Pharma R&D Today

Ideas and Insight supporting all stages of Drug Discovery & Development

Select category
Search this blog

Review of Alzheimer’s Association International Conference (AAIC16) Part 2: Focus on Secretases

Posted on August 24th, 2016 by in Pharma R&D


In a previous blog, I provided a few thoughts on tau-related drug discovery news coming out of the recent Alzheimer’s Association International Conference (AAIC16) in Toronto, Canada. In this post, I will highlight some of the research from the conference on gamma and beta-secretase, which are fundamental small molecule targets for attacking the Amyloid Cascade Hypothesis. Despite some significant failures in past clinical trials with amyloid-directed drugs, targeting this putative cascade remains the prevailing framework for disease-modifying therapeutic strategies for Alzheimer’s disease (AD).

Gamma Secretase: “Not Dead Yet”

In a classic original Monty Python movie, a poor villager in plague-ridden London is being offered to the death cart collection despite his protests that he’s “not dead yet”. Similarly, despite the failures of two gamma-secretase inhibitors (GSIs) in late stage AD trials, gamma-secretase (GS) as a therapeutic target remains a viable strategy. The negative outcomes for these two GSIs, semagacestat and avagacestat, were remarkably similar and included evidence for impaired cognition. Clearly, new approaches to drugging GS are needed.

A better fundamental understanding of GS is required to aide in the discovery of next generation GSIs. As evidenced by presentations at an Emerging Concepts session focused on Secretase Biology, these novel insights appear to be forthcoming. (As an aside, this year’s AAIC16 saw a welcome return of basic science to the main program with the appearance of several Emerging Concepts platform sessions). Bart DeStrooper opened the Emerging Concepts: Secretase Biology session with a review of gamma secretase biology and a proposal that GSIs much better than first generation GSIs now could be developed. In particular, a better understanding of the various subtypes of gamma-secretase now allows for targeting PS-1 vs PS-2-containing GS complexes. Given their differential tissue and sub-cellular distribution, this may afford a certain level of substrate selectivity allowing scientists to tailor APP-preferring GSIs for AD and Notch-preferring GSIs for cancer.

Sjors Scheres followed with a cryo-EM analysis of GS which has yielded unprecedented molecular understanding of this multimeric protein complex. These structural insights into substrate binding and active site conformations in the presence and absence of inhibitors should prove as invaluable in designing next generation GSIs as the X-ray crystal structure of BACE1 has enabled development of potent BACE1 inhibitors.

Structural insights could also lead to more rationale design of gamma-secretase modulators (GSMs) which don’t inhibit GS per se, but rather shift the cleavage preferences towards shorter forms of Abeta. These GSMs don’t interfere with notch-signaling, a process that early GSIs blocked leading to adverse effects. Pfizer presented early clinical data on their GSM, PF-06648671, which yielded significant reductions in plasma Abeta 40 and 42 following a high (360 mg) single dose. A study with multiple ascending doses of PF-06648671 evaluating safety and tolerability as well as CSF ABeta pharmacodynamics is ongoing.

Thus, gamma secretase remains a viable therapeutic target, but not without challenges. In particular, the next generation GSIs or GSMs will need to overcome the stigma associated with this target emanating from the negative outcomes of earlier GSI clinical trials. And if the BACE1 inhibitors currently in clinical trial (see below) are successful, then investing in gamma secretase for AD will become a very hard sell.

The BACE Race

Discovering inhibitors of Beta-secretase, also known as Beta-APP Cleaving Enzyme 1 (BACE1), has been the “Holy Grail” of AD research since the initial description of the proteolytic processing of Abeta from the amyloid precursor protein (APP). Initially cloned in 1999, it took over a decade of research to understand how to drug BACE1, but now there are several potent inhibitors in clinical development.

Leading the pack

Verubecestat from Merck is the BACE1 inhibitor furthest in development. At this conference, there were only a couple of presentations on the baseline characteristics, demographics and biomarker measurements collected at baseline for the trial. The real action for this molecule will be coming next year when the Phase 3 trial is expected to readout. Similarly, BACE inhibitor AZD3293, the Lilly-AstraZeneca joint venture, is in Phase 3 and thus not a lot of new information was presented on this molecule – just some PK/PD modeling and the Phase 1b study in Japanese subjects.

Fast followers

The real excitement at the meeting in the secretase area of research came with the presentations on several new BACE1 inhibitors now in the clinic. While there are several BACE1 inhibitors in development, the molecules from Janssen, Lilly and Roche had the largest visibility at this conference and I will briefly highlight each below.

Janssen had several presentations on JNJ-54861911, a BACE1 inhibitor they licensed in and are co-developing with Shionogi. Along with good safety and tolerability, robust reductions in plasma and CSF Abeta 1-40 were reported in healthy Japanese subjects following a single dose (25-100 mg) or after four weeks daily dosing of 10 or 50 mg JNJ-54861911. Similar plasma and CSF Abeta pharmacodynamics were observed in subjects with prodromal AD (CDR 0.5) or asymptomatic at-risk for developing AD (ARAD) following four weeks of dosing at 10 or 50 mg.

Eli Lilly presented both several posters on the preclinical development of LY3202626 as well as the initial Phase 1 data in healthy subjects and a small study in AD patients. LY3203636 produced dose-dependent reductions in Abeta 1-x as well as proximal biomarkers, sAPPbeta and C99 in transgenic mice, and plasma and CSF Abeta in beagle dogs following low doses of LY3202626. The molecule was safe and well tolerated in Phase 1 trials and extremely potent, yielding 50% reduction in CSF Abeta 1-40 following a single 1 mg dose to healthy subjects.

Roche presented preclinical and early clinical data on CNP520, a BACE1 inhibitor being co-developed with Amgen. CNP520 yielded dose-dependent reduction in brain and CSF Abetas in rat and dog, respectively. Interestingly, in chronic dosing in mouse and other non-clinical species, there was no evidence of hypopigmentation, which is thought to be due to inhibition of BACE2, a highly related protease that is involved in melanosome function. Currently, the other BACE1 inhibitors in development are generally non-selective for BACE1 over BACE2. Given that two of these BACE nonselective compounds, verubecestat and AZD3293, have advanced into late stage Phase 3 trials, the clinical significance of this potential BACE2 inhibition remains to be determined. In first in human studies. CNP520 was safe and well tolerated and following 14 days of dosing produced > 40% reductions in CSF Abeta 42 at doses of 10 mg or higher.

Stumbling at the wire?

While there was much excitement around the excellent early safety and tolerability as well as the robust Abeta pharmacodynamics observed with these multiple BACE1 inhibitors, a word of caution is always necessary. In lessons learned from the gamma-secretase inhibitor development, APP is not the only substrate for these secretases, and while BACE1 appears to be not as promiscuous as gamma secretase, there are a variety of important non-APP substrates described for BACE1 such as neuregulin 1. Despite these potential stumbling blocks, the progression of BACE1 inhibitors into late stage Phase 3 studies gives investigators cautious optimism that BACE1 can be safely targeted. The bigger question is whether chronic reduction of steady state Abeta levels in mild AD will have a meaningful impact on the clinical progression of the disease. The answer to this big question awaits completion of the large Phase 3 trials currently ongoing or about to commence.

Conflict of interest: As a former employee at Eli Lilly and Co, I was involved in the early preclinical development of semagacestat and LY3202626.


All opinions shared in this post are the author’s own.

R&D Solutions for Pharma & Life Sciences

We're happy to discuss your needs and show you how Elsevier's Solution can help.

Contact Sales