Hypothesis: glutaminyl cyclase inhibitors decrease risks of Alzheimer’s disease and related dementias

Abstract

Alzheimer’s disease and related dementias (ADRD) comprise several progressive and incurable neurodegenerative disorders that some have classified as amyloidosis. With increased aging of the world’s population, the prevalence of the sporadic form of ADRD, which comprises over 99% of cases, continues to rise at an alarming rate. The enormous societal burdens of ADRD already rival those of the many other major chronic diseases causing premature morbidity and mortality in the USA and worldwide such as cardiovascular disease and cancer. At present, there is an insufficient totality of evidence concerning the efficacy and safety of any pharmacologic agents to delay slow progression or reduce complications of ADRD. In this context, glutaminyl cyclase (QC) inhibitors have shown some early possible evidence of efficacy with a reassuring safety profile. To reliably test the glutaminyl cyclase (QC) and any other promising hypotheses will require cogent data from large-scale randomized trials of sufficient size and duration.

Keywords:

• Alzheimer’s disease and related dementias
• glutaminyl cyclase inhibitors
• neurotherapeutics
• pharmacologic therapies
• randomized trials

As has been the case in the development of pharmacologic agents for Alzheimer’s disease and related dementias (ADRD), advances in medical knowledge proceed on several fronts, optimally simultaneously [1]. Basic researchers address the crucial hypothesis of why a pharmacologic agent has net benefit. Clinical researchers address the equally crucial and complementary hypothesis of whether a pharmacologic agent has net benefit. For hypothesis testing of large benefits, which means relative risks of two or greater, randomized trials are neither necessary nor desirable. For smaller and more moderate effects, however, the amount of uncontrolled or uncontrollable confounding inherent in all observational analytic study designs, no matter how large or well designed, may be as big as the effect sizes being sought. In such circumstances, the large-scale randomized trial represents the most reliable design strategy. For common and serious diseases, such small to moderate effects can have a large clinical and public health impact.

ADRD is one such common and serious category of several disorders classified as amyloidosis [2]. ADRD is characterized by abnormal folding of proteins into oligomeric diffusible and insoluble deposited material. Such aggregates are typically resistant to degradation by proteases and damage the affected cells and tissues. The deposition of the amyloid peptides amyloid-beta (Aβ) and hyperphosphorylated tau, representing the typical hallmarks of ADRD, is restricted to the brain. In affected patients, cognitive function becomes progressively impaired by processes, which result in premature death. ADRD is the most common form of dementia and by 2050, is predicted to affect 1 in 85 or over 80 million people worldwide. In the USA, ADRD is the sixth leading cause of death among adults and fifth among those aged 65 years and older. Since medical therapeutic options, including pharmacologic agents, are currently limited, informal caregivers, especially family members, provide the majority of care. As a consequence, financial and health-related burdens are enormous.

Currently used pharmacologic agents include inhibitors of acetylcholine esterase (AChEI) or putative antagonism of the N-methyl-D-aspartate (NMDA) receptor. These pharmacologic agents may offer limited short-term symptomatic relief but neither halt nor even delay progression of ADRD. To better understand the pathophysiology of ADRD, recent research has targeted a plethora of biomarkers as potential causal factors. These include such factors as the apolipoprotein E allele and hippocampal volume, but the evidence to support their utility is both inconsistent and limited. Several alternative pharmacologic agents have been hypothesized and tested. Among these strategies, one of the most widely studied, publicized, and as a consequence, financially well supported is the reduction of Aβ load in the brain. The amyloid cascade hypothesis is that the accumulation of A in the brain leads to neuronal dysfunction and death. The postulated mechanisms include reduced clearance and/or overproduction that initiates inflammation and intracellular aggregation of hyperphosphorylated tau. Thus, it has been hypothesized that either suppression of Aβ formation or enhanced clearance will halt the amyloid cascade at an early stage and, thereby, prevent neuronal dysfunction and death [3]. To date, however, the results of randomized trials designed a priori to test this hypothesis have generally been disappointing [4,5]. Semagacestat, a gamma-secretase (γ-secretase) inhibitor was hypothesized to decrease Aβ formation, but was, in fact, associated with worsening of cognitive function. In addition, a γ-secretase modulator also did not show efficacy. Most recently, the anti-Aβ antibodies bapineuzumab and solanezumab showed no statistically significant overall benefits, although subgroup analyses interpreted to be promising, are being used to formulate not test hypotheses [1].

It is tempting to speculate about why the plausible and rational hypothesis to reduce Aβ load did not translate into clinical benefits. Characterization of the sequence of pathological events, provided by recently developed imaging techniques, suggests that Aβ deposition can precede the first signs of cognitive impairment by many years. In moderate and late stage ADRD, the Aβ load in brain does not increase with clinical deterioration, and similarly high Aβ loads have been observed in the brains of cognitively normal human subjects [6].

For these and other cogent reasons, earlier initiation of the use of pharmacologic agents in early ADRD, such as during the early clinical stages of mild cognitive impairment (MCI) or, if plausible, even before the onset of symptomatic changes may be necessary and desirable. However, since ADRD is characterized by insidious onset, the timing of promising therapies is challenging to both clinicians and researchers.

In addition, the development of evidence-based practice guidelines is hampered by the incomplete totality of evidence. Some suggest that MCI has prognostic value at all stages, while others have challenged the legitimacy of these purported prodromal or subclinical stages of ADRD. Nonetheless, ADRD’s devastating clinical and psychological impact on patients and families is well documented, creating a perceived need for more immediately available behavioral and psychosocial interventions [7].

Pharmacologically, while Aβ may still be an appropriate neurochemical target, the specific mechanisms of benefit have not yet been clearly elucidated. The heterogeneity of Aβ, however, suggests that more precisely targeted interventions are likely to be necessary. For example, wild-type Aβ itself may play a protective physiological role [8]. In addition, C-terminal shorter forms of Aβ (Aβ38 and Aβ40) have been described to prevent the aggregation of the presumably more amyloidogenic peptide Aβ42 [9]. Nonetheless, most Aβ-directed pharmacological agents reduce all forms of Aβ, which could negate any possible benefits due to the loss of this protective physiologic function.

Recent research in ADRD has demonstrated that the Aβ peptides display very large heterogeneity. The majority of the AB peptides are N-terminally truncated and might be modified. It is unclear, however, whether these A forms have been processed subsequent to formation, or whether multiple alternative pathways of the processing of the amyloid precursor pathway may lead to the heterogeneity [10]. Taken together, these recent findings support the plausibility of interventions to accelerate selective Aβ clearance so that only toxic aspects are targeted. This hypothesis is supported by the observation that it is peptide clearance rather than Aβ overproduction that is reduced in AD [11].

One novel approach stemming from this hypothesis involves targeting pyroglutamate (pE, pGlu) modified Aβ [12]. These forms of the peptide are N-terminally truncated, and post-translationally modified at glutamate residues 3 or 11 of Aβ. As a consequence of pE formation, Aβ is rendered more hydrophobic, insoluble and stable toward degradation. In familial forms of AD caused by mutations in the protein PS1, pE-modified Aβ represents the dominant species [12]. Similarly, pGlu-Aβ has been shown to accumulate in the course of development of sporadic ADRD. Since pGlu-modified Aβ could function as a seed of amyloid/tau plaques, glutaminyl cyclase (QC) inhibition might specifically reduce the insoluble aggregates with Aβ peptides. This, however, is under-represented in normal aging. These findings are consistent with previous findings that demonstrate a crucial role of water soluble, oligomeric Aβ as the distinguishing factor between physiological aging and ADRD. With respect to pGlu-Aβ, recent studies support the hypothesis that these forms generate mixed oligomers with significantly enhanced surface hydrophobicity. These species exerted superior neurotoxicity over full-length A oligomers. pE-Aβ is able to transfer its molecular properties and ‘infect’ other, nonmodified peptides to form neurotoxic oligomers. With regard to inter-relationship of Aβ with tau, it is interesting to note that pEAβ-induced toxicity of oligomers on neuronal cells is absent when the neuronal cells have been prepared from tau-KO mice. Likewise, tg mice usually develop massive gliosis and neurodegeneration within 3 month of age. Cross-breeding of tg mice expressing only a small amount of the N-truncated and modified Aβ with tau-KO mice leads to a complete ablation of the pGluAβ dependent toxicity in the bigenic mice. These findings suggest that pEA is upstream in the toxicity cascade [13]. In summary, the emerging totality of evidence suggests pGlu-modified Aβ’s crucial role as driver of the amyloid cascade. Several trials are ongoing to test this hypothesis using agents either targeting the synthesis of these amino acid modifications or enhancing clearance.

The enzyme QC catalyzes the formation of pyroglutamate-Aβ [14]. The expression of this enzyme correlates with the appearance of pEAβ and, notably, QC is upregulated early in ADRD in humans. It is tempting to speculate that inhibition of QC might prevent the formation of pGlu-Aβ and suppress downstream pathophysiology. This, in turn, may lead to enhanced clearance of A due to reduced sequestration of Aβ to diffusible and insoluble aggregated material. In essence, the strategy aims at retrieval of Aβ clearance and, thus, remains consistent with current overarching paradigms that test the amyloid hypothesis. In contrast to other approaches, however, it does not modulate the molecular events of Aβ formation and therefore preserves the potential physiological function of Aβ. Recent data have been reported in preclinical studies applying pEAβz-specific antibodies to transgenic mice, hereby overexpressing amyloid precursor protein (APP) [15]. The pGlu-AB seeds AB oligomers, and the toxicity of these oligomers seems to result from changes in the secondary and tertiary AB structure conferred by pGlu-AB, leading to symptomatic impairment and neuronal cell death. QC targets pGlu-AB via two modes of action: inhibiting its formation with QC inhibitors, and increasing pGlu-AB clearance with specific pGlu-AB antibodies [16].

Promising early results have been reported about the safety and tolerability of QC inhibitors. They appear to be well tolerated in both young and elderly patients, but with substantially different rates of metabolism. Such data are a prerequisite to test whether QC inhibitors have a favorable benefit to risk ratio on clinical outcomes.

Based on observations of modified peptide abundance in the familial PS1 mutation in ADRD in humans, pilot studies in rapid progressing disease variants may also provide importantly relevant data. In addition, based on activity in this restricted population in what is likely to be shared pathobiology, it may be possible to expand testing to the broader general population of ADRD patients. Furthermore, it would be feasible to study efficacy and tolerability among patients with nonsporadic and aggressive forms of ADRD.

Gains in life expectancy are likely to continue and, if so, the incidence of ADRD will continue to rise. ADRD’s devastating global clinical and public health burdens include premature morbidity and mortality analogous to many other major chronic diseases such as cardiovascular disease and cancer. It is therefore not surprising that early promising findings of plausible but unproven benefits, while inadequately tested, continue to guide clinical practice. Given their reassuring safety profile and early evidence of efficacy, QC inhibitors represent an important and promising hypothesis. To reliably test this and other promising hypotheses about ADRD will require cogent data from large-scale randomized trials of sufficient size and duration.

Financial & competing interests disclosure

CH Hennekens is funded by the Charles E. Schmidt College of Medicine of Florida Atlantic University and serves as an independent scientist in an advisory role to investigators and sponsors as Chair or Member of Data and Safety Monitoring Boards for Amgen, AstraZeneca, Bayer, Bristol Myers-Squibb, British Heart Foundation, Cadila, Canadian Institutes of Health Research, DalCor, Genzyme, Lilly, Regeneron, Sanofi, Sunovion and the Wellcome Foundation. to legal counsel for Pfizer and Takeda, the United States (U.S.) Food and Drug Administration, and UpToDate; and receives royalties for authorship or editorship of 3 textbooks and as coinventor on patents for inflammatory markers and CV disease that are held by Brigham and Women’s Hospital; and has an investment management relationship with the West-Bacon Group within SunTrust Investment Services, which has discretionary investment authority and does not own any common or preferred stock in any pharmaceutical or medical device company. R Zivin owns common shares of Johnson & Johnson, Medgenics Palatin Technologies and Neurotrope Bioscience. JM Gaziano has received investigator-initiated research funding from the National Institutes of Health, the Department of Veterans Affairs, and Pfizer (formerly Wyeth). He reported serving as a consultant Santarus and as a consultant to and expert witness for Bayer. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.