Fluid biomarkers for frontotemporal lobar degeneration and related diseases — discovery and clinical validation

Abstract

Frontotemporal dementia (FTD) spectrum disease encompasses a wide range of progressive clinical syndromes, including behavioral changes, progressive language deterioration, sometimes in combination with motor neuron disease or parkinsonian syndromes. These syndromes are often associated with frontotemporal lobar degeneration (FTLD), a neurodegenerative condition defined by the loss of neurons in the frontal and temporal lobes of the brain and the presence of insoluble protein deposits. Although less common than Alzheimer’s disease (AD), it is a leading cause of early onset dementia. A significant proportion of individuals affected by FTD have a history of dementia among relatives, and up to a third have mutations in genes that are determinately causing the disease. There are currently no effective treatments for the disease, which may in part be due to large clinical and pathophysiological heterogeneity, which hampers development of effective drugs that target the underlying biology of the disease. Biomarkers – quantifiable measures reflecting biological processes – are likely to be important for clinical management, understanding disease evolution, and therapeutic development. The work in this thesis sought to expand the understanding of biomarkers for the FTD disease spectrum and move the field one step closer to biomarker-based management of these conditions. In paper I, we sought to develop reference limits for a blood-based biomarker of neuronal damage, neurofilament light (NfL), which has been shown to be of use in differential diagnostics where FTLD is suspected, in disease monitoring, and therapeutic trials. We developed age-stratified reference limits across the lifespan for plasma NfL based on its concentration in healthy individuals, facilitating its interpretation in clinical settings. In paper II, we sought to explore novel preanalytical techniques to simplify the handling of blood samples in remote settings. We found that delayed processing of blood samples, in which NfL was then measured, did not meaningfully change its stability. Further, “filter papers” (so called dried blood or plasma spots) were shown to be a complementary option. Both simplifications suggest the feasibility of remote testing in epidemiological studies and possibly to remotely monitor disease or treatments in the future. To expand the armamentarium of biomarkers reflecting FTLD pathophysiology, papers III and IV utilized untargeted mass spectrometry to explore all quantifiable proteins (“the proteome”) in the cerebrospinal fluid (CSF). In paper III, we found a range of proteins quantifiable in CSF discriminating different causes of genetic FTLD from non-mutation carriers, as well as distinct proteomic signatures in each genetic form. Some markers were also changed already in presymptomatic mutation carriers. In paper IV, we observed that many proteins previously identified in paper III – linked to synaptic dysfunction, neuronal loss, immune responses, and metabolic processes – were also altered in individuals with different FTLD subtypes. In paper IV, we additionally found that there were both overlapping, and distinct changes when comparing FTLD with other causes of progressive cognitive impairment, such as AD and Lewy body disease. Notably, FTLD subtypes were surprisingly similar in paper IV. In paper V, we investigated a novel method to measure ~120 proteins of potential relevance to neurodegenerative disease in plasma of individuals with genetic FTD. We found clear changes in established biomarkers, such as NfL, but also in less explored candidates, that should be subject to follow-up studies. Altogether, the work in this thesis expanded the utility of NfL in the context of FTLD and provided novel candidates for further research in biomarkers of FTLD pathophysiology, which may ultimately lead to a panel of clinically useful biomarkers able distinguish between FTLD forms and enable their treatment.