TO THE EDITOR:
Egan et al.1 reported adverse neurological effects including decreased memory and cognition in prodromal Alzheimer’s disease (AD) patients in a trail of verubecestat. Henley et al.2 observed similar findings in a trial of atabecestat in presymptomatic AD subjects. These compounds inhibit β-site amyloid precursor protein (APP) cleaving enzyme (BACE). We conducted neurobiological studies of verbucestat and two other BACE inhibitors (lanabecestat and LY2886721). All 3 compounds significantly inhibited long-term potentiation (LTP), a key electrophysiological correlate of memory, in mouse hippocampus (Figure 1A). Two of the 3 inhibitors decreased paired-pulse facilitation, a measure of presynaptic neurotransmitter release, and all 3 reduced multi-pulse bursts that reflect both presynaptic and postsynaptic plasticity (Figure 1B, C). These data provide a mechanistic explanation for the impaired memory and cognition documented in these two human trials: BACE inhibitors interfere with the physiological processing of known neuronal substrates of BACE (e.g., neuregulin; Scn2b) required for neurotransmission. One needs to identify BACE1 inhibitors which selectively inhibit APP processing more than that of other substrates to enable safe, chronic inhibition of this amyloid-generating protease.
Figure. 1.
Verubecestat, lanabecestat and LY2886721 impair synaptic plasticity in mouse hippocampus. A. All 3 BACE inhibitors decrease LTP in hippocampal CA1 slices after exposure for 5 hr: verubecestat 123 ± 8% (n=6) vs. vehicle (DMSO) 152 ± 7% (n=6), p<0.01; lanabecestat 134 ± 8% (n=7) vs. vehicle (DMSO) 155 ± 7% (n=7), p<0.05; LY2886721 132 ± 3% (n=6) vs. vehicle (DMSO) 151 ± 8% (n=6), p<0.05 (means ± SEM; unpaired Student’s t-test). B. Two of the 3 inhibitors impair paired-pulse facilitation (60 ms interstimulus intervals): verubecestat 1.23 ± 0.05 vs. vehicle (DMSO) 1.72 ± 0.075, p<0.001; LY2886721 vs. vehicle (DMSO) 1.45 ± 0.05, p<0.01. C. All 3 inhibitors significantly impair short-term plasticity in burst stimulations, especially by the 3rd-8th pulses. Insets in B-C represent typical field excitatory postsynaptic potentials (fEPSPs) recorded in vehicle (black) or after 5 hr verubecestat exposure (red). Horizontal calibration bar, 20 ms; vertical bar, 0.5 mV. Note that 10 μM doses enabled sufficient slice penetration in vitro to quantify acute effects; verubecestat administered chronically to rats at 25 mg/kg/day for 3 mo resulted in >1 μM free compound in brain 3 hr after final dose1.
Acknowledgments
Supported by R01 AG006173 from the National Institute on Aging
Footnotes
Disclosure forms provided by the author are available with the full text of this letter at NEJM.org
References
- 1.Egan MF, Kost J, Voss T, et al. Randomized trial of verubecestat for prodromal Alzheimer’s disease. N Engl J Med 2019;380:1408–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Henley D, Raghavan N, Sperling R, et al. Preliminary results of a trial of atabecestat in preclinical Alzheimer’s disease. N Engl J Med 2019;380;15. [DOI] [PubMed] [Google Scholar]
- 1.Kennedy ME, Stamford AW, Chen X, et al. The BACE1 inhibitor verubecestat (MK-8931) reduces CNS β-Amyloid in animal models and in Alzheimer’s disease patients. Sci Transl Med 2016;8;1–14. [DOI] [PubMed] [Google Scholar]