Page 105 - AN-3-2

Advanced Neurology X chromosome-mediated risk in Alzheimer’s disease

Table 5. Analysis based on ancestral family history, based on UPDB population, indicating that the Cognitive Disorders
the sex of the affected probands, and the sex of the affected Clinic population tended to evaluate incident patients.
parental side We have previously pointed out that the female-to-male
24
ratio of AD in population-based age-specific incidence
Proband Lineage
(n=1850) Paternal (father’s Maternal (mother’s studies was 1.0, and that it dropped to 0.6 when looking at
first cousins) first cousins) population-based age-specific incidence of aMCI (usually
Female (1211) 629 582 an early clinical stage of what will evolve to be AD). The
greater prevalence of AD in women is due to their longer
Male (639) 343 296
lifespan, or the shorter lifespan of men who die at younger
Notes: (1) The parental lineage of AD was determined by third-degree ages from competing causes of mortality. It is also likely
relatives (data taken from Tables 2 and 3). (2) The calculated OR is 0.93 that males documented in the UPDB, including those who
(629:582/343:296), with a 95% CI of 0.77 – 1.13 (P=0.48).
also might have early manifestations of AD, died of other
causes, and only these non-AD diagnoses were reflected in
4. Discussion
their death certificates.
4.1. Preamble The 20-year gap between the average age at presentation
Due to the paucity of loci predisposing to AD that had of the memory disorders clinic population and the age at
been identified on the X chromosome, we were prompted death of the deceased population documented in the UPDB
to develop a new method to estimate the risk for AD suggests that the clinic was seeing patients with relatively
conveyed via the X chromosome, because of evidence early-presenting LOAD, given that the usual average duration
17
that the risk for the disease may depend on the sex of the of the disease after diagnosis is 3 – 8 years. In the Baltimore
affected parent, and possibly on the sex of the offspring. Longitudinal Study of Aging, the median survival times
The first application of this method indicated that the risk ranged from 8.3 years for persons diagnosed as having AD at
borne by the X chromosome might be considerable. We 65 years old to 3.4 years for persons diagnosed as having AD
then decided to see if a similar result would be obtained at 90 years old. Thus, the two studies reported on different
25
in an existing dataset of patients with AD and a thorough populations: one on a population with relatively early onset
history of the presence or absence of AD in their families, LOAD, while another on a population with a much later onset
i.e., the UPDS. Our initial results did not replicate in the of the disease. This idea is reinforced by the observation that
new dataset. In this section, we discuss several reasons why the median age of the AD population in UPDB, which is
limitations in the second dataset may have biased the result 85 years (55% of whom died after 1978), is also higher than
26
toward “null,” and how the lessons learned may influence the average for the Utah population: In Utah, life expectancy
future research to advance understanding of the role of the at birth for males increased from 72.4 years in 1980 to
26
X chromosome in conferring risk for AD. 77.1 years in 2020, and for females from 78.6 to 80.9 years.
This raises the possibility that some of the affected probands
4.2. Methodological critique in the UPDB suffered from a recently-recognized limbic-
27
The magnitude of the difference between the results obtained predominant age-related TDP-43 encephalopathy (LATE),
in this and the original study suggests that a comparison rather than AD. Limitations of using death certificate data to
17
of the populations between the two studies is warranted, define cases of AD have been discussed previously. 6
specifically looking for factors that may bias the OR in the The third methodological difference is that while the
present study toward unity, that were not present in the affected probands identified from the clinic population
original study. This comparison is presented in Table 6. did not include individuals with a bi-parental history of
The chief difference between the two studies is that dementia, the available UPDB data did not permit the
one had a population largely of carefully selected, newly exclusion of such individuals. Thus, this would bias the OR
diagnosed, incident patients and the other included in the UPDB data toward unity.
a population of deceased individuals, in whom the Another conspicuous difference has less to do with
cumulative lifetime risk to develop AD has been able methods than with yield. Less than 1% of individuals
to assert itself. Factors that impact the initial or early identified with AD in the UPDB were identified as having
presentation (incidence) of LOAD, which are appropriate a parent with AD, compared to over 50% of the cognitive
for consideration as risk factors, may not be detectable in disorders clinic population with a unilateral family history
a population of deceased individuals. The female-to-male of AD or dementia. It is challenging to explain how overall
ratio of the probands in the previous study (Cognitive low case finding of AD parents of probands with AD would
Disorders Clinic) was closer to unity than that of the affect the chances of finding male patients whose fathers

Volume 3 Issue 2 (2024) 6 doi: 10.36922/an.3122