Brain: Chinese scientists have revealed for the first time the changes in adult cerebrospinal fluid and imaging AD biomarkers over time

*For medical professionals only

The pathological process of Alzheimer's disease (AD) is known to be so long that it has been "silently" planted

At present, the number of AD patients in China has exceeded 13 million, and with the deepening of China's aging degree, the number of patients is increasing rapidly[1].

Recently, the research results of Professor Xiong Chengjie's team at the University of Washington School of Medicine were published in the prestigious journal Brain

Specifically, the ratio of CSF Aβ42 to Aβ42/Aβ40 in the 45-50 age range decreased longitudinally compared with the baseline age range of 18-45 years, and the total tau protein level (Tau) and phosphorylated Tau181 (pTau181) increased

The above results show the earliest age range for possible neuropathological changes in AD, which gives accurate enlightenment for the relevant experimental design, which is of great clinical significance

Research paper homepage

Clinically, by measuring the Aβ42, Aβ42/Aβ40 ratio, Tau, pTau181 and PET scans Aβ plaque imaging, magnetic resonance (MRI) scanning of brain structure, etc.

This year, it was found that Aβ was deposited in neuronal cells early in AD, and the subsequent formation of age plaques is actually the remains of neuronal apoptosis [3].

Therefore, most of the participants will not (too late) have obvious symptoms during the trial, resulting in a large number of resources being wasted, ineffective trials occur from time to time, many AD new drugs are broken in the sand, and many pharmaceutical companies have withdrawn

In the face of such confusing AD, the research team plans to start with age (maximum risk factor) and APOE ɛ4 (maximum genetic risk factor), aiming to find the regular/relative order of biomarkers over time and reveal the potential effects of APOE ɛ4, so that it is possible to accurately "search" people who only have symptoms in the next few years, and have the opportunity to establish reasonable inclusion/exclusion criteria, greatly improve the efficiency of the trial, and make it possible for better treatment/prevention programs to break the ice

The research team has previously developed a novel algorithm based on age and PET scan data that can assess when people who are now cognitively normal will develop AD symptoms and can estimate the time before cognitive impairment develops [4].


Combining the above, the research team reused eight ongoing longitudinal studies to form a large biomarker and cognitive database that included a total of 2703 participants with baseline ages ranging from 18 to 103 years, with an average age of about 66 years, a female proportion of about 60%, and an APOE ɛ4 carrier ratio of more than 30%.


According to the age, it is divided into 8 reasonable ranges of 18-45, 45-50, 50-55, 55-60, 60-65, 65-70, 70-75, and 75-104 years, and the median follow-up time is 5-6 years
.

After screening out participants who did not meet the 3 consecutive observations, the team divided the participants into cognitive cohorts (2137), MRI cohorts (544, hippocampal volume and cortical thickness), PET PiB cohorts (285, mean cortical and pre-wedge SUVR), and cerebrospinal fluid cohorts (249, Aβ42, Aβ42/Aβ40 ratios, Tau, and pTau181) by marker type, and used a linear mixed-effects (LME) model.
The marker rate of change in different age ranges was compared separately to infer the regular/relative order of biomarker change over time and to reveal how APOE ɛ4 changed the longitudinal trajectory
.

For "acceleration", the research team also gave a special explanation: it does not represent an acceleration of the rate of change within the participants, but rather an increase/decrease
in the average rate of change of biomarkers from one younger baseline age range to the next older age range.

This is very helpful
for the understanding of research logic.

The research team first explored the longitudinal trajectory of the baseline age range of 18-45 years, and the annual change rate of learning ability was statistically different in the cognitive queue composite analysis (P=0.
0093); The annual rate of change of the ratio of Aβ42 and Aβ42/Aβ40 was 9.
03 pg/ml (P=0.
0028) and 0.
0002 (P=0.
0077), respectively; MRI cortical thickness decreases over time with significant differences (P<0.
0001);P and a tendency to decrease is also observed for iB SUVR, Tau, and pTau181 (P>0.
05
).

The annual rate of change of major AD biomarkers and their 95% CIs relative to the baseline age range

For participants over 45 years of baseline age, the only age range in which a significant acceleration in Aβ42 reduction was observed in the cerebrospinal fluid queue was 45-50 years (P=0.
0055); The annual rate of change of the Aβ42/Aβ40 ratio showed an accelerated decline in the 45-50 years and 65-70 years (P<0.
0001); the change rate of Tau and pTau181 showed a positive trend across all age ranges and showed a statistical difference between 45-50 years and 65-70 years (P<0.
05).
<b10>

In the PET PiB cohort, both the mean cortex and the anterior wedge SUVR were in the baseline age 50-55 years range, and for the first time significantly accelerated increases were observed (P<0.
05), until the change began to slow after<b10> the 65-70 age range.

In the MRI queue, the volume of hippocampuses in the baseline age range of 55-60 years was more than twice that of the previous age range (P=0.
008), and the decrease continued to accelerate in the three ranges of 60-75 years (P<0.
0001<b10>).

Only the first/only significant accelerated reduction in cortical thickness in the 65-70 years of age ranges (P<0.
0001).
<b12>

Finally, cognitive performance declined significantly for the first time in the baseline age range of 65-70 years, with an annual decline rate more than twice as fast as the 60-65 age range (P<0.
0001), followed by a sustained decline<b10>.

Longitudinal trajectories of major biomarkers in different baseline age ranges (solid line)

The data points and lines are the mean expressions of AD biomarkers fitted based on the LME model

The black dashed curve is a cubic spline smoothing curve over the entire age span

In addition, APOE ɛ4 may alter the longitudinal trajectory of Aβ metabolism and neuronal injury, which is also important
for AD prevention trial design.

Its potential mechanism of effects is well worth exploring, so the research team established separate LME models for APOE ɛ4 carriers and non-carriers, and independently repeated the similar analysis
described above with the APOE ɛ4 state as a covariate.

The results showed that in the 18-45 age range, only APOE ɛ4 non-carriers could observe a significant increase in the ratio of Aβ42 to Aβ42/Aβ40 and a significant decrease in PiB SUVR.
Once the baseline age was over 45 years, the difference between the two was more prominent, and the rate of change in all cerebrospinal fluid markers, PiB SUVR, and cognitive ability of APOE ɛ4 carriers was significantly greater than that of non-carriers, and the first accelerated longitudinal changes occurred in the 45-50/50-55 age range, and in these two same age ranges, only PiB SUVR cortical thickness and cognitive ability showed the earliest longitudinal acceleration changes
in non-carriers.

It was thought that the longitudinal regulatory effect of APOE ɛ4 on biomarkers may occur after old age (65 years and older), but these results suggest that APOE ɛ4 began to regulate biomarkers earlier than expected (in the 18-45 age range
).

Longitudinal trajectories (solid lines) of APOE ɛ4 carriers (red) and non-carriers (green) for different baseline age ranges

The data points and lines are the average expression of AD markers fitted based on the LME model

Dotted line: Cubic spline smoothing curve over the entire age span (black for APOE ɛ4 carriers, dark purple for APOE ɛ4 non-carriers)

Finally, after Benjamini-Hochberg sensitivity correction analysis, it was concluded that the ratio of Aβ42 (FDR P=0.
0386) and Aβ42/Aβ40 (FDR P=0.
0015) decreased in the baseline age range of 45-50 years, the Tau accelerated increase (FDR P=0.
0551), and the PiB mean cortical SUVR increased (FDR P=0.
0551).
Accelerated decrease in hippocampal volume between 55-60 years (FDR P=0.
03); Cortical thickness (FDR P<0.
0001) and cognitive ability (FDR P<0.
0001) both accelerated declines in the 65-70 age range; in the 65-70 age range, additional significant acceleration changes occurred in MRI markers (FDR P<0.
0001) and cognitive ability (FDR P<0.
0013) in the 65-70 age range<b10>.

Of course, there are some shortcomings in this study, such as the fact that the participants did not include people younger than 18 years old, which may miss the potential mechanism of biomarker changes at this age; In addition, the sample size and longitudinal follow-up time are still limited, the proportion of Caucasian races is relatively high, and the inclusion of races is relatively single
.

In general, this study solves some of the "old and difficult problems" of previous cross-sectional studies, such as short follow-up time and relatively small sample size
.

Crucially, this study has the potential to suggest a precise optimal time window
for the experimental protocol.

For example, primary prevention trials for Aβ and Tau may need to start in the 45-55 age range; Secondary cognitive prophylaxis trials are to be conducted in individuals who already have positive biomarker results, starting in the 65-70 age range, is more appropriate
.

I hope that in the near future, the "eraser in the brain" will never erase any traces of beauty and life! "People trapped in time" can see a new dawn
because of the law of time change of the markers derived from the above research.

References:

1.
Alzheimer's Disease Report China 2021 (AD Report).

2.
Luo J, Agboola F, Grant E, et al.
Accelerated longitudinal changes and ordering of Alzheimer disease biomarkers across the adult lifespan [published online ahead of print, 2022 Aug 4].
Brain.
2022; awac238.
doi:10.
1093/brain/awac238.

3.
Ju YH, Bhalla M, Hyeon SJ, et al.
Astrocytic urea cycle detoxifies Aβ-derived ammonia while impairing memory in Alzheimer's disease.
Cell Metab.
2022; 34(8):1104-1120.
e8.
doi:10.
1016/j.
cmet.
2022.
05.
011.

4.
Schindler SE, Li Y, Buckles VD, et al.
Predicting Symptom Onset in Sporadic Alzheimer Disease With Amyloid PET.
Neurology.
2021; 97(18):e1823-e1834.
doi:10.
1212/WNL.
0000000000012775.

The author of this article is Mr.
Origin

Editor-in-charge Daisi Rain