Immunisation of patients with Alzheimer's disease with full-length amyloid-beta peptide (Abeta(42)) can clear amyloid plaques from the brain. Our aim was to assess the relation between Abeta(42) immune response, degree of plaque removal, and long-term clinical outcomes.In June, 2003, consent for long-term clinical follow-up, post-mortem neuropathological examination, or both, was sought from 80 patients (or their carers) who had entered a phase I randomised, placebo-controlled trial of immunisation with Abeta(42) (AN1792, Elan Pharmaceuticals) in September, 2000. The follow-up study was completed in September, 2006. Plaques were assessed in terms of the percentage area of the cortex with Abeta immunostaining (Abeta load) and in terms of characteristic histological features reflecting plaque removal. Survival of all 80 individuals until severe dementia or death was assessed with a Cox proportional hazard model.20 participants--15 in the AN1792 group, five in the placebo group--died before follow-up started. A further 22 patients--19 in the AN1792 group, three in the placebo group--died during follow-up. Nine of the deceased patients, all in the AN1792 group, had given consent for post-mortem analysis; one of these who did not die with Alzheimer's disease was excluded. In the remaining eight participants who received immunisation and who were examined neuropathologically, mean Abeta load was lower than in an unimmunised control group that was matched for age at death (2.1% [SE 0.7] in treated participants vs 5.1% [0.9] in controls; mean difference 3.0%, 95% CI 0.6-5.4; p=0.02). Although there was considerable variation in Abeta load and degree of plaque removal among immunised participants, the degree of plaque removal varied significantly with mean antibody response attained during the treatment study period (Kruskal-Wallis p=0.02). Seven of the eight immunised patients who underwent post-mortem assessment, including those with virtually complete plaque removal, had severe end stage dementia before death. In the whole cohort, there was no evidence of improved survival (hazard ratio 0.93, 95% CI 0.43-3.11; p=0.86) or of an improvement in the time to severe dementia (1.18, 0.45-3.11; p=0.73) in the AN1792 group versus the placebo group.Although immunisation with Abeta(42) resulted in clearance of amyloid plaques in patients with Alzheimer's disease, this clearance did not prevent progressive neurodegeneration.