187 Summary and general discussion Clinical implications / implications for future research In this thesis, we show biomarkers provide valuable predictive information for cognitively normal individuals with SCD who present at a memory clinic. Individuals presenting at a memory clinic form an especially clinically relevant population, because they actively seek help and an answer to their worries and questions. The construct of SCD might become increasingly important for clinicians, because the number of individuals with subjective cognitive concerns who seek medical help is increasing. Individuals with SCD often have many questions or are worried about their risks of future dementia. Up until recently, not much was known about these early stages in which individuals were still cognitively normal, and clinicians did not always know the answers to their questions. With this thesis, we show that abnormality of biomarkers in individuals with SCD is not a benign finding, and biomarkers can be used to identify those at risk. It still needs to be elucidated, however, whether these results can be directly translated to other populations of SCD. For example, individuals who seek medical help more often have abnormal amyloid values (i.e. more often have preclinical AD) than individuals with SCD in the general population, who do not seek medical help (4, 8, 31-33). Future research should compare these populations and focus on the generalizability of the findings. The chapters in this thesis consistently show that biomarkers are associated with risks of cognitive decline or clinical progression on a group level. Ultimately, however, individuals presenting to a physician would want to know the specific risks of their own situation. There are two main issues that need to be addressed for a future in which individualized risk profiling is possible. First, studies should investigate larger populations with a longer follow-up duration. Additionally, as we showed in chapter 3 for amyloid burden specifically, dichotomization could result in loss of information, since information lies in the whole range of data points. As an alternative, future studies might focus on biomarkers as continuous variables, which could make the estimation of individual risks more accurate. Second, in addition to future studies focusing on more accurate predictions, they should consider the practical implications of early detection of biomarker abnormalities. Many physicians struggle with the disclosure of biomarker information. It could lead to distress when an individual hears they have abnormal biomarkers consistent with AD, since no cure is available yet, nor can an accurate prognosis be given yet. However, some find it unethical to withhold information that an individual could use to make plans for the future or to make changes in their lifestyle. Studies investigating this topic show individuals themselves also have ambivalent views regarding these topics (34). Studies investigating this topic will facilitate shared and informed decision making. 8