In search of an anti-inflammatory drug for Alzheimer disease

Evidence suggests that chronic neuroinflammation has an important role in the pathogenesis of Alzheimer disease. However, in a new clinical trial, the tetracycline antibiotic minocycline, which has anti-inflammatory properties, failed to delay disease progression in individuals with mild Alzheimer disease.

Chronic microglial activation occurs in many neurodegenerative diseases, including chro nic traumatic encephalopathy, amyotrophic lateral sclerosis, Parkinson disease (PD) and Alzheimer disease (AD). In the brains of patients with AD, this microglial activa tion is observed at all stages of the disease and is accompanied by increased levels of proinflammatory mediators, such as tumour necrosis factor (TNF), IL1β, IL6, prostag landins, reactive oxygen species and reactive nitrogen species 1 . Furthermore, genomewide association studies (GWAS) have identified associations between vari ants in inflammationrelated genes, includ ing the gene that encodes triggering receptor expressed on myeloid cells 2 (TREM2) and the risk of developing AD 2,3 . TREM2 is pres ent on the surface of microglial cells and responds to damageassociated molecular patterns (DAMPs) and bacterial lipopoly saccharides by activating phago cytosis and promoting microglial survival; the link between TREM2 and AD is considered to be one of the strongest arguments for a role of neuroinflammation in the disease 3 . These genetic and histological findings suggest that progression of AD is driven -at least partly -by a selfperpetuating cycle of inflam matory neurotoxicity. In this cycle, injured neurons release DAMPs 4 , which acti vate microglia. The activated microglia then secrete neurotoxic cytokines (for example, TNF) and free radicals (for example, super oxide with the Standardized MiniMental State Examination (sMMSE), and patients' ability to carry out basic activities of daily life, such as eating, dressing or taking medication, was assessed with the Bristol Activities of Daily Living Scale (BADLS). No statistically signif icant differences were seen between partici pants receiving minocycline and participants receiving placebo. However, a closer look at the data reveals that the mean change in sMMSE score over 24 months was 4.1 points in the group that received placebo and 3.3 points in the group that received 400 mg of minocycline, which suggests that deterio ration was ~20% slower among participants who received 400 mg minocycline than among those who received placebo (Fig. 1). However, at the 400 mg dose, the adverse effects of minocycline were severe and only 28% of participants in this group completed the study.
Howard and colleagues did a diligent job of discussing possible explanations for the negative outcome. The first explanation given in the paper is that neuroinflammation "may be a reaction to pathologic character istics of the disease rather than an impor tant factor in neurodegeneration". However, results of GWAS indicate that >60% of the genes linked to lateonset sporadic AD are inflammationrelated, which strongly and nitric oxide), which damage neurons, thus beginning the cycle again. Consequ ently, treatments that target chronic neuro inflammation might be able to modify the course of AD.
In a new clinical trial published in JAMA Neurology, Howard et al. 5 tested the efficacy of the tetracycline antibiotic minocycline for the treatment of mild AD. Minocycline was chosen for the trial on the basis of two sys tematic reviews that identified the drug as a highpriority candidate for repurposing as a treatment for AD. The drug has a variety of offtarget effects, including immuno suppression, and is therefore already used as a secondline agent for the management of rheumatoid arthritis. Evidence from animal studies indicates that minocycline reduces neutrophilmediated tissue injury via inhibi tion of neutrophil migration and degranu lation, as well as suppression of oxygen radi cal formation 6 . In animal models of familial AD, minocycline has antiinflammatory prop erties; for example, it reduces levels of IL1β, TNF, IL4 and IL10 (reF. 7 ). In these preclini cal studies, minocycline has been tested at a daily dose of ~50 mg/kg. This translates into a dose of 400 mg for the average person, which is considerably higher than the 200 mg that is given to treat infections.
In the trial by Howard et al. 5 , partici pants received a daily dose of either 200 mg minocycline, 400 mg minocycline or pla cebo. Cognitive performance was measured

Erika Gyengesi and Gerald Münch
Evidence suggests that chronic neuroinflammation has an important role in the pathogenesis of Alzheimer disease. However, in a new clinical trial, the tetracycline antibiotic minocycline, which has anti-inflammatory properties, failed to delay disease progression in individuals with mild Alzheimer disease. suggests a role for inflammation in AD patho genesis. Furthermore, a large, retrospective case-control study of the electronic health records of 56 million individuals indicated that TNFblocking agents reduce the risk of AD in patients with rheumatoid arthritis and psoriasis 8 . For example, patients with rheumatoid arthritis who were being treated with etanercept, adalimumab or infliximab were less likely to be diagnosed with AD than patients with untreated rheumatoid arthritis.
The second potential explanation for the failure of the trial is that, because the asso ciation between microglial activation and neurodegeneration is complex, minocycline might be having unexpected effects and could even interfere with the supportive function of microglia. One of the biggest problems with minocycline is that it was not designed to be an antiinflammatory drug, and its exact tar get and mode of action are not entirely clear. Whether the doses of minocycline given in the trial by Howard et al. 5 can decrease levels of the relevant proinflammatory cytokines, radicals and neurotoxins is not yet known.