PLOS ONE is excited to publish the Early Diagnosis and Treatment of Alzheimer’s Disease Collection. As the global population ages, the impact and prevalence on Alzheimer’s disease is predicted to rise, making early diagnosis and treatment crucial. PLOS ONE launched a call for papers last year inviting submissions in this important area. This Collection showcases the submissions chosen for inclusion by the guest editors.
We received over 50 submissions, covering topics related to both the diagnosis and treatment of the disease using preclinical methods and clinical data.
Much of the preclinical work in the Collection provides insights that will be valuable for development of future therapeutics, with topics including gene delivery of modified antibodies and possible mechanisms for genes known to act as risk factors for Alzheimer’s. Further studies investigate the potential role of inflammatory responses as risk factors for the disease and the effects of modulation of the immune system in treating pre-symptomatic Alzheimer’s disease in an animal model.
These works are complemented by several clinical papers, many of which bring us closer to early detection of Alzheimer’s disease. This includes studies detecting Alzheimer’s disease with blood-based biomarkers and the correlation between Amyloid-β PET imaging results and cerebrospinal fluid biomarkers.
A number of the papers also make use of machine learning and other computational techniques to analyse genetic, PET imaging and MRI datasets, demonstrating how these can be used to improve detection of the disease.
At launch, the Collection includes 20 papers, with several more to be added as they reach publication. We invite you to revisit the Collection again to read the latest research on this topic.
To celebrate its launch the Guest Editors have selected some of their favourite papers featured in the Collection and provided a short summary.
Yona Levites, University of Florida, USA
This study by Elmer and Collegues assesses effects of single chain variable fragment coupled with mutated Fc domain retaining FcRn binding, but lacking Fc gamma receptor (FcγR) binding (silent scFv-IgG) on amyloid accumulation and aggregation in Alzheimer’s disease transgenic mouse model. The molecule has increased affinity to amyloid as compared to scFv, but toxicity. Interestingly, peripheral AAV delivery resulted in detectable levels of scFv-IgG in the brain. Intracranial injection of anti-amyloid scFv IgG resulted in significant reduction of amyloid burden in TgAPP mice. Future studies will potentially focus on improving the effects of silent scFv-IgG delivered peripherally.
Jussi Tohka, University of Eastern Finland, Finland
The current diagnostic criteria allow the use of Cerebro Spinal Fluid (CSF) biomarkers to provide pathophysiological support for the diagnosis of AD. The paper by Rhodius-Meester and colleagues argues that diagnostic guidelines do not specify which patients should receive CSF testing and the appropriate use criteria, which have been proposed as a guideline, are challenging to translate to clinical practice. Hence, this study evaluated a computerized decision support system to select patients for CSF biomarker (CSF beta-amyloid 1–42 (AB42), total tau and tau phosphorylated at threonine 181 (p-tau)) determination. In the computerized decision support approach of the study, diagnosis is first attempted using only neuropsychology, MRI and APOE data. Depending on the confidence of the first diagnosis, the clinician can decide either not to order CSF testing or visualize the effects of the simulated normal and AD-like CSF values on the workflow. Only if the potential change in the confidence of the diagnosis due to CSF is large enough, the clinician orders CSF testing. The study indicated that this approach could support clinicians in making a balanced decision in ordering additional CSF testing. Specifically, the computerized decision support approach of the study restricted CSF testing to only 26% of cases, without compromising diagnostic accuracy.
Roberta Diaz Brinton, The University of Arizona
The computational analysis by Potashkin, Bottero, Santiago and Quinn yielded new insights regarding gene networks affected in Alzheimer’s while also strengthening the role of metabolic networks in the pathophysiology of the disease. Each of the gene networks identified by this team, glucose homeostasis, glucocorticoid signaling, sleep regulation, and memory, are familiar to those working in the field. The critical insight derived from this analysis is that multiple gene networks are involved in Alzheimer’s and by extension that a single therapeutic target does not address the complexity of the disease. The gene network analyses were conducted on postmortem brain tissue and thus represent the brain in the end stages of the disease. A challenge for the field is to develop accessible and predictive biomarkers of earlier stages of the disease when therapeutic interventions can potentially delay or reverse the disease process. The findings provided by Potashkin and colleagues provide a platform on which to build just such predictive biomarkers.
Michael Weiner, University of California San Francisco, USA
In this paper comparing amyloid PET with Flutemetamol and measurements of Aβ42 in cerebrospinal fluid, Müller and co-workers found a significant correlation between 18F-Flutemetamol PET classification and the three CSF biomarkers. The highest correlation was between Aβ42 and 18F-Flutemetamol PET. Good correlations between CSF Aβ42 and amyloid PET have been previously reported in many studies. An “optimal cut-off value for Aβ42 “ was used and yielded an improvement in sensitivity, while maintaining a high specificity, for a positive 18F-Flutemetamol PET. 18F-Flutemetamol PET was found to be the best predictor of a clinical AD diagnosis. The authors stated that “whether biomarkers are to be included in the clinical criteria to further improve their sensitivity is still under investigation.”