Brain atrophy is a prominent pathological feature of AD, considered to be caused by neuronal loss and correlating with clinical disease manifestation. However, there are no circulating biomarkers that accurately identify cell death in the brain. The identification of such biomarkers is a major unmet need that could revolutionize the understanding, diagnosis and monitoring of AD.

It is well established that blood contains short-lived nucleosome-size fragments of circulating, cell-free DNA (cfDNA) derived from cells that have recently died. Analysis of cfDNA is already being used to identify mutations in fetuses, to monitor tumor dynamics and to detect graft rejection. These applications rely on genetic differences between the host and the tissue of interest, and are therefore blind to cfDNA derived from dying cells with normal genome, as occurs in AD.

We have developed an approach to identify the tissue origins of cfDNA, using cell type-specific DNA methylation patterns. Methylation patterns are highly specific to individual cell types, conserved among individuals and in pathologies, and preserved on cfDNA. In extensive preliminary studies we have identified multiple brain-specific methylation markers (including neuronal, oligodendrocyte and astrocyte-specific markers), and developed a method to identify these in cfDNA derived from plasma, based on PCR on bisulfite-converted DNA followed by next generation sequencing. We also demonstrated the presence of elevated brain-derived cfDNA in the plasma of people with brain damage, including highly encouraging data from plasma of people with early AD.

We propose to establish a method for highly sensitive, specific and accurate detection of death of specific cell types in the brain, based on specific methylation patterns of cfDNA, and to test it method in 750 prospective plasma samples from people with early AD and healthy controls. This novel type of biomarker may transform research and diagnosis of AD.