AD trial: Cognitive impairment exacerbated with reduced iron in brain

Brain iron-lowering therapy with deferiprone appears to do more harm than good in patients with Alzheimer's disease (AD), being associated with increased cognitive decline in a phase II study.

In a cohort of 81 patients with amyloid-confirmed mild cognitive impairment or early AD, treatment with deferiprone 15 mg/kg twice a day for 12 months resulted in accelerated cognitive deterioration, as assessed using a neuropsychological test battery (NTB) of memory, executive function, and attention tasks, compared with placebo (change in NTB composite z score, −0.80, 95 percent confidence interval [CI], −0.98 to −0.62 vs −0.30, 95 percent CI, −0.54 to −0.06; β for interaction, −0.50, 95 percent CI, −0.80 to −0.20; p=0.002). [JAMA Neurol 2024;doi:10.1001/jamaneurol.2024.3733]

“The deterioration in cognitive performance was mostly for executive dysfunction, a result that was consistent with accelerated volume loss to frontal lobe areas,” the investigators noted.

Secondary analysis of individual tests of the NTB showed that deferiprone-treated patients exhibited a more pronounced decline in executive function (β for interaction range, −0.64 to −0.35; p<0.001 for the Delis-Kaplan Executive Function System; p=0.02 for the WAIS-IV Coding Subtest) than in memory (β for interaction range, −0.19 to –0.12; p=0.34 for the Cogstate International Shopping List Test; p=0.69 for the Cogstate Groton Maze Learning Test) and attention (β for interaction range, −0.54 to –0.10; p=0.01 for the Cogstate Identification test; p=0.72 for the Cogstate Detection test).

At 12 months, quantitative susceptibility mapping data confirmed that compared with placebo, deferiprone led to decreased iron levels in the hippocampus (−0.36 vs 0.32 ppb; β for interaction, −0.68, 95 percent CI, −1.27 to −0.09; p=0.03). More importantly, while the hippocampal volume did not significantly differ between the treatment groups, the deferiprone group had greater brain volume loss in frontal areas, particularly in the insula, lateral orbitofrontal, medial orbitofrontal, and pars triangularis cortices.

“[The] increased brain volume loss in the frontal regions caused by deferiprone in the current trial is consistent with accelerated disease progression, not just symptom exacerbation, although these results are from an exploratory analysis with lower patient numbers than in the main outcome measure,” the investigators explained.

Overall, the present study challenges the notion that iron accumulation in the brain contributes to AD progression, the investigators said. Contrary to expectations, lowering brain iron levels with deferiprone led to worsened cognition and accelerated brain atrophy, suggesting a more complex role of iron in AD, they added. [JAMA Neurol 2017;74:122-125; Mol Psychiatry 2020;25:2932-2941; Alzheimers Dement 2021;17:1244-1256; J Neurol Neurosurg Psychiatry 2023;94:211-219]

“Iron sequestration in AD pathology has been proposed to cause functional iron deficiency. Iron trapped within the AD proteinopathy might be biologically important, given that participants who received active amyloid β immunization with AN1792 had iron redistribution into neurons in brain areas cleared of plaque… A forfeiture of bioavailable iron might contribute to metabolic failure in AD,” the investigators pointed out. [Brain Sci 2023;13:511; Brain Commun 2022;4:fcac021]

No anaemia despite iron loss

Of the patients included in the study, 53 were randomly assigned to the deferiprone group (mean age 73.0 years, 54.7 percent male) and 28 to the placebo group (mean age 71.6 years, 60.7 percent). A total of 54 patients completed the study, with seven (25.0 percent) and 20 (37.7 percent) withdrawing from the placebo and deferiprone groups, respectively.

Serious adverse events (AEs) were reported, with four deferiprone-treated patients (7.5 percent) having neutropenia that was deemed to be probably or possibly related to the study drug.

“Deferiprone caused a marked reduction in plasma ferritin levels (102.9 μg/L) and decreased QSM with large effects to several brain regions. The frequency of neutropenia [in the present study] was higher than in a Parkinson disease trial (1.6 percent) and a meta-analysis of deferiprone trials (4.4 percent),” the investigators noted.  [N Engl J Med 2022;387:2045-2055; Am J Hematol 2016;91:1026-1031]

“However, no patient developed agranulocytosis. While one participant taking deferiprone developed anaemia, deferiprone did not otherwise lower haemoglobin in our study, indicating that the degree of iron loss was not sufficient to cause anaemia in most patients,” they added.