Yoro-Zohoun et al. (Reference Yoro-Zohoun2021) investigated the association between the apolipoprotein E (APOE) ɛ4 allele and neuropsychiatric symptoms (NPS) in 322 older adults from Central Africa. The sample had a high prevalence of the APOE ɛ4 allele, with 40% of older adults having at least one ɛ4 allele. Interestingly, participants who were homozygous APOE ɛ4 allele carriers had less NPS in the previous 30 days compared to non-carriers in an analysis adjusted for sociodemographic variables and cognitive status. Non-significant trends were also found in participants who were heterozygous for the APOE ɛ4 allele, suggesting lower odds of NPS among these participants compared to non-APOE ɛ4 allele carriers.
Based on studies in white populations, the APOE ɛ4 allele is the strongest genetic risk factor for Alzheimer’s disease (AD), with evidence suggesting that APOE status influences β-amyloid aggregation and clearance (Kim et al., Reference Kim, Basak and Holtzman2009). Although the APOE ɛ4 allele was shown to be a risk factor for AD among individuals with African ancestry in a recent large, genome-wide association study (Kunkle et al., Reference Kunkle2020), the effect of the APOE ɛ4 allele on dementia risk among these individuals seems to be weaker than in white individuals (Barnes and Bennett, Reference Barnes and Bennett2014). The study by Yoro-Zohoun et al. (2021) reports that 40% of the 322 older adults in a sample from Central African had at least one ɛ4 allele, which is similar to that found by Evans et al. in the USA among black participants (42%) but higher than among white participants (27%) (Evans et al., Reference Evans2003). However, African Americans showed a weaker association between the APOE ɛ4 allele and cognitive decline and AD than that found in the white participants (Evans et al., Reference Evans2003). Moreover, in a study comparing this association among African Americans and Yoruba, the association between the APOE ɛ4 allele with AD incidence and cognitive decline was weaker in the Yoruba people than in the African Americans (Hendrie et al., Reference Hendrie2014). The frequency of β-amyloid accumulation seems to be lower in individuals with African ancestry, suggesting that the weaker effect of APOE on cognitive outcomes could be mediated by lower β-amyloid deposition (Schlesinger et al., Reference Schlesinger2013). In line with this finding, compared to white participants, lower concentrations of AD biomarkers have been found in the cerebrospinal fluid of African Americans who had at least one APOE ɛ4 allele compared to non-Hispanic whites and African Americans without ɛ4 alleles (Morris et al., Reference Morris2019).
NPSs are common during the course of dementia and are an important source of burden for patients with dementia and their caregivers (Sink et al., Reference Sink, Holden and Yaffe2005). In older people without dementia, some, but not all NPS, seem to be associated with dementia incidence and cognitive decline (Brodaty et al., Reference Brodaty2012). In a study that included 873 community dwelling adults aged 70–90 years in Australia, only baseline depression was associated with dementia at follow-up and only anxiety and agitation were significantly associated with cognitive decline, while overall rates of NPS at baseline were not associated with mild cognitive impairment (MCI), dementia, or cognitive decline over the 2 years follow-up (Brodaty et al., Reference Brodaty2012). In Mexico, however, a study conducted with 1355 older adults (Acosta et al., Reference Acosta, Borges, Aguirre-Hernandez, Sosa and Prince2018) found that five NPSs (delusions, hallucinations, anxiety, aberrant motor behavior and depression) were associated with dementia incidence over a period of 3 years follow-up and the risk increased with the increased number of symptoms. These increased risks may be influenced by APOE ɛ4. Pink et al. (Reference Pink2015) found that among 332 MCI cases, baseline agitation, nighttime behaviors, depression, and apathy elevated the risk of incident dementia, and there was a synergistic interaction of depression and apathy with APOE ɛ4 in further elevating the risk of incident dementia.
The investigation of risk factors for NPS is essential to promote patient and caregiver well-being, and the hypothesis that the APOE ɛ4 allele could be a risk factor for NPS in older populations is reasonable. However, current evidence on the association between NPS and the APOE ɛ4 allele is mixed (Panza et al., Reference Panza2012), and most studies have been conducted in white populations with clinical dementia. Studies on individuals with African ancestry and in cognitively normal individuals are scarce. Yoro-Zohoun et al. investigated the association between APOE ɛ4 and NPS and found less participants with at least one NPS in the last 30 days among those who were homozygous for the ɛ4 allele, and a trend of less participants with at least one NPS in those who were heterozygous for the APOE ɛ4 allele (Yoro-Zohoun et al., Reference Yoro-Zohoun2021). It is worth noting that the presence of NPS was defined by Yoro-Zohoun et al. as any score greater than zero, and the severity and frequency of NPS were not taken into consideration in the analyses.
The study by Yoro-Zohoun et al. not only provides valuable information on an African population but also has other distinct features compared to previous studies. For example, 91% of the sample had no formal education (Yoro-Zohoun et al., Reference Yoro-Zohoun2021). This low educational level adds complexity to cognitive evaluations as performance in most tests is impaired by low educational attainment, which tends to be lower in racial and ethnic minorities. Indeed, although a high prevalence of dementia has been found among African Americans, they have shown a slower cognitive decline rate, a lower incidence rate of dementia, and longer survival than non-Hispanic whites (Mayeda et al., Reference Mayeda, Glymour, Quesenberry, Johnson, Perez-Stable and Whitmer2017; Weuve et al., Reference Weuve2018). This suggests that dementia may have been over-diagnosed in African Americans due to cross-sectional evaluations using cognitive tests that are not suitable for settings where low education is common (Barnes and Bennett, Reference Barnes and Bennett2014). In the same way, caregivers with low education may have more difficulty in understanding and responding to the Neuropsychiatry Inventory (Camozzato et al., Reference Camozzato, Godinho, Kochhann, Massochini and Chaves2015), the same instrument used in the Yoro-Zohoun et al. study. Therefore, some degree of measurement error in the study is to be expected, and longitudinal evaluations in the EPIDEMCA study may minimize measurement error regarding the NPS evaluation. Moreover, NPSs were evaluated regarding their presence in the last 30 days (Yoro-Zohoun et al., Reference Yoro-Zohoun2021), whereas evaluation of major neuropsychiatric disorders during the life course, not only over a short period of time, is needed to give more robust information to allow comparisons to be made with the APOE genotype.
Another important aspect to highlight is that 57% of the sample were cognitively normal. It is important to examine the association between the APOE e4 allele and NPS stratified by cognitive status. The APOE e4 allele would be expected to be a risk factor for NPS in participants with MCI and dementia. Yoro-Zohoun et al. found a trend toward a higher risk of delusion among APOE ɛ4 allele carriers in the whole sample (Reference Yoro-Zohoun2021). Delusion is most common in older adults with advanced dementia, and a stratified analysis by cognitive status might have shown different directions for the association between the APOE ɛ4 allele and NPS depending on the participant’s cognitive status.
The prevalence of at least one NPS was high (60%) in this study. Depression (39%), anxiety (27%), and irritability (23%) were the most frequent NPS in the 30 days before the evaluation. These symptoms are common in older adults and are probably more related to environmental factors than genes such as APOE (Fiske et al., Reference Fiske, Wetherell and Gatz2009; Lotrich, Reference Lotrich2011; Petkus et al., Reference Petkus, Gatz, Reynolds, Kremen and Wetherell2016). Clinical comorbidities, caregiver stress, death of family members and friends, traumas during the life course, and changes in socioeconomic status are common during old age and are important triggers of neuropsychiatric disorders (Fiske et al., Reference Fiske, Wetherell and Gatz2009). Although symptoms of depression, anxiety, and irritability can also be present in patients with mild and moderate dementia, it is likely that most participants with these symptoms are cognitively normal.
Finally, we would like to highlight the possible association of the APOE ɛ2 allele with cognitive and neuropsychiatric outcomes that was not investigated in the current study. There has been a significant amount of research into the evaluation of the ɛ4 allele as a risk factor for AD, but less attention has been given to the potential protective effect of the APOE ɛ2 allele on dementia risk (Li et al., Reference Li, Shue, Zhao, Shinohara and Bu2020). The EPIDEMCA sample had a higher frequency of APOE ɛ2 allele (24%) compared to 17% in non-Hispanic whites and 19% in African Americans (Evans et al., Reference Evans2003). Therefore, it would be interesting to investigate the relationship between the APOE ɛ2 allele and NPS in the EPIDEMCA study.
We congratulate Yoro-Zohoun et al. for the study and look forward to future longitudinal studies that investigate the associations between the APOE genotype and cognitive and neuropsychiatric outcomes in Central Africa.
You have
Access
Yoro-Zohoun et al. (Reference Yoro-Zohoun2021) investigated the association between the apolipoprotein E (APOE) ɛ4 allele and neuropsychiatric symptoms (NPS) in 322 older adults from Central Africa. The sample had a high prevalence of the APOE ɛ4 allele, with 40% of older adults having at least one ɛ4 allele. Interestingly, participants who were homozygous APOE ɛ4 allele carriers had less NPS in the previous 30 days compared to non-carriers in an analysis adjusted for sociodemographic variables and cognitive status. Non-significant trends were also found in participants who were heterozygous for the APOE ɛ4 allele, suggesting lower odds of NPS among these participants compared to non-APOE ɛ4 allele carriers.
Based on studies in white populations, the APOE ɛ4 allele is the strongest genetic risk factor for Alzheimer’s disease (AD), with evidence suggesting that APOE status influences β-amyloid aggregation and clearance (Kim et al., Reference Kim, Basak and Holtzman2009). Although the APOE ɛ4 allele was shown to be a risk factor for AD among individuals with African ancestry in a recent large, genome-wide association study (Kunkle et al., Reference Kunkle2020), the effect of the APOE ɛ4 allele on dementia risk among these individuals seems to be weaker than in white individuals (Barnes and Bennett, Reference Barnes and Bennett2014). The study by Yoro-Zohoun et al. (2021) reports that 40% of the 322 older adults in a sample from Central African had at least one ɛ4 allele, which is similar to that found by Evans et al. in the USA among black participants (42%) but higher than among white participants (27%) (Evans et al., Reference Evans2003). However, African Americans showed a weaker association between the APOE ɛ4 allele and cognitive decline and AD than that found in the white participants (Evans et al., Reference Evans2003). Moreover, in a study comparing this association among African Americans and Yoruba, the association between the APOE ɛ4 allele with AD incidence and cognitive decline was weaker in the Yoruba people than in the African Americans (Hendrie et al., Reference Hendrie2014). The frequency of β-amyloid accumulation seems to be lower in individuals with African ancestry, suggesting that the weaker effect of APOE on cognitive outcomes could be mediated by lower β-amyloid deposition (Schlesinger et al., Reference Schlesinger2013). In line with this finding, compared to white participants, lower concentrations of AD biomarkers have been found in the cerebrospinal fluid of African Americans who had at least one APOE ɛ4 allele compared to non-Hispanic whites and African Americans without ɛ4 alleles (Morris et al., Reference Morris2019).
NPSs are common during the course of dementia and are an important source of burden for patients with dementia and their caregivers (Sink et al., Reference Sink, Holden and Yaffe2005). In older people without dementia, some, but not all NPS, seem to be associated with dementia incidence and cognitive decline (Brodaty et al., Reference Brodaty2012). In a study that included 873 community dwelling adults aged 70–90 years in Australia, only baseline depression was associated with dementia at follow-up and only anxiety and agitation were significantly associated with cognitive decline, while overall rates of NPS at baseline were not associated with mild cognitive impairment (MCI), dementia, or cognitive decline over the 2 years follow-up (Brodaty et al., Reference Brodaty2012). In Mexico, however, a study conducted with 1355 older adults (Acosta et al., Reference Acosta, Borges, Aguirre-Hernandez, Sosa and Prince2018) found that five NPSs (delusions, hallucinations, anxiety, aberrant motor behavior and depression) were associated with dementia incidence over a period of 3 years follow-up and the risk increased with the increased number of symptoms. These increased risks may be influenced by APOE ɛ4. Pink et al. (Reference Pink2015) found that among 332 MCI cases, baseline agitation, nighttime behaviors, depression, and apathy elevated the risk of incident dementia, and there was a synergistic interaction of depression and apathy with APOE ɛ4 in further elevating the risk of incident dementia.
The investigation of risk factors for NPS is essential to promote patient and caregiver well-being, and the hypothesis that the APOE ɛ4 allele could be a risk factor for NPS in older populations is reasonable. However, current evidence on the association between NPS and the APOE ɛ4 allele is mixed (Panza et al., Reference Panza2012), and most studies have been conducted in white populations with clinical dementia. Studies on individuals with African ancestry and in cognitively normal individuals are scarce. Yoro-Zohoun et al. investigated the association between APOE ɛ4 and NPS and found less participants with at least one NPS in the last 30 days among those who were homozygous for the ɛ4 allele, and a trend of less participants with at least one NPS in those who were heterozygous for the APOE ɛ4 allele (Yoro-Zohoun et al., Reference Yoro-Zohoun2021). It is worth noting that the presence of NPS was defined by Yoro-Zohoun et al. as any score greater than zero, and the severity and frequency of NPS were not taken into consideration in the analyses.
The study by Yoro-Zohoun et al. not only provides valuable information on an African population but also has other distinct features compared to previous studies. For example, 91% of the sample had no formal education (Yoro-Zohoun et al., Reference Yoro-Zohoun2021). This low educational level adds complexity to cognitive evaluations as performance in most tests is impaired by low educational attainment, which tends to be lower in racial and ethnic minorities. Indeed, although a high prevalence of dementia has been found among African Americans, they have shown a slower cognitive decline rate, a lower incidence rate of dementia, and longer survival than non-Hispanic whites (Mayeda et al., Reference Mayeda, Glymour, Quesenberry, Johnson, Perez-Stable and Whitmer2017; Weuve et al., Reference Weuve2018). This suggests that dementia may have been over-diagnosed in African Americans due to cross-sectional evaluations using cognitive tests that are not suitable for settings where low education is common (Barnes and Bennett, Reference Barnes and Bennett2014). In the same way, caregivers with low education may have more difficulty in understanding and responding to the Neuropsychiatry Inventory (Camozzato et al., Reference Camozzato, Godinho, Kochhann, Massochini and Chaves2015), the same instrument used in the Yoro-Zohoun et al. study. Therefore, some degree of measurement error in the study is to be expected, and longitudinal evaluations in the EPIDEMCA study may minimize measurement error regarding the NPS evaluation. Moreover, NPSs were evaluated regarding their presence in the last 30 days (Yoro-Zohoun et al., Reference Yoro-Zohoun2021), whereas evaluation of major neuropsychiatric disorders during the life course, not only over a short period of time, is needed to give more robust information to allow comparisons to be made with the APOE genotype.
Another important aspect to highlight is that 57% of the sample were cognitively normal. It is important to examine the association between the APOE e4 allele and NPS stratified by cognitive status. The APOE e4 allele would be expected to be a risk factor for NPS in participants with MCI and dementia. Yoro-Zohoun et al. found a trend toward a higher risk of delusion among APOE ɛ4 allele carriers in the whole sample (Reference Yoro-Zohoun2021). Delusion is most common in older adults with advanced dementia, and a stratified analysis by cognitive status might have shown different directions for the association between the APOE ɛ4 allele and NPS depending on the participant’s cognitive status.
The prevalence of at least one NPS was high (60%) in this study. Depression (39%), anxiety (27%), and irritability (23%) were the most frequent NPS in the 30 days before the evaluation. These symptoms are common in older adults and are probably more related to environmental factors than genes such as APOE (Fiske et al., Reference Fiske, Wetherell and Gatz2009; Lotrich, Reference Lotrich2011; Petkus et al., Reference Petkus, Gatz, Reynolds, Kremen and Wetherell2016). Clinical comorbidities, caregiver stress, death of family members and friends, traumas during the life course, and changes in socioeconomic status are common during old age and are important triggers of neuropsychiatric disorders (Fiske et al., Reference Fiske, Wetherell and Gatz2009). Although symptoms of depression, anxiety, and irritability can also be present in patients with mild and moderate dementia, it is likely that most participants with these symptoms are cognitively normal.
Finally, we would like to highlight the possible association of the APOE ɛ2 allele with cognitive and neuropsychiatric outcomes that was not investigated in the current study. There has been a significant amount of research into the evaluation of the ɛ4 allele as a risk factor for AD, but less attention has been given to the potential protective effect of the APOE ɛ2 allele on dementia risk (Li et al., Reference Li, Shue, Zhao, Shinohara and Bu2020). The EPIDEMCA sample had a higher frequency of APOE ɛ2 allele (24%) compared to 17% in non-Hispanic whites and 19% in African Americans (Evans et al., Reference Evans2003). Therefore, it would be interesting to investigate the relationship between the APOE ɛ2 allele and NPS in the EPIDEMCA study.
We congratulate Yoro-Zohoun et al. for the study and look forward to future longitudinal studies that investigate the associations between the APOE genotype and cognitive and neuropsychiatric outcomes in Central Africa.