Abstract
Alzheimer’s disease (AD) can be preceded by subtle memory decline that can last a decade or more before progressing to what would be diagnosed as the mild cognitive impairment stage. During this early stage of decline, individuals and even their caregivers can fail to perceive any serious difficulty or need to consult a physician. Herein, we present evidence in support of these concerns, and demonstrate how this can interfere not only with clinical trials of AD but also those involving cognitive performance of elderly individuals without intentional reference to AD.
Keywords
Alzheimer’s disease (AD) can be preceded by subtle memory decline that can last a decade or more before progressing to what would be diagnosed as the mild cognitive impairment stage [1]. During this early stage of decline, individuals and even their caregivers can fail to perceive any serious difficulty, and therefore will not consult a physician [2]. This unfortunately delays intervention, including that of relatively simple lifestyle modifications that could delay [3–9] or even reverse [10, 11] cognitive decline provided they are initiated sufficiently early.
We previously conducted a trial evaluating the impact of a nutraceutical formulation on cognitive performance of community-dwelling individuals over a large age range (18–86) all of whom self-reported no known nor suspected cognitive decline. Participants completed a test of simple memory (California Verbal Learning II) and executive function (the Trail-Making test) at baseline and following 3 months of randomization to treatment or placebo [12]. During initial screening of data, we noted that a number of individuals presented substantially declined in performance on the part B of the Trail-making test (by 111–146 seconds) at 3 months versus their respective baseline performance. These individuals, all of whom were randomized to the treated cohort, were 76, 77, 79, and 82 years of age. Notably, this level of decline was not typical of the total cohort of 27 participants >74 years of age. This decline was specifically in part B of the test (in which participants must connect circles alternating in numbers and letters; i.e., 1, A, 2, B, 3, C, 4, D, etc.) and therefore requires intact executive function. A similar extent of decline was not observed in part A (in which participants connect circles containing only numbers) and therefore is based on vision, hand-eye coordination, and neuromuscular control. These findings indicate a specific decline in cognitive performance rather than age-related decline in physical/motor function. An additional two participants in this cohort declined in Trails B by 42 and 44 seconds, respectively. No such marked decline in this test were observed in the cohort randomized to placebo. We elected not to classify these individuals as outliers, and therefore made no overall conclusion regarding participants >74 years of age. Rather, we only noted that those individuals >74 years of age that did improve displayed the same extent of improvement of individuals <74 years of age [12].
During recent retrospective analyses of participant performance, we noted further differences between the elderly (>74 years of age) versus younger (<74 years of age) cohorts. The change in performance for individuals <74 years of age after 3 months of treatment ranged from –38 to +38 seconds, while for those >74 years of age ranged from –38 to +146 seconds (a 2.4x greater range compared to that of individuals <74 years of age). Following randomization, the treatment and placebo cohorts <74 years of age were statistically identical in terms of gender, education level, and baseline scores on the Trail-making test. By contrast, the treatment and placebo >74 years of age were identical in terms of gender and education level, but baseline scores on the Trail-making test were statistically lower in the cohort randomized to nutraceutical formulation versus placebo. As described in Chan et al. [12], those individuals >74 years of age that were classified as responders (i.e., who improved in the Trails test) improved to the same extent as participants≤74 years of age, indicating that neither the significant decline of the participants listed above nor the wide range in baseline scores were characteristic of the entire cohort of aged participants. Moreover, if the individuals listed above were considered as outliers, the remainder of individuals in the treated cohort >74 years of age completed the Trail-making test 9.7±6.1 seconds faster (mean±standard error); while this value did not differ statistically from that of the placebo cohort (which completed the test 3.1±5.1 seconds faster), it was comparable to the extent of improvement of that of participants≤74 years of age (7.0±2.1 seconds faster) [12].
Based on the wide range of scores of >74 aged cohort in Chan et al. [12], this group likely consists of a diverse group of individuals with respect to cognitive performance: some with cognitive performance considered normal for their respective age, some with impending dementia, and some who underwent demonstrable decline in performance during our analyses. Since these participants self-reported no known nor suspected cognitive decline, their respective cognitive difficulties likely did not pervade day-to-day performance, and were revealed only by the Trail-making test. In support of this possibility, the four individuals displaying extreme decline in the Trail-making test did not display a similar decline in simple word recall on the California Verbal Learning tests. This possibility is further supported by the lack of a physician’s diagnosis confirming normal cognitive performance for their respective age.
It is further noteworthy that statistically-significant improvement with this same treatment was observed in phase II studies for individuals diagnosed with AD and, separately, mild cognitive impairment [13, 14]; the requirement of diagnosis was an inclusion criterion in these studies, which avoided the diversity of cognitive performance, and unanticipated decline, observed with the elderly cohort in Chan et al. [12].
This differential performance between participants <74 and >74 years of age in Chan et al. is consistent with the frequency of AD; clinical cognitive decline is rare among individuals <65 years of age, is present in 3% of individuals 65–74 years of age, and present in 17% of individuals 75–84 years of age (http://www.alz.org/documents_custom/2017-facts-and-figures.pdf). In this regard, 17% of the 27 participants >74 years of age correlates with the four participants who underwent marked decline in executive function. These analyses support that a degree of cognitive decline can occur prior to that warranting consulting a physician [1, 2]. Moreover, since AD is underdiagnosed and underreported, many individuals and/or their caregivers may remain unaware or unwilling to accept that they are experiencing the early stages of clinical cognitive decline [15–19]. Even if an individual is aware of memory difficulties, it is unclear whether or not assistance will be sought [20] especially in the absence of any family history of AD [21]. Herein, we provide data in support of these concerns by demonstrating that individuals who self-reported lack of any cognitive difficulties displayed severe decline as detected by an executive function test. These findings extend these concerns by indicating that studies aimed at normal cognition can be compromised by onset of AD-related cognitive decline. In retrospect, the findings discussed herein suggest inclusion of an independent tool (such as the Mini-Mental State Exam at each interval during studies to monitor cognitive performance among elderly individuals. Such would provide independent evaluation of performance in the context of onset of AD-related decline, which could justify excluding certain participants at the outset as well as considering individuals as outliers should they display a critical decline in Mini-Mental State Exam scores during a trial.
