Subjective cognitive decline
Why studying SCD in the context of Alzheimer’s disease?
One of the main challenges in AD research is to detect the disease as early as possible in order to intervene when the neurodegenerative processes are still limited – and hopefully stoppable or even reversible. The field has thus progressively moved towards the identification of the earliest manifestations of the disease.
In this context, individuals who present with a subjective cognitive decline (SCD), although not showing any abnormal score on a classical neuropsychological examination, have become a population of high interest. Indeed, multiple longitudinal studies have been published in the last decade, showing that these individuals have a higher risk of developing dementia compared to their non-complaining peers (Reisberg et al, 2010; Jessen et al, 2014; Kaup et al, 2015). As a consequence, SCD is now considered as a concept of interest to investigate preclinical AD.
However, individuals with SCD constitute a heterogeneous population: not all older individuals with a cognitive complaint will develop AD. This relatively recent field of research is also heterogeneous itself, because of the variety of approaches used to define and study SCD. Indeed, SCD has been assessed in volunteers from the community or more rarely from population-based samples; in these cases, a variety of questions or questionnaires have been used to quantify SCD, as highlighted by a review from the SCD-Initiative (Rabin et al, JAD 2015). Other studies have used a different approach, specifically assessing patients recruited from a memory clinic that is, patients who sought help because of SCD. It is then of very high importance to further our understanding the drivers and correlated of SCD in older adults, and potentially to determine relevant characteristics of SCD that are associated with preclinical AD versus other etiologies (eg. normal aging, anxiety, or other neurological and psychological conditions, etc).
Our work on SCD and awareness of memory deficits
To reach this goal, we took advantage of the unique dataset obtained in the IMAP+ project, including cognitively normal older adults recruited from both the community and from a memory clinic (which they attended because of SCD) for whom state-of-the-art AD imaging biomarkers were measured. We have also contributed to, and benefited from, an international initiative on SCD (SCD-Initiative Working Group) established to develop a conceptual framework and research criteria for SCD (Jessen et al., Alz&Dem 2014).
In line with our main goals, we have specified the relationships between SCD and the presence of AD biomarkers in cognitively normal individuals. Particularly, we showed that, the severity of self-reported cognitive difficulties as assesses with an auto-questionnaire was associated with an increased prevalence of amyloid deposition (Perrotin, La Joie et al., Alz&Dem in press), in line with the original data acquired from the Berkeley Aging Cohort at UC Berkeley (Perrotin et al., Arch Neurol 2012). In addition, an AD pattern of brain atrophy, involving the hippocampus, was specifically found in individuals recruited from a memory clinic compared to SCD recruited from the community (Perrotin, La Joie et al., Alz&Dem in press). In a more specific analysis of structural MRI, we showed that these medical help seeking patients showed a profile of hippocampal subfield atrophy similar to that observed in AD (see section on hippocampal subfields) and different from cognitively intact elderly (Perrotin et al., JAD 2015).
In line with our objective of better discriminating SCD due to AD versus other etiologies (normal aging, anxiety…), we wondered if certain types of self-reported cognitive difficulties were specifically related to AD biomarkers or mild cognitive deficits in non-demented older adults. In this perspective we analyzed each item of our SCD questionnaires and looked for associations with amyloid imaging and clinical features (medical help seeking, diagnosis of MCI). We showed that, in cognitively normal individuals recruited from the community, the presence of amyloid was associated with a multidomain rather than a memory-specific SCD. In patients recruited from the clinic, endorsement of temporal disorientation items was associated with the presence of both detectable memory deficits and amyloid deposition (La Joie et al., Alz&Dem:DADM 2016).
Moving further down the course of Alzheimer’s disease, we also assessed SCD in patients at the dementia stage of the disease. Notably, we were interested in how self-reported cognitive difficulties related to actual measures of cognition function, and how this association varied with cognitive and brain integrity. Using multimodal neuroimaging, we showed that anosognosia (the impaired awareness of one’s own memory deficits) was related to a functional disconnection between midline cortical regions and the medial temporal lobe, areas involved in both memory processes and self-representations (Perrotin et al., Ann Neurol 2015).
Ongoing studies in the lab focus on the comparison between self- and informant-reports of cognitive difficulties, to assess their relative and combined contribution to the prediction of the presence of abnormal AD biomarkers across different clinical stages. In addition, we are using the longitudinal data acquired in the IMAP study to assess how SCD at baseline predicts cognitive decline and brain alterations over a 36-month follow up.