is that one of the reasons the anti-inflammatory treatment studies failed lies in the clinical progression of Alzheimer's disease. The discrepancy may be due to the fact that the bulk of epidemiological studies of NSAIDs have examined the likely incidence of AD prior to the onset of clinical symptoms. In therapeutic studies, NSAIDs are administered to patients already presenting with illness severe enough to exceed the clinical detection threshold. Thus it may be that NSAID as a therapy administered following the onset of AD may not be optimally effective.

Todd Golde: I agree. Same issue may be at work with statins.

Kiminobu Sugaya: I guess that is true. Once neurons are lost, NSAIDs itself cannot do anything.

Gabrielle Strobel: There is evidence that neuronal death in the entorhinal cortex is extensive before people are diagnosed, and the majority of neurons in the most affected areas are probably gone by the time someone has even "mild" AD. Short of a cheap early biomarker (which we don't yet have), what could actually be done now to make sure future treatment trials focus on the earliest possible cases?

Huntington Potter: Hi Giulio, I agree with almost all that you wrote. The main thing is to identify why inflammation contributes. Our data say that inflammatory proteins such as apoE and ACT are essential for amyloid formation, and that amyloid formation is essential for cognitive decline in mice.

Douglas Galasko: If inflammation, COX activation, or other pathways are active as part of AD pathology, then why shouldn't suppressing inflammation be beneficial for the surviving neuron pool?

Sue Griffin: Doug, if IL-1 does upregulate AβPP, the stress protein, then IL-1 suppression should be good for healthy neurons exposed to IL-1 and other inappropriately elevated proteins.

Huntington Potter: Suppressing inflammation should prevent the disease if started long before clinical symptoms.

Kerry O'Banion: This will clearly be an issue for any therapy: vaccination, Aβ blockers, etc. Some clinical studies with mixed NSAIDs had modest positive results. Would anyone like to comment on these?

Huntington Potter: You have a good point. Mixed NSAIDS may work better because of additive effects or because we really don't know the important target. It is interesting that Sue Griffin and colleagues have shown that IL-1 polymorphisms are great risk factors (Griffin et al. 2000).

Todd Golde: But the only therapeutic study with any positive result was the Arizona indomethacin study (Rogers et al., 1993). This was not a good study. Other therapeutic studies have been disappointing. Interestingly, indomethacin lowers Aβ42.

Douglas Galasko: On the other hand, no clinical trial with mixed NSAIDs in symptomatic AD has included a high dose (may be necessary to decrease amyloid??). It is possible that higher doses may be needed for treating symptomatic disease and lower doses could be effective in prevention.

Giulio Pasinetti: Secondly, there is significant interest in studying the role of NSAIDs on beta amyloid metabolism. This may not impinge on neuronal death immediately, but may influence the clinical dementia in the early stages of the disease. Thus inflammation as a function of the progression of the disease is not necessarily static.

Kiminobu Sugaya: Some of NSAIDs' effects on Aβ may not go through COX inhibition.

Huntington Potter: It does appear that COX may not be the most important target.

Giulio Pasinetti: It is certainly true that some NSAIDs may not require COX at high doses in in-vitro situations.

Yong Shen: I agree with Giulio because we found some NSAIDs at high doses can inhibit certain cytokines but not through COX-2.

Todd Golde: I think we must all keep in mind that all NSAIDs (except newer COX 2 derivatives) are dirty drugs with lots of targets....

Sue Griffin: Todd, why would one think something so mild as NSAIDs would have a noticeable effect on a person who already "has" AD?

Todd Golde: Sue, why do you call an NSAID mild?

Sue Griffin: Todd, mild at the low dose that may be effective, see the Netherlands epidemiological study, (see ARF news story) and mild relative to vaccines.

Alexei Koudinov: Is there any idea on how to explain the inflammation associated with the immunization trial in 4 AD patients in France?

Sarah Sagi: Could anyone comment on the reduced plaque load seen in mice treated after Aβ deposition begins?

Huntington Potter: I think there are a number of reasons for the inflammation. One is that the antibodies may have attacked the vascular amyloid and opened up the BBB.

Huntington Potter: Does anyone know whether the clinical trial inflammation was due to infection?

Gabrielle Strobel: Hunt, we will discuss this question, in as much detail as possible Absent official data from Elan/AHP, during our March 5 discussion. Come join us! (See vaccine live chat).

Yong Shen: I wonder if microglial activation is good or bad in the early stage of AD. If we inhibit in early stages maybe we also face the issue of reduction of amyloid clearance.

Kiminobu Sugaya: I think microglia activation is bad, if they produce nitric oxide. NO may damage mitochondrial function and eventually kill the neuron.

Sue Griffin: On the other hand, microglial activation could be good on a limited scale, e.g. for astrocyte activation and subsequent release of growth, repair, and maintenance factors.

Yong Shen: Yes, neurons become very vulnerable to infection and to Aβ too, I believe.

Huntington Potter: Would there be any bad side effects in the brain from suppressing inflammation for many years?

Sue Griffin: Hunt, I don't think so since it appears that in "normals" the level of activation is very low until old age.

Giulio Pasinetti: Hunt, we do not anticipate there to be any such consequences, however, tolerable drugs without peripheral side-effects must be used.

Douglas Galasko: Hunt, we don't know. However, NSAIDs that cross the BBB have been given to arthritis patients chronically, without ill effect.

Paul Aisen: Hunt, I am not sure whether it is bad for the brain, but it is certainly dangerous to suppress COX in the periphery for many years.

Todd Golde: Back to NSAIDs. It would really help if the epidemiologists could provide detailed data on what drugs people took. We have completed a survey of all FDA-approved NSAIDs, both in cultured cells and in Tg2576 mice. Only indomethacin, ibuprofen, flurbiprofen, meclofenamic acid, and fenoprofen lower Aβ42. (See ARF news story.) If naproxen or other non-Aβ lowering NSAIDs show efficacy, then their efficacy is much more likely to be due to anti-inflammatory properties.

Gabrielle Strobel: Todd, why do you think the recent Rotterdam study on NSAID and AD protection found no difference between the different NSAIDs? It did list the different drugs people took and the doses. Naproxen did lower AD risk. (See ARF news story.)

Todd Golde: That study did tell us what drugs people took but did not break down the data into efficacy based on individual drugs. The only group which showed protection was the group taking drug for over two years. There is no data as to what drugs they took.

Paul Aisen: The epidemiological studies lack the power to distinguish among NSAIDs.

Amy Graves: Todd, the numbers in the Rotterdam study may have been too small to break them down by type of NSAID.

Sue Griffin: Todd, what about John Breitner's data on aspirin--that looked pretty good compared to ibuprofen.

Todd Golde: Aspirin has not shown efficacy in the Rotterdam study, nor was the data significant in the BLS series, (see Stewart et al., 1997.)

Huntington Potter: Todd, I'm not sure that suppressing Aβ is the important thing NSAIDs are doing. We find that Aβ is harmless in the absence of apoE. There is no amyloid and no cognitive decline. The key is to suppress the inflammation-induced production of things like apoE or ACT

Todd Golde: Hunt, I make no claims that the NSAID effect is due to Aβ42 lowering, just that data is out there to address this if looked at properly.

Douglas Galasko: Hunt, there is always apoE in the human brain, so NSAIDs could be useful if they lower Aβ.

Huntington Potter: True, but apoE goes up during inflammation and neuronal damage.

Alexei Koudinov: …and it is likely activity-dependent during normal conditions, (see Cedazo-Mingues et al. 2001).

Todd Golde: Hunt, I agree, I will also place a side bet that statins work by lowering apoE, not Aβ. I'll collect in ten years

Huntington Potter: Nonetheless, it is interesting that Aβ42 is affected. We know that AβPP translation is upregulated by IL-1 (Rogers et al.). It would make sense that the processing might be affected by such upregulation and that NSAIDs might change that.

Kerry O'Banion: Of course, the naproxen question is a big one. To the best of my knowledge, the "mixed inhibitor" for all current trials (treatment and preventative) is naproxen, primarily because it of it's dosing convenience. So how would negative results be viewed here?

Todd Golde: Kerry, if naproxen and vioxx don't work I'd have to put my money on lowering Aβ42.

Kerry O'Banion: Todd, you are not the only one putting your money there!

Todd Golde: Hunt, unpublished data from my lab suggest that the effect is independent of NSAIDs on Aβ42 is not due to transcriptional/translational events.

Huntington Potter: Todd, no but I completely agree and have been trying to convince others. If cholesterol is reduced, there is less need for apoE.

Kerry O'Banion: Lots of interesting food for thought. NSAIDs are dirty drugs; as they are "cleaned up" (e.g. COX-2 inhibitors) are we going to lose effectiveness? Thinking about targets, the main ones I consider are: 1) inflammation, which may depend on both COX-1 (microglial) and COX-2 activities; 2) neuronal expression of COX-2, which in certain models seems to make neurons more vulnerable; and 3) the new information on Aβ processing influences. Any others come to people's mind?

Giulio Pasinetti: We recently submitted a paper suggesting that the overexpression in the brain of COX-2 transgenic mice potentiates amyloid plaque deposition, and altered metabolism of AβPP.

Sue Griffin: Giulio, good for you. This really supports the idea that everything that drives increases in AβPP would be potentially detrimental to neurons in the surround. These include head injury, etc., and in all cases evidence of microglial activation with all the accompanying goodies.

Kerry O'Banion: Agreed, Giulio's work sounds very interesting.

Giulio Pasinetti: We believe that a putative mechanism by which COX-2 influences features of AD-type neuropathology is through affecting AβPP metabolism. However, although our data demonstrate that COX-2 promotes amyloidosis, we believe that a non-selective COX inhibitor may be most effective. Opinions?

Caregiver: My mother is 78 and has early Alzheimer's. She is currently on 375mg of naproxen per day plus 10mg aricept, 100mg Vitamin E, 400mg of folic acid, and the B-50 vitamin complex. Is she taking enough naproxen to ensure an optimal result?

Douglas Galasko: Giulio, could you comment on the negative clinical trials of selective COX-2 inhibitors in AD, which have been reported at meetings, though are not yet published. Is the problem that they started treatment too late? Is prevention the only way to go? Is there evidence for COX activation or other inflammatory markers being very prominent in the brain at the stage of MCI?

Sue Griffin: Since we're gambling, I think something that stops the process early is going to be best-too late when those nasty plaques and tangles are there.

Douglas Galasko: Sue, but those nasty plaques and tangles trigger inflammation. Is there much inflammation before they are present?

Sue Griffin: Doug, yes. Look at just after a head injury. Look at epilepsy, Down's, even just getting pretty old.

Eddie Koo: I agree with Sue. Treating Aβ or whatever after the neurons and synapses are fried just doesn't seem like it will be very effective,

Todd Golde: Sue, Martin Rosser and Nick Fox's work on serial MRIs makes a good case that, prior to symptoms, the brain is atrophying in patients with FAD mutants (Fox et al., 2001).

Paul Aisen: The disease progresses for many years after diagnosis. It should be possible to influence the rate of progression.

Giulio Pasinetti: We believe that not only is late intervention a problem, but also the lack of proper characterization of disease state. Our goal nonetheless is to intervene as early in the process as possible. The fact that COX-2 is elevated does not necessarily mean specific inhibitors are the way to go.

Gabrielle Strobel: To get at the earliest cases in treatment trials, could finding and enrolling at-risk people be coordinated with prospective imaging studies that follow questionable AD cases to see who later converts to AD and what brain areas can predict conversion?

Kerry O'Banion: Giulio, I agree. If microglial activation is at the heart of inflammation in AD (they are the first reactors to Aβ and major producers of IL-1), then current data suggests a mixed inhibitor with COX-1 activity would be necessary.

Sue Griffin: Sorry, I just have to: I did say this in 1989 (Griffin et al. 1989.)

Agnieszka: Our behavioral data (preliminary) in Tg2576 mice also indicate a better effect of ibuprofen if treatment begins before the onset of behavioral deficits

Sue Griffin: Kim, I like that.

Todd Golde: In general, how do we know that microglial activation is bad? Tony Wyss-Coray's studies, presented at the Neuroscience meeting last fall, show that complement activation may play a role in normal Aβ clearance.

Sue Griffin: Todd, I really don't know what to make of Tony's data except I don't doubt it. Maybe it's like in a wholesale or last-ditch effort, lots of potential actions come into play.

Eddie Koo: It seems that there is the sentiment that "inflammation" in AD is bad. I just don't think the data is there for this conclusion.

Huntington Potter: Maybe not, but NSAIDS seem to help in both mice and men.

Todd Golde: Agree with Eddie.

Giulio Pasinetti: Eddie, certain aspects of inflammation are not necessarily detrimental and perhaps even neuroprotective (e.g. C5a anaphylotoxin protects through MAP kinase pathway.)

Eddie Koo: I seem to recall that at the Neuroscience meeting, someone showed that giving lipopolysaccharide (LPS) to activate the inflammatory response also reduced amyloid pathology. My sense is that some of the inflammatory response in brain is just an inefficient way to get rid of the amyloid. So deciding on which is good and which is bad will be important but difficult. [Editor: In San Diego, Dave Morgan's group showed that one-time hippocampal injection of LPS reduced Aβ load in APP/PS1-transgenic mice (though not fibrillar amyloid), suggesting microglial inflammation may be beneficial. However, Don Price and VE Koliatsos' group showed that systemic LPS-induced inflammation caused accumulation of APP in wildtype and transgenic mice and suggest it is detrimental. Find Abstracts at]

Sue Griffin: Eddie, inefficient is the key word. In a last-ditch effort no holds are barred and even less effective strategies maybe brought into play.

Giulio Pasinetti: Can we collectively reconcile a goal for future clinical trials, to address as many of the most pertinent issues as possible.

Sue Griffin: Giulio, I think we need a way to group the patients. If the IL-1 genetics holds up, then we could select groups and try early preventive strategies discussed here.

Todd Golde: Hunt, but we do not know why? Obviously we need to study NSAIDs that are and are not Aβ42-lowering agents, and Aβ42-lowering agents that are not NSAIDs.

Huntington Potter: It seems the conclusion that treatment must begin early for anti-inflammation to work is now the conventional wisdom Yes/No?

Agnieszka: Hunt, I agree but don't know if it's conventional wisdom.

Todd Golde: Hunt, I think this is a general statement: "Inflammation and Aβ-prevent both and things are likely to work better."

Amy Graves: Just as there are many years post-diagnosis to monitor progression, it is likely from epidemiological data that the pathology goes on for many years before onset. It seems the epi studies show that NSAIDs reduce the incidence because we're catching people before the earliest symptoms when the neurons can still be repaired. Treating them later is likely too late.

Douglas Galasko: Before going into trials, which are extremely expensive, some attempts to define doses, mechanisms of action/treatment targets and possibly biomarkers would be helpful. A trial at the stage of Mild Cognitive Impairment is much cheaper than a primary prevention trial. I still wouldn't like to discard the idea of trying to slow progression in symptomatic AD.

Sue Griffin: Right, Doug.

Todd Golde: Agree with Doug.

Giulio Pasinetti: All: Do you think that treating elderly population prophylactically with tolerable NSAIDs (as in the Hopkins ADAPT Study) is the way to go?

Eddie Koo: I would put my money on prevention.

Amy Graves: Eddie, this is essentially what the epidemiological studies are.

Gabrielle Strobel: Joe Rogers called in. He could not connect but says prevention trials are key, not treatment.

Giulio Pasinetti: We agree, Joe.

Kerry O'Banion: Agree, prevention is where all therapeutics for neurodegeneration would work best.

Alexei Koudinov: Thanks for Dr. Rogers note.

Sue Griffin: Let's work on Joe's comment.

Todd Golde: The issue with prevention (as Giulio pointed out) is $$$$$$$$$$$ and time. Even the estrogen study has been underpowered. We need better predictors of who is going to get disease.

Huntington Potter: I agree. Prevention is key, which brings up the need to have a very early diagnostic. Indeed, if we had one it would look like it failed because so many "normals" would be positive because they would not yet show symptoms.

Giulio Pasinetti: Todd and Hunt, biomarkers are evolving with new technologies, e.g. microarray and proteomics.

Alexei Koudinov: We already spent 15 years on the amyloid hypothesis.

Sue Griffin: Bravo, Alexei!

Huntington Potter: We now completely believe in the amyloid hypothesis based on behavior in mice and indeed that it is the process or product (ADDLs?) of amyloid formation that causes the cognitive deficit.

Todd Golde: To the DOUBTERS: people spent years trying to disprove the relationship of cholesterol to atherosclerosis.

Sue Griffin: Good point Todd. Of course tombstones can be like that since they are what we see most in a graveyard.

Alexei Koudinov: The relation of cholesterol and atherosclerosis was hardly proved, as you can pick from last year's essay "The soft science of dietary fat," in Science 2001 Mar 30;291(5513):2536-45. See also Kai Simons' review "How Cells Handle Cholesterol" and a recent letter to the editor on studies of dietary fat and heart disease Science 2002 Feb 22;295(5559):1464-5.

Eddie Koo: Alexei, almost everything we do is correlative and not proof.

Giulio Pasinetti: Todd, perhaps the NSAIDs are truly a wonder drug re:cholesterol, yet another mode for NSAIDs' 1. antiinflammatory, 2. Aβ processing and potentially more, unknown mechanisms?

Todd Golde: Giulio, I agree it may be that both actions are needed and that they are synergistic.

Giulio Pasinetti: Please keep in mind that this is only a taste of what we can discuss more intensely in the satellite symposium to be held at the International Alzheimer Congress in Stockholm, see conference calendar for contact info.

Sue Griffin: Todd, what did you think about my comment for predictors and grouping of patients?

Todd Golde: Sue, sorry I missed it.

Sue Griffin: Todd, I said maybe genetic markers like IL-1 will maybe help grouping patients for trials.

Todd Golde: Sue, I agree completely with the need for multiple predictive tests. This is what is done in atherosclerosis, genetics, biomarkers, family history, lifestyle....

Giulio Pasinetti: Sue, definitely grouping is very good, however, this would be a monumental effort to undertake, especially in MCI studies.

Sue Griffin: Giulio, it really is not too difficult now that genotyping is just about routine. Of course lots of IRB questions and fuss over it.

Huntington Potter: I agree. Multiple markers will be essential. We may even be able at a very early age to give a risk assessment. Then only at-risk or paranoid people need to start therapy for what could be decades.

Douglas Galasko: Hunt, a clinical trial can't last decades. AD prevention trials are extremely expensive, and typically need 3000+ subjects. Do funding agencies and big Pharma have the stomach for NSAIDs in this setting? At present the best predictor of when someone will convert to AD is age, and stratifying for age saves a little money in prevention studies. Even if we improve the odds of who is at risk, we are unlikely to be able to shorten the duration of follow-up in these trials (typically 3-4 years).

Sue Griffin: Right Hunt. And yes, Doug, I think Pharma has a hard time if generics will do pretty good-say lower the risk to 0.2.

Huntington Potter: One group of humans that always gets AD on a predictable schedule are the Down syndrome individuals. Are any of the longer-term NSAIDs trials going to focus on this unique set of patients?

Paul Aisen: Hunt, the first large-scale trial in Down's is just starting-looking at vitamin E.

Huntington Potter: Vitamin E is interesting, but I would prefer a NSAID or melatonin trial.

Sue Griffin: Hunt, it would be great and we are now going to do a little genotyping (not the usual suspects) on some to try to encourage that idea.

Huntington Potter: Great.

Todd Golde: Basically, development of AD therapeutics is going to be difficult if efficacy can't be proven in some sort of therapeutic setting.

Douglas Galasko: So, I still think there is much to be gained from studying MCI or mild AD.

Giulio Pasinetti: Our goal is to combine biomarkers with properly characterized MCI cases.

Giulio Pasinetti: Guys, are you aware of scores such as sum of boxes that may aid in sub-characterizing MCI, and focusing target populations?

Douglas Galasko: Giulio, MCI is a much better deal than primary prevention. An interesting aside is that there is no completed/published MCI clinical trial in AD, so we don't know which methods of assessment are most efficient for trials. The ADCS trial has included multicenter imaging, so there will be data comparing imaging to pencil and paper for tracking change.

Amy Graves: Doug, MCI is okay to "test" but ultimately we will want to know who to give the preventions to.

Giulio Pasinetti: Doug, as we have stated, the sum of box scores with biomarkers as an initial target population.

Paul Aisen: Perhaps we are getting closer to imaging plaques and tangles in pre-symptomatic individuals (see ARF news story).

Huntington Potter: I hear from Jim Mortimer that the nun study seems to get good predictive power from hippocampal volume measures by MRI.

Amy Graves: …before they have symptoms.

Sue Griffin: MRI costs are so high how can we afford that large-scale?

Paul Aisen: Sue, even more expensive, but perhaps more promising, is PET imaging

Sarah Sagi: Which biomarkers can we test for predictive power? Is there a consensus?

Giulio Pasinetti: Sarah, this is the $40,000 question, we are all striving to answer that.

Kerry O'Banion: Good points about MCI. On another issue, any clinicians out there have a sense of how frequently current AD patients and elderly are "prescribed" NSAIDS? In a few years we will have open trial data showing up. Someone could be thinking about this prospectively.

Huntington Potter: Actually, too many MCI pill poppers could mess up future trials.

Todd Golde: Doug et al, MCI still might be too late. The brain is shrinking, but it's a good place to start.

Giulio Pasinetti: Todd, how early do you think we should start?

Todd Golde: Giulio, anybody's answer to this is pure speculation.

Sue Griffin: All are right who say MCI is too late, see John Morris report. (Morris & Price, 2001).

Huntington Potter: I would focus first on genetic risk factors, then on inflammatory markers.

Sue Griffin: Agreed, Hunt; those are the real drivers. Then NSAIDS should be one of the strategies against overproduction of apoE4, which Steve Barger showed was detrimental (Barger & Harmon, 1997).

Todd Golde: What happens when we treat somebody with symptoms and show that the drug lowers Aβ in CSF and plasma, but has no effect clinically? Should we do the prevention trial (assuming the drug is safe)?

Kerry O'Banion: Todd, I think the prevention trials are required.

Kiminobu Sugaya: I worry about the physiological function of APP. If we suppress it, what is going to happen?

Gabrielle Strobel: Paul, Marilyn Albert has a combined MRI measurement of 4 brain areas in undiagnosed people who report a memory complaint that is 93% accurate in predicting who will convert. Could that help identify early subjects?

Sue Griffin: Gabrielle, yes.

Paul Aisen:Gabrielle, yes, but memory complaint may indicate MCI; I would like to identify people pre-MCI.

Giulio Pasinetti: Gabrielle, this is encouraging, and an appropriate index for targeting.

Huntington Potter: Memory complainers are often apoE4.

Gabrielle Strobel: Their major problem is telling apart future converters from those with normal age-related memory loss who do not progress to AD.

Huntington Potter: Gabrielle, right.

Huntington Potter: If any of this works, then we really have a problem because earlier diagnosis will pick up 30 percent of the total population.

Todd Golde: All: With the increased use of statins, it should be possible to tap into that population to enable a true prevention study. The patients are there, there just has to be commitment to enroll them in studies to assess cognitive function.

Douglas Galasko: All: we can try to measure Aβ and amyloid in the brain, plasma, or CSF in patients to see whether treatment has an effect. Is there data on useful, clinically relevant markers of inflammation (or other pet pathways) that should be added to clinical trials?

Kiminobu Sugaya: Doug, why Aβ concentration in serum? It is not correlated with AD pathology.

Douglas Galasko: Kiminobu, plasma is convenient. Also, plasma Aβ may tell us something about total amyloid production from day to day (elevated in Down's, FAD, and in people with a family history of AD).

Kiminobu Sugaya: Doug, if the Aβ in serum is a marker for familial AD and Down's, that kind of population can be screened by gene not measurement of Aβ.

Douglas Galasko: Kiminobu, presumably Aβ in plasma is produced by the same secretases as in the brain. Clearance may be slightly different. It remains to be proven, but it is possible that measuring plasma Aβ could help tell if an anti-amyloid drug is working

Paul Aisen: Doug, given the general futility of measuring plasma Aβ in sporadic AD, I think we will need a measure of Aβ in brain.

Paul Aisen: If we can identify a low-cost, safe, effective remedy, we can treat the entire aging population; but NSAIDs definitely do not fit into this category.

Sue Griffin: Exactly why not, Paul?

Paul Aisen: NSAIDs are toxic, the drug class causing the most iatrogenic illness.

Giulio Pasinetti: Paul, not all NSAIDs are toxic, tolerability has been a road block in several studies since Mr. Bayer invented salicylate.

Todd Golde: Paul, statins fit the bill (at least the safety part).

Gabrielle Strobel: Todd, Bayer pulled their statin for demyelinating side effects. Do we have long-enough exposure on the others to be sure this will not happen again?

Sue Griffin: Right. Todd wish statin did, then we'd just call it a vitamin.

Paul Aisen: Statins are safer, but not that safe (remember Baycol).

Todd Golde: All drugs have side-effects, the clinical experience with statins is that they are SAFE relative to almost any other drug out there. Compared to NSAIDs they are sugar pills

Huntington Potter: Besides, my neurons need cholesterol.

Todd Golde: Hunt, your brain cholesterol will not change with statin treatment (other precursors may).

Alexei Koudinov: I am not sure about that (see article by Eckert et al., 2001.)

Huntington Potter: But it is brain cholesterol that may be crucial.

Kiminobu Sugaya: Cholesterol may directly relate to the AβPP and other receptors like apoE2R, because they have NPXY in the C' terminal.

Alexei Koudinov: please see recent apoE forum transcript for additional possibilities.

Huntington Potter: Maybe low-dose delayed-release aspirin fits the bill.

Todd Golde: Hunt, epidemiology does not support aspirin.

Giulio Pasinetti: Does anyone have an idea of the influence of NSAID on cholesterol (mechanisms)?

Giulio Pasinetti: Doug, is AD pathology or dementia what we are interested in?

Douglas Galasko: Both Giulio. If we could index AD pathology presymptomatically in older folks, that would be a great target to treat. Even then we'll need to use clinical endpoints to validate whether such an approach helps.

Giulio Pasinetti: Doug, absolutely, we agree and we have encouraging data.

Todd Golde: Will academics be the only ones who really push for preventive trials?

Giulio Pasinetti: Todd, no, I'm sure Pharma will be right behind them.

Douglas Galasko: Todd, I think you're right. For Big Pharma, the marker of however many millions of AD patients are out there is the most appealing. A preventive treatment needs to meet higher proof of effectiveness, and cannot be toxic...

Todd Golde: I know from talking to Pharma the last thing they want is run a preventive trial. In fact I do not think they ever will unless the NIH supports it.

Sue Griffin: If we could help each other reach our goals-what would everybody like? I want to do a really big study of MCI and IL-1 genotypes and haplotypes. What about you all?

Huntington Potter: I agree Sue. That is the best we can do now.

Todd Golde: Sue, any genetics study should be comprehensive, looking at all potential markers. Picking one's own favorite marker is not going to be that informative.....

Sue Griffin: Oh, picking IL-1 is like dynamiting a pond for fish-it's a regulator of lots of bad un's…

Huntington Potter: It is much easier to add a slew of genetic markers. They may even interact with each other, like apoE4 and ACT may.

Todd Golde: Hunt, I agree they almost certainly will.

Sue Griffin: The little "marker battery" we have now is a given, of course. I didn't mean to leave that out.

Giulio Pasinetti: All, what are the best directives for future trials of NSAIDs?

Kiminobu Sugaya: We should combine some other treatment method with NSAIDs to see a more dramatic effect.

Huntington Potter: There is an NIH primary prevention trial starting with certain NSAIDs and MCI patients, I believe.

Giulio Pasinetti: Hunt, yes there is, it is the ADAPT trial by John Breitner.

Kerry O'Banion: I hope ibuprofen is among those NSAIDs!

Sue Griffin: Me too, Kerry.

Todd Golde: Yes, but how is recruiting going? Is it powered enough? Are they using the right drug (naproxen)?

Giulio Pasinetti: ALL , we have less then 10 min remaining. I would appreciate final statements on this chat. As I said before, we expect to see you at the satellite symposium in Stockholm.

Douglas Galasko: Comparative long-term studies of different compounds in animal models, and dose-finding studies in humans would be an excellent preface to clinical trials. These would be good vehicles to explore Aβ, IL-1, apoE and other variables of interest before betting everything on the clinical trial. As a counter-example, the NIH primary prevention trial uses naproxen and a selective COX-2 inhibitor, designed by extrapolation from epidemiological studies and the general concept that inflammation is bad.

Kerry O'Banion: If the ADAPT trial is celecoxib vs naproxen, then expect no info for Aβ- processing effects!

Todd Golde: Agree with Doug. As the ELAN trial points out.

Huntington Potter: Early diagnosis is key. Starting treatment early is essential. The drugs must be safer than most because of the long term of treatment. Moreover, I am in favor of finding drugs that specifically target the Aβ-apoE interaction. They are less likely to have other anti-inflammatory side effects.

Todd Golde: Many obstacles are likely to be encountered during the clinical testing of any disease-modifying therapy for AD. Some of these obstacles may be insurmountable and others may not.

Kiminobu Sugaya: NSAIDs may stop inflammation cascades in AD, but it does not regenerate neurons. It seems too toxic to use in prevention. So, what we can add to it?

Sue Griffin: Definition of markers is essential, as is characterization of their functions. The latter in order to search for other points of entry for cheaper safer and more effective drugs.

Todd Golde: Another side bet: neither celecoxib or naproxen will work. Any takers? I like dark beer.

Kerry O'Banion: I'll just share that beer with you.

Huntington Potter: I agree, Todd.

Kiminobu Sugaya: Is Aβ always a bad guy? How about the physiological function of AβPP?

Gabrielle Strobel: Kiminobu, we are planning an upcoming chat on this topic.

Huntington Potter: I agree that AβPP and even Aβ may have important functions. Both are harmless unless amyloid starts. The key is to prevent the conversion and leave normal levels of Aβ intact.

Sue Griffin: Hunt, too bad we can't limit the original production of AβPP to just the right amount-conversion may be too late.

Sue Griffin: I don't think Aβ is always bad. And for sure AβPP is not. Apparently we have to have it to respond to injury, although sometimes the response is overwhelming and inappropriate.

Kerry O'Banion: Sue, nice view on the issue.

Kiminobu Sugaya: I found AβPP in stem cell differentiation. In low concentration, it increases neuronal differentiation and at high concentration increases glial differentiation. [Editor: did I edit this sentence correctly? And: is this data published? Citation? Thanks.]

Todd Golde: The data is clear that Aβ is bad when it aggregates.

Kiminobu Sugaya: Suppressing AβPP too much could be dangerous.

Gabrielle Strobel: We have passed the end of the hour. I'd like to give Giulio the opportunity for a final statement. Where should we go, Giulio?

Giulio Pasinetti: In summary, the general consensus is that prevention (MCI and earlier) as early as possible is key. Future identification of biomarkers in concert with imaging studies will also be necessary. We also need to identify more tolerable NSAID compounds, and further characterization of how these compounds elicit their effects. Thank you all for participating.

Gabrielle Strobel: Thanks everyone for contributing. You are most welcome to continue this discussion via email and to send comments. Great, lively chat! Goodbye everybody.


Background Text
By Giulio Maria Pasinetti, Mount Sinai School of Medicine

Epidemiological evidence suggest that non-steroidal anti-inflammatory drugs (NSAID) may protect against Alzheimer disease (AD). However, therapeutic studies with NSAIDs, including cyclooxygenase (COX) inhibitors and steroids have not supported such epidemiological evidence. The apparent inconsistency may be due to the fact that the epidemiological evidence is based on studies examining AD before clinical manifestations are apparent, while therapeutic studies have been carried out on people with illnesses severe enough to exceed the clinical detection threshold. Thus, it is conceivable that therapeutic strategies administered during early Alzheimer' s disease dementia or moderate dementia may not be optimally effective. Alternatively, the influence of inflammatory activity in the brain for cases at high risk to develop Alzheimer's disease, e.g. MCI cases, as a potential target of anti-inflammatory drugs in clinical studies maybe more suitable to be studied.

The primary action of NSAID's is inhibition of the cyclooxygenase (COX) enzymes COX. COX enzymes exist in an inducible form COX-2, that has been found to be elevated in the AD brain, and a constitutive form COX-1. Both COX-1 and COX-2 are known to be involved in numerous inflammatory activities as well as normal neuronal functions. In-vitro it has been demonstrated that non-selective inhibitors of COX can preferentially decrease the levels of the highly amyloidogenic b-amyloid (Ab)1-42 peptide. Recent studies testing non-selective NSAID's in murine models of AD neuropathology indicated that the frequency of Ab plaque deposits in the brains of these animals can be significantly reduced by treatment with the non-selective COX inhibitor ibuprofen.

These studies and epidemiological data strongly support a therapeutic potential for NSAID's in the treatment of AD. Upon this premise, industry and academia are devoting a tremendous amount of resources to the testing of anti-inflammatory drugs for the treatment of Alzheimer's disease (AD). However, given the large number of candidate anti-inflammatory drugs and their widely divergent activities, it is essential to optimize drug selection and study design. A better understanding of the influence of inflammatory activity in AD, and identification of the specific mechanisms which play an early role in the disease's progression will greatly will improve the likelihood of success in efforts to find an effective anti-inflammatory treatment strategy.

We would like to discuss recent developments reinforcing anti-inflammatory drugs as therapeutic in the treatment of AD amyloidosis, and the relevance of understanding the role of COX and other inflammatory mediators in AD neuropathology and the clinical progression of AD dementia. These discussions may provide important criterion for the design of clinical trials of anti-inflammatory drugs in AD.


  1. The Role of NSAIDs and Cyclooxygenase on Beta-amyloidosis and Clinical Dementia

    Several recent developments in the study of anti-inflammatory drugs as possible therapeutic agents to prevent or treat AD must be reconciled:

    1. Epidemiologic evidence continues to suggest that chronic exposure to traditional non-selective NSAIDs reduces subsequent risk of AD. The recent compelling analysis of data from the Rotterdam Study is consistent with prior epidemiologic studies. It suggests that long term use of drugs such as ibuprofen and naproxen by prescription, particularly when used for periods exceeding two years, dramatically reduces risk of AD.

    2. Cell culture and transgenic mouse studies indicate that some NSAIDs, but not others, have a favorable impact on amyloid peptide generation and deposition. Ibuprofen is effective in these models, but naproxen and selective COX-2 inhibitors are not. The effective NSAIDs seem to reduce amyloidogenic cleavage of APP by a mechanism independent of COX inhibition.

    3. The major trials of anti-inflammatory drugs for the treatment of AD have been negative. Studies of prednisone, celecoxib and hydroxychloroquine have not demonstrated any benefit in the treatment of patients with mild to moderate AD.

    Some unresolved questions:

    • Do the epidemiologic studies reflect a causal relationship between NSAID use and reduced risk?
    • To what extent does the artificiality of the model systems cloud applicability to AD in humans?
    • Can the negative results of specific treatment regimens be generalized?
    • How feasible is long term treatment with non-selective NSAIDs in the frail population at risk for AD?
    • If such treatment is to be effective, how early must it be instituted?
    • Is it time to move away from the anti-inflammatory treatment strategy?
  2. Our published work indicates that (1) COX-2 expression in the AD brain tissues involves blood-derived macrophages, not neurons; (2) macrophages are infiltrating the neuritic plaques and phagocytizing amyloid-beta. COX-2 activation in the brain appears to be associated with inflammatory activation of peripheral monocytes/macrophages. Does the current work in other laboratories support neuronal COX-2 expression

  3. It appears that expression of Cox-1 and 2 varies with species, dose, route and type of insult. Neurons, microglia, astrocytes and also the peripheral macrophages that migrate into the brain following injury can all express Cox-2. The expression of Cox-1 has also been noted in microglia. I am not sure how much specificity can be expected in Cox-2 expression based on these observations. My hunch is that invading macrophages and microglia may make up a large portion of the Cox-2 activity in inflammatory states but that needs to be certainly challenged and verified with better techniques in localizing COX-2 reactivity in different environments.


    . In vivo expression of cyclooxygenase-2 in rat brain following intraparenchymal injection of bacterial endotoxin and inflammatory cytokines. J Neuropathol Exp Neurol. 1999 Nov;58(11):1184-91. PubMed.

    . Cyclooxygenase-2 is highly expressed in microglial-like cells in a murine model of prion disease. Glia. 2000 Feb 15;29(4):392-6. PubMed.

    . Inflammatory responses to amyloidosis in a transgenic mouse model of Alzheimer's disease. Am J Pathol. 2001 Apr;158(4):1345-54. PubMed.

    . Cyclooxygenase expression in microglia and neurons in Alzheimer's disease and control brain. Acta Neuropathol. 2001 Jan;101(1):2-8. PubMed.

  4. Strong correlations exist between neuroinflammation as observed in Alzheimer's disease brain, antiinflammatory protection as observed in epidemiological studies, and therapeutic benefit as observed in clinical trials of traditional NSAIDs. Nevertheless, there are many comments in the literature expressing doubt about these correlations. They are mostly attributable to some badly targeted clinical trials that have ended in failure.

    It is now incontrovertible that neuroinflammation is a prominent characteristic of AD pathogenesis. More than a thousand reports on the subject have so far been published. The most recent papers can be found in a special issue of the Neurobiology of Aging devoted to the subject (Vol. 22:6, Nov-Dec, 2001).

    It is also incontrovertible that antiinflammatory agents protect against AD. More than 20 epidemiological studies involving diverse subjects from 4 continents have confirmed this phenomenon. The most recent report was that of Veld et al. (see ARF news story) who analyzed the Rotterdam cohort of 6,989 subjects. They found that use of traditional NSAIDs for 2 years or more provided an 80 percent protection against AD. In other studies, some protection, but of a considerably smaller degree, was found for steroids (Breitner et al., 1995; Canadian Study of Health and Aging, 1994; McGeer et al, 1996.) A correlation of sorts is the identification of fewer reactive microglia in postmortem brains of NSAIDs users compared with controls (Mackenzie & Munoz, 1998), although no reductions were found for steroid users (Mackenzie, 2000).

    Translation of these findings into practice has been inexplicably weak. The only clinical trial within the boundaries of these prominent signposts was conducted nine years ago (Rogers et al., 1993). In that small, double-blind, placebo-controlled study of patients with probable AD, therapeutic doses of the COX-1 inhibiting NSAID indomethacin appeared to arrest progress of the disease. Five of 24 treated patients and 1 of 20 placebo patients withdrew because of gastrointestinal side effects. Two test patients and 5 controls withdrew for other reasons. Overall, 58 percent of test patients and 70 percent of controls patients completed the trial.

    An equally small trial of the mixed COX inhibitor diclofenac was not straightforward because the prostaglandin analog misoprostol was added as a gastroprotective agent (Scharf et al., 1999.) That strategy was not successful since 12 of 24 test patients dopped out because of GI problems. Even so, there were non-significant trends towards less deterioration in the test group.

    Other, poorly targeted clinical trials have been outright failures. Hydroxychloroquine proved to be of no benefit (Van Gool et al., 2001) but there were no data, either epidemiological or pathological, to indicate that it would work. Hydroxychloroquine is not an antiinflammatory agent. Rather it is an antimalarial drug which, for reasons yet unknown, is helpful in treating rheumatoid arthritis.

    Similarly, a trial of the COX-2 inhibitor celecoxib failed (1). Once again there was no persuasive evidence to suggest that it would be effective. Selective COX-2 inhibitors have not been in use long enough for epidemiological data to accummulate, but immunohistochemical evidence was available which indicated they should fail. Unlike COX-1, COX-2 is highly expressed in normal pyramidal neurons, as well as in pyramidal neurons of AD cases (Yasojima et al., 1999). Therefore, selective COX-2 inhibitors will primarily target pyramidal neurons rather than microglia.

    The prednisone trial also failed, despite the mild epidemiological support. The problem here was probably the use of a subthreshold dose of 10 mg/day (2). Based on the doses of steroids required to control acute relapses of multiple sclerosis, levels approximately three times higher would probably have been required. Whether such levels could be tolerated for any length of time is questionable, given the many physiological functions of glucocorticoids in the body. An additional problem is the known toxicity of steroids to hippocampal neurons.

    The preferred target for NSAID treatment of AD is clearly COX-1. GI side effects are an inevitable problem with such treatment. The question to be considered is whether to put brains ahead of guts. Currently approved treatments for AD do not alter neuronal loss. They only mask the disease by boosting cholinergic function. Since 100 percent of AD cases will inevitably deteriorate and only some will experience GI side effects, the balance would appear to favor NSAID treatment. Nevertheless, better strategies are required if the full benefits of COX-1 inhibition in brain are to be obtained. Perhaps the NO-NSAIDs, which have greatly reduced GI toxicity (Wallace JL et al., 1994) while retaining the beneficial effects of their parent NSAIDs, will provide the answer." (See also ARF news story.)

    See also:
    (1) Sainati SM et al. Results of a double-blind, placebo-controlled study of Celebrex for the progression of Alzheimer's disease. 6th International Stockholm-Springfield Symposium of Advances in Alzheimer Therapy, page 180..

    (2) Aisen P et al. Treatment of Alzheimer's disease with prednisone: results of pilot study and design of multicenter trial. J Am Geriatric Soc 43, SA27 (1995)


    . Delayed onset of Alzheimer's disease with nonsteroidal anti-inflammatory and histamine H2 blocking drugs. Neurobiol Aging. 1995 Jul-Aug;16(4):523-30. PubMed.

    The Canadian Study of Health and Aging: risk factors for Alzheimer's disease in Canada. Neurology. 1994 Nov;44(11):2073-80. PubMed.

    . Arthritis and anti-inflammatory agents as possible protective factors for Alzheimer's disease: a review of 17 epidemiologic studies. Neurology. 1996 Aug;47(2):425-32. PubMed.

    . Nonsteroidal anti-inflammatory drug use and Alzheimer-type pathology in aging. Neurology. 1998 Apr;50(4):986-90. PubMed.

    . Anti-inflammatory drugs and Alzheimer-type pathology in aging. Neurology. 2000 Feb 8;54(3):732-4. PubMed.

    . Clinical trial of indomethacin in Alzheimer's disease. Neurology. 1993 Aug;43(8):1609-11. PubMed.

    . A double-blind, placebo-controlled trial of diclofenac/misoprostol in Alzheimer's disease. Neurology. 1999 Jul 13;53(1):197-201. PubMed.

    . Effect of hydroxychloroquine on progression of dementia in early Alzheimer's disease: an 18-month randomised, double-blind, placebo-controlled study. Lancet. 2001 Aug 11;358(9280):455-60. PubMed.

    . Distribution of cyclooxygenase-1 and cyclooxygenase-2 mRNAs and proteins in human brain and peripheral organs. Brain Res. 1999 Jun 5;830(2):226-36. PubMed.

    . Novel nonsteroidal anti-inflammatory drug derivatives with markedly reduced ulcerogenic properties in the rat. Gastroenterology. 1994 Jul;107(1):173-9. PubMed.

Make a Comment

To make a comment you must login or register.


Webinar Citations

  1. From Epidemiology to Therapeutic Trials with Anti-inflammatory Drugs in Alzheimer's Disease: The Role of NSAIDs and Cyclooxygenase on Beta-amyloidosis and Clinical Dementia
  2. Alzheimer Immunotherapy Trial Grounded: Time to Reassess Safety and Vaccine Design

News Citations

  1. NO-Releasing NSAID Reduces b-Amyloid, Activates Microglia
  2. Anti-inflammatory Drugs Side-Step COX Cascade to Target Aβ
  3. New PET Probe to Aid Diagnosis and Monitoring of Alzheimer's Disease

Other Citations

  1. See Recent Review

External Citations

  1. Griffin et al. 2000
  2. Rogers et al., 1993
  3. Stewart et al., 1997
  4. Cedazo-Mingues et al. 2001
  5. Fox et al., 2001
  6. Griffin et al. 1989
  8. Hopkins ADAPT Study
  9. "How Cells Handle Cholesterol"
  10. Morris & Price, 2001
  11. Barger & Harmon, 1997
  12. Eckert et al., 2001
  13. ADAPT trial

Further Reading


  1. . The anti-inflammatory agents aspirin and salicylate inhibit the activity of I(kappa)B kinase-beta. Nature. 1998 Nov 5;396(6706):77-80. PubMed.