As the research field’s collective eyes are shifting toward the γ-secretase modulators, what do scientists actually have on that score? At AD/PD 2011, several academic and biopharma groups presented on the topic ranging from fundamental mechanisms to late preclinical data. On the basic science front, researchers have made progress in understanding how γ-secretase modulators might work. For example, Taisuke Tomita and Takeshi Iwatsubo of the University of Tokyo made light-activatable probes with the research compound GSM-1. Using those, and also other methods, the Japanese scientists showed that GSM-1 binds to the N-terminal end of presenilin-1, in particular at the hydrophobic region of transmembrane domain 1. This part of the protein is at a distance from the hydrophilic pore deep in the membrane that harbors the active site of the enzyme complex, indicating that the mode by which GSM-1 tweaks APP cleavage is allosteric. A poster by Hiroyuki Amino and colleagues at the pharma company Eisai showed essentially the same thing for that company’s GSM E2012 (see more on this compound below), and scientists in other labs have independently seen GSM binding to presenilin-1’s N-terminal fragment. Allosteric modulation frequently works through conformational changes at the substrate binding site or the catalytic pocket of an enzyme, and can be quite tractable for drug development (see also Uemura et al., 2009).

On the translational science front, Sascha Weggen of Heinrich Heine University, Düsseldorf, Germany, updated his prior work showing that amino acid sequence matters. Previously, Weggen’s lab had shown that some AD-causing presenilin mutations respond poorly to GSMs (Czirr et al., 2007). Since then, the group has studied this hunch systematically. At AD/PD 2011, Weggen reported that, indeed, GSMs look feeble when tested against almost all familial PS-1 mutations the field has used over the years to model AD in transgenic mice and cell culture. In contrast, the same compounds work fine in models using pathogenic APP mutations. This does not mean GSMs would be ineffective in most patients; after all, the vast majority of AD patients have no presenilin mutation, but it does mean that scientists developing GSMs must choose their models carefully when they screen for such compounds and evaluate them preclinically (see also Hahn et al., 2011). Weggen noted that, in his experiments, the resistance of one presenilin mutation was broken by two recently published GSMs, but not by other second-generation compounds he tested, such as E2012 and BB25, an analog of GSM-1. Harald Steiner of Ludwig-Maximilians-Universität in Munich, Germany, reinforced this point in his talk. He showed that potent compounds such as E2012 and GSM-1 do reduce Aβ42 generation by many mutant forms of presenilin-1, but his group used higher concentrations than Weggen’s. This data appeared online last month (Kretner et al., 2011).

On the preclinical side, Kathy Rogers of EnVivo Pharmaceuticals, a biotech company in Watertown, Massachusetts, told the audience that its GSM, called EVP-0015962, decreased Aβ42 while concomitantly increasing Aβ38, but not the APP-CTFs.

The γ-secretase complex generates Aβ peptides of varying length, and the longer ones aggregate, particularly Aβ42. Modulating the enzyme means that it keeps chopping away, but processes APP slightly differently, such that it produces less Aβ42 and more Aβ38, or shorter isoforms but leaves the total amount of Aβ and its cytoplasmic tail AICD unchanged. (Whether the short forms are harmless has not been formally proven, but taking out Aβ42 is widely thought to stop the formation of oligomers and fibrils.)

Rogers showed data to suggest that EVP-0015962 behaves in the desired way, i.e., down with Aβ42, up with Aβ38, no effect on total load, in four different cell types. It leaves AICD generation unchanged, and neither α- nor β-CTF pile up. “We shift the site of cleavage, not the rate of cleavage,” Rogers said. The compound does not affect Notch cleavage. When added to food once, it reduced Aβ42 in the brains of wild-type and Tg2576 mice. Fed to Tg2576 mutant APP-transgenic mice for a year, neither APP’s α nor β-CTF went up. With this chronic exposure, the compound reduced neuroinflammation measured as cortical astrocyte and microglial activation, as well as plaque load in the hippocampus, Rogers showed. A stepwise biochemical extraction protocol showed that the compound reduced Aβ42 in all three pools analyzed: cytosolic, membrane-bound, and aggregated. It also reversed the mice’s memory deficit in the contextual fear-conditioning test. All effects were dose-dependent.

Incidentally, this compound has no effect in wild-type mice; hence, it is not a cognitive enhancer, Rogers said. The compound was well tolerated in mice at the doses tested, Rogers said. How about the all-important pharmacokinetics and pharmacodynamics? Rogers showed a small number of data in rats, where the compound lowers Aβ42 while upping Aβ38 in the CSF, and clears out over the course of some six hours. After the talk, EnVivo scientists said they have calculated a minimally efficacious dose and are currently studying the compound in non-human primates.

Responding to an audience question about whether this compound is an NSAID—those are the anti-inflammatory drugs that originally led scientists on the trail of GSMs years ago—Rogers said the company’s chemists started out with an NSAID and then modified it to where it now has no Cox1 or 2 activity. Another question from the audience touched on intestinal goblet cells, asking whether their numbers were up with EVP-0015962. Nope, said Rogers, they were not.

This question pertains to the intestinal toxicity that arises when γ-secretase inhibitors (GSIs) inhibit Notch. It came up again on a poster by scientists at the pharmaceutical company Eisai Company, a pharmaceutical company that developed donepezil. In Barcelona, Mai Uesugi and colleagues showed how they had compared gene expression in rats treated with either their GSM E2012 or a GSI, looking specifically for changes that would indicate interference with Notch signaling. In theory, a GSM might shift cleavage of Notch as well, and in this way perhaps deprive the cell of certain Notch fragments that are needed for subsequent signaling. On the poster, the scientists showed that the GSI, but not the GSM, reduced expression of target genes downstream of Notch signaling in rat intestine. The GSI, but not GSM, led to increased numbers of goblet cells in the gut as well.

Led by Christa Nagy, Eisai also presented a poster with some single-dose human pk/pd data on 10 different doses ranging from 1 to 400 mg of E2012. The poster suggested that E2012 reduces mostly Aβ42, but also Aβ40 in plasma of healthy volunteers. Eisai had previously tested this drug in Phase 1 but stopped when a high dose group appeared to develop a problem with the lenses of their eyes. Apparently, the FDA permitted resumption of clinical testing for E2012 in 2008, according to an article in FierceBiotech, but no trials with this compound are listed on or the IHO clinical trials listing.

At least one other GSM is currently in trials, however. That is Chiesi’s CHF5074 (Lanzillota et al., 2010; Imbimbo et al., 2009). That compound is getting ready for a Phase 2 dose-finding trial in people with MCI to start this month in Italy and New Jersey, U.S. Chiesi, an Italian company, did not present at AD/PD 2011, nor did Satori Pharmaceuticals, a U.S. biotech company that has a preclinical second-generation GSM. Other companies did, though, for example, the Swiss pharma giant Hoffmann-La Roche, which showed data on aminothiazole GSMs.

While everyone is looking for the right GSM, plenty of questions remain. One is when a person would have to start taking it. For example, the Tg2576 mice in the EnVivo study started taking it at five months of age, when Aβ levels are high but before the mice have deposited amyloid. How much would such a drug help in people, who tend to have abundant amyloid deposition in their brains years before they develop symptoms of AD? At AD/PD 2011, Eddie Koo of the University of California, San Diego, spoke for many scientists when he emphasized the need to treat early and to test drugs early. Because scientists think GSMs are safer than GSIs, they hope to use them toward that end. But will a GSM be enough? Joanna Jankowsky at Baylor College of Medicine, Texas, reported long-term findings of her studies treating transgenic mice with GSIs. She cautioned that shutting off Aβ production might not suffice once the brain is riddled with plaques. At that point, a GSM plus immunotherapy together might be required. This kind of treatment is straightforward to model in mice, but not practical in humans yet. In humans, clinical trials are beginning to inch their way back from mild to moderate AD toward prodromal AD, a stage where amyloid pathology is already present. Neither prevention nor combination therapy with GSMs and immunotherapy is a reality yet, but efforts such as the Dominantly Inherited Alzheimer Network (DIAN), API, and ADCS A4 are actively working toward at least secondary prevention trials, where amyloid is present but symptoms are not (for extensive coverage of these efforts, see ARF London conference series; ARF DC series; ARF Webinar).

Another potential hurdle is looming on the horizon. There were whispers at AD/PD and previous conferences that vasogenic edema—a mysterious side effect that first cropped up in Phase 1 trials of passive immunotherapy—might be something all anti-amyloid therapies may have to contend with. Immunotherapy trials already are under orders from the FDA to closely monitor patients for these edemas, and this practice is gradually creating more data about them. One early idea had been that rapid transport of large amounts of amyloid from the parenchyma to the brain’s blood vessels might cause transient fluid retention as capillaries become temporarily more permeable to serum proteins, and that this would subsequently resolve as the brain gradually clears the amyloid. Since last winter, however, some scientists have speculated privately that even γ-secretase inhibitors have caused these fluid shifts in recent trials. In Barcelona, Reisa Sperling of Brigham and Women’s Hospital, Boston, said nothing in her talk about specific trials and declined comment afterward. But she did say that vasogenic edema could potentially be a general complication of all Aβ-lowering strategies. Interestingly, the Chiesi trial of its GSM makes MRI including the FLAIR sequences that visualize vasogenic edemas an inclusion criterion.

Sperling showed data on two women who developed vasogenic edema in immunotherapy trials at her site. One had no symptoms and the edema was gone after three months. The other woman was confused and her MMSE dropped; subsequent imaging procedures showed that she had a microhemorrhage, which resolved on its own. The cases beyond those two examples range from clinically invisible to locally inflammatory and requiring steroids, Sperling said. Vasogenic edema typically shows up after the first or second infusion, and in most cases, patients resume dosing after a while. Thus far, vasogenic edema has not changed patients’ outcomes, she added.

“I believe vasogenic edema has to do with early, rapid shifts in Aβ load,” Sperling said. One possible explanation might be that rapid movement of Aβ plugs up drainage through the perivascular space, though there is also some data for clearance happening right at the vessel and for toxicity to the vessel. “There is an important balance between production and clearance of Aβ. Vasogenic edema is not specific to immunotherapy, but arises when this balance changes at the vessel temporarily,” Sperling said.

ApoE4 genotype and CAA are risk factors; in fact, some people with CAA have those edemas spontaneously even in the absence of dementia or AD. Vascular amyloid somehow is a common underlying pathophysiology, but the exact relationship among vasogenic edema, Aβ, and microbleeds remains elusive at this point (for more on microbleeds, see ARF 2011 HAI conference story).

“Ideally we need to treat 10 years before people get symptoms. We can see that vessel amyloid is associated with vasogenic edema or microbleeds in models even before they are symptomatic. That means we will have to be careful with vasogenic edema even in preclinical trials in the future,” Sperling concluded.—Gabrielle Strobel.

This is Part 2 of a two-part series. See also Part 1.


  1. Comment by Sascha Weggen and Stefanie Hahn
    To understand how presenilin mutations affect γ-secretase modulators (GSMs), it's worth elaborating on the differences in the studies by Hahn et al. (Hahn et al., 2011) and Kretner et al. (Kretner et al., 2011). Both studies used similar compounds. Both used E-2012 as an example of a potent non-acidic GSM without a carboxylic acid group. In addition, Kretner et al. used GSM-1 as a potent acidic GSM, whereas we used BB25, which is a close analogue of GSM-1 with similar potency. Kretner et al. concluded from their results that the attenuating effects of most presenilin mutations could be overcome by second-generation GSMs. In contrast, Hahn et al. concluded the opposite: that the efficacy of second-generation GSMs is reduced by presenilin mutations, in a similar fashion to what was previously reported by both of our groups for the less potent NSAID-type GSMs such as sulindac sulfide (Czirr et al., 2007; Page et al., 2008). The different outcomes of the two studies appear to be due to differences in the compound concentrations used and the statistical analysis.

    Kretner et al. used higher GSM concentrations. We conducted our experiments in a dose-dependent fashion with a wide range of concentrations from 50-1,000 nM for E-2012 and BB25, whereas Kretner et al. used 1 μM in most experiments. The IC50 of these compounds in cell-based assays is around 50-200 nM. Our experience is that, if you go high enough with the compound concentration, one can break almost every presenilin mutation. However, we do not think that these concentrations are a realistic approximation of the in-vivo situation. A general problem with current GSM structures is that they are highly lipophilic. This means that high doses will be required to reach effective free drug concentrations in vivo. In this respect, concentrations five- to 20-fold above the in-vitro IC50 seem overly optimistic.

    We are also not convinced that second-generation GSMs are principally different from the older NSAID-type GSMs. Kretner et al. suggested this because they found that many more mutants did not respond to sulindac sulfide as compared to GSM-1. However, they used sulindac sulfide at 50 μM, which is around its IC50 in cell-based assays, and GSM-1 at 1 μM, which is five times its IC50. Higher sulindac sulfide concentrations are very toxic on cells. Therefore, it is simply not possible to do the experiment, but if one could go high enough with the sulindac sulfide concentrations without inducing toxicity, we would predict sulindac sulfide would behave exactly the same as the new GSMs.

    More important are the differences in the statistical analyses between our studies. Kretner et al. tested for significance of drug treatment versus vehicle control in each cell line separately. In many mutant cell lines, they observed a significant drop in Aβ42 levels after drug treatment and concluded that the mutants do respond to the drug. In contrast, we tested (at a given drug concentration) for a significant difference between the Aβ42 reductions in cell lines expressing mutant presenilin versus a control cell line expressing wild-type PS1. In this way, one can observe a significant drug effect with presenilin mutants, but the effect is still less in comparison to wild-type presenilin.

    In conclusion, our studies seem to report contradictory results for similar or identical compounds, but these differences can be explained by the experimental design and the statistical analysis. So what is the truth? As an example, imagine treatment of a sporadic AD patient and a PS1 FAD patient with a GSM. Let's say the wild-type patient shows a 35 percent drop in CSF Aβ42 after GSM treatment. Most companies would likely be pleased with such an effect size. Now, in line with the data in both of our studies, the presenilin mutation in the FAD patient might attenuate the drug effect and cause only a 20 percent reduction in Aβ42. If you do a statistical analysis in the FAD patient of treatment versus non-treatment, a 20 percent drop is likely statistically significant. So is the 15 percent difference between treatment efficacy in the wild-type and the mutant patient. In this sense, the statistics in both studies are correct. However, although it may be possible to significantly lower Aβ42 in the FAD patient, the 15 percent difference in Aβ42 reduction could have profound implications for the clinical efficacy of the compound. Obviously, we do not know whether FAD patients with presenilin mutations will ever be treated with a GSM in the future, and this thought experiment might be a purely academic exercise. Nevertheless, these two papers nicely illustrate how differences in statistics can lead to completely different interpretations of the data. By just looking at the experimental raw data, there is very little difference between our studies.


    . Presenilin-1 but not amyloid precursor protein mutations present in mouse models of Alzheimer's disease attenuate the response of cultured cells to γ-secretase modulators regardless of their potency and structure. J Neurochem. 2011 Feb;116(3):385-95. PubMed.

    . Attenuated Abeta42 responses to low potency gamma-secretase modulators can be overcome for many pathogenic presenilin mutants by second-generation compounds. J Biol Chem. 2011 Apr 29;286(17):15240-51. PubMed.

    . Generation of Abeta38 and Abeta42 is independently and differentially affected by familial Alzheimer disease-associated presenilin mutations and gamma-secretase modulation. J Biol Chem. 2008 Jan 11;283(2):677-83. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. London: Europe-U.S. Regulators Mull Prevention Trial in Familial AD
  2. Scientists and Regulators Discuss Preclinical AD Trials
  3. Miami: Astrocytes, Antidepressants, Microbleeds, and More
  4. Barcelona: Live and Learn—γ-Secretase Inhibitors Fade, Modulators Rise

Webinar Citations

  1. Treating Before Symptoms—ADCS Invites Ideas for Clinical Trials in Very Early AD

Paper Citations

  1. . Allosteric modulation of PS1/gamma-secretase conformation correlates with amyloid beta(42/40) ratio. PLoS One. 2009;4(11):e7893. PubMed.
  2. . Insensitivity to Abeta42-lowering nonsteroidal anti-inflammatory drugs and gamma-secretase inhibitors is common among aggressive presenilin-1 mutations. J Biol Chem. 2007 Aug 24;282(34):24504-13. PubMed.
  3. . Presenilin-1 but not amyloid precursor protein mutations present in mouse models of Alzheimer's disease attenuate the response of cultured cells to γ-secretase modulators regardless of their potency and structure. J Neurochem. 2011 Feb;116(3):385-95. PubMed.
  4. . Attenuated Abeta42 responses to low potency gamma-secretase modulators can be overcome for many pathogenic presenilin mutants by second-generation compounds. J Biol Chem. 2011 Apr 29;286(17):15240-51. PubMed.
  5. . The γ-secretase modulator CHF5074 reduces the accumulation of native hyperphosphorylated tau in a transgenic mouse model of Alzheimer's disease. J Mol Neurosci. 2011 Sep;45(1):22-31. PubMed.
  6. . CHF5074, a novel gamma-secretase modulator, attenuates brain beta-amyloid pathology and learning deficit in a mouse model of Alzheimer's disease. Br J Pharmacol. 2009 Mar;156(6):982-93. PubMed.
  7. . Alzheimer's prevention initiative: a proposal to evaluate presymptomatic treatments as quickly as possible. Biomark Med. 2010 Feb;4(1):3-14. PubMed.

External Citations

  1. article in FierceBiotech
  2. Phase 2 dose-finding trial
  3. Dominantly Inherited Alzheimer Network

Further Reading


  1. . Attenuated Abeta42 responses to low potency gamma-secretase modulators can be overcome for many pathogenic presenilin mutants by second-generation compounds. J Biol Chem. 2011 Apr 29;286(17):15240-51. PubMed.
  2. . Presenilin-1 but not amyloid precursor protein mutations present in mouse models of Alzheimer's disease attenuate the response of cultured cells to γ-secretase modulators regardless of their potency and structure. J Neurochem. 2011 Feb;116(3):385-95. PubMed.