Background

Atomoxetine is a norepinephrine uptake inhibitor. It blocks a norepinephrine transporter and boosts noradrenaline levels in the brain. It is one of the few non-stimulant drugs used to treat attention deficit hyperactivity disorder in children, and is prescribed widely around the world. Norepinephrine is mainly produced by neurons of the locus coeruleus, a midbrain region that shows neuropathology and degeneration early in both AD and Parkinson’s disease (e.g., Jacobs et al., 2021). This and other lines of research implicate loss of noradrenaline in Alzheimer's disease (see, e.g., Dec 2010 conference news), hence atomoxetine's value as an add-on medication was investigated.

In preclinical work using the 5XFAD mice, increasing brain noradrenaline using atomoxetine and the noradrenaline precursor L-DOPS suppressed glial activation and Aβ deposition, and improved performance in the Morris water maze (Kalinin et al., 2011).

Findings

Between 2003 and 2006, Eli Lilly conducted a six-month Phase 2/3 repurposing trial at eight sites in the United States to evaluate the effectiveness of 25-80 mg/day of atomoxetine in 92 patients with mild to moderate Alzheimer's disease who were on stable doses of standard acetyl cholinesterase inhibitor therapy. Cognition as measured by the ADAS-Cog was the primary outcome of this trial. Atomoxetine in these patients was reported to be generally safe, though with an increase in heart rate; however, it did not benefit cognition in these patients (Mohs et al., 2009).

From 2012 to 2018, a Phase 2, 12-month biomarker study at Emory University assessed atomoxetine in people with mild cognitive impairment. It enrolled 39 people who had to have CSF Aβ and tau levels indicative of AD, to receive placebo or flexible doses of atomoxetine starting at 10 mg per day and increasing weekly to 100 mg or a maximum tolerated dose. In a crossover design, participants took drug or placebo for six months each, followed by an optional two-year open-label extension. Blood and CSF biomarkers were assessed at baseline, six, and 12 months. The primary outcome was change in neuroinflammation measured by CSF IL1 and thymus-expressed chemokine concentrations. Other primaries were adverse events and drop-out rate. Results are published (Levey et al., 2021). Atomoxetine treatment led to increased plasma and CSF norepinephrine concentrations, demonstrating target engagement. IL-1α and thymus-expressed chemokine were undetectable in most samples. CSF tau and p-tau181 were reduced by 5-6 percent compared to placebo, Aβ42 was unchanged. FDG-PET and functional MRI measures found increased glucose uptake and inter-network connectivity in the hippocampus and temporal lobe circuits with treatment. The differences persisted for six months after treatment. The regimen was safe and tolerable, with all but two participants reaching the 100 mg dose. Common adverse events were gastrointestinal symptoms, dry mouth, and dizziness. Other side effects were consistent with atomoxetine’s established profile and included increased heart rate, weight loss, and a trend for higher blood pressure. A proteomics analysis noted normalization of CSF biomarkers related to synaptic function, brain metabolism, and inflammation. As expected, there were no significant effects on cognition and clinical outcomes.

Atomoxetine is also being tested as a cognitive enhancer in people with Parkinson’s disease. It gave mixed results in several small trials (see analysis in Ghosh et al., 2020). One study indicated that Parkinson’s patients with low locus coeruleus volumes respond to atomoxetine (O’Callaghan et al., 2021).

For details on atomoxetine trials, see clinicaltrials.gov.