Cognitive impairment is a recognized effect of drug misuse, including the use of opiates. The pathological basis for this is unknown but the temporal and frontal cortices have been implicated. We have shown previously that deposits of hyperphosphorylated tau in drug user brains exceed those seen in age-matched controls. The present quantitative study of hyperphosphorylated tau and beta amyloid in drug user brains allows comparison with the related pathology in Alzheimer's disease. Brains were obtained from the Edinburgh Medical Research Council Brain Banks, comprising 39 human immunodeficiency virus negative drug users, five subjects with Alzheimer's disease and 37 age-matched, cognitively normal controls, all legally and ethically approved for research. Hyperphosphorylated tau positive (AT8, AT100) neuropil threads were significantly increased in the frontal and temporal cortex, and in the locus coeruleus, of drug users aged >30 years (all P = 0.04). Under the age of 30 years, drug users showed a similar increase in neuropil threads compared with controls, but this reached significance only in the frontal cortex (P = 0.03). Immunopositivity for both three- and four-repeat tau was present in drug user brains. There was a direct relationship between the numbers of neuropil threads and of neurofibrillary tangles: neurofibrillary tangles were sparse in brains that had neuropil thread counts below 200 cm(2). Hyperphosphorylated tau positive neuropil threads increased at a faster rate in drug users than in controls and the levels of the phosphorylating enzyme, GSK-3, was raised in drug user brains. Beta amyloid (AB4, AB42 and 4G8) was raised in drug user brains (mainly as shadow plaques) but not significantly different from controls and there was no correlation between high beta amyloid and hyperphosphorylated tau in individual cases. Hyperphosphorylated tau levels correlated significantly (P = 0.038) with microglial activation in drug users but not in controls. The levels of hyperphosphorylated tau in drug users fell far short of those seen in Alzheimer's disease but overlapped with those in elderly controls. We conclude that drug users show early Alzheimer's disease-related brain pathology that may be the basis for cognitive impairment and that neuroinflammation is an early accompanying feature. This provides an opportunity to study the pathogenesis of tau pathology in the human brain.