Tau and Amyloid - Sex Differences
Julie Wisch
March 5, 2019
After reading Rhesa Sperling and Melissa Murray’s recent papers, I wanted to also look at sex differences in our cohort.
I started with 165 cognitively normal ADRC participants. They are well matched, although women had higher overall levels of tau than men.
## Stratified by GENDER
## female male
## n 100 64
## Age (mean (sd)) 66.67 (7.45) 67.83 (8.25)
## EDUC (mean (sd)) 16.05 (2.48) 16.25 (2.27)
## CDR = 0 (%) 100 (100.0) 64 (100.0)
## apoe4 = 1 (%) 32 ( 32.0) 21 ( 32.8)
## PUP_fSUVR_rsf_TOT_CORTMEAN.x (mean (sd)) 1.32 (0.57) 1.32 (0.59)
## PUP_fSUVR_rsf_TOT_CORTMEAN.y (mean (sd)) 1.19 (0.14) 1.11 (0.12)
## Stratified by GENDER
## p test
## n
## Age (mean (sd)) 0.352
## EDUC (mean (sd)) 0.603
## CDR = 0 (%) NA
## apoe4 = 1 (%) 1.000
## PUP_fSUVR_rsf_TOT_CORTMEAN.x (mean (sd)) 0.999
## PUP_fSUVR_rsf_TOT_CORTMEAN.y (mean (sd)) <0.001Sperling’s paper found no clear association with sex for tau. However, she did find higher entorhinal cortical tau in women for a given level of tau.
Murray’s path paper said, “There is an interaction betweeen age and sex, resulting in selective neuroanatomic susceptibility to neurofibrillary pathology in different decades of life.”
When I included age as a covariate, I saw an association of sex with tau in the temporal and occipital lobes, as well as a trend in the parietal lobe. We also saw an age-sex interaction in the frontal lobe.
This is the model I started with for each lobe: Tau ~ Amyloid + Sex + Age + Amyloid:Sex + Amyloid:Age + Age:Sex + Amyloid:Age:Sex + ApoeStatus
After maximum likelihood model selection, here are the final models for tau accumulation in each lobe:
##
## Call:
## lm(formula = Frontal.tau ~ Frontal.amyloid + Age + Frontal.amyloid:Age +
## Age:GENDER + Age:GENDER:Frontal.amyloid, data = df)
##
## Coefficients:
## (Intercept) Frontal.amyloid
## 0.8071632 1.6195614
## Age Frontal.amyloid:Age
## -0.0117424 -0.0175198
## Age:GENDERmale Frontal.amyloid:Age:GENDERmale
## 0.0008918 -0.0014485##
## Call:
## lm(formula = Parietal.tau ~ Parietal.amyloid, data = df)
##
## Coefficients:
## (Intercept) Parietal.amyloid
## 2.167e-16 3.466e-01##
## Call:
## lm(formula = Temporal.tau ~ Temporal.amyloid + GENDER, data = df)
##
## Coefficients:
## (Intercept) Temporal.amyloid GENDERmale
## 0.1465 0.4211 -0.3753##
## Call:
## lm(formula = Occipital.tau ~ GENDER, data = df)
##
## Coefficients:
## (Intercept) GENDERmale
## 0.1915 -0.4906It’s easy to see in these figures that women have more tau for a given level of amyloid in the temporal and occipital regions. The trend for a sex difference in the parietal region is also easy to see in the figures.
## TableGrob (2 x 2) "arrange": 4 grobs
## z cells name grob
## 1 1 (1-1,1-1) arrange gtable[layout]
## 2 2 (1-1,2-2) arrange gtable[layout]
## 3 3 (2-2,1-1) arrange gtable[layout]
## 4 4 (2-2,2-2) arrange gtable[layout]
## TableGrob (2 x 2) "arrange": 4 grobs
## z cells name grob
## 1 1 (1-1,1-1) arrange gtable[layout]
## 2 2 (1-1,2-2) arrange gtable[layout]
## 3 3 (2-2,1-1) arrange gtable[layout]
## 4 4 (2-2,2-2) arrange gtable[layout]
## TableGrob (2 x 2) "arrange": 4 grobs
## z cells name grob
## 1 1 (1-1,1-1) arrange gtable[layout]
## 2 2 (1-1,2-2) arrange gtable[layout]
## 3 3 (2-2,1-1) arrange gtable[layout]
## 4 4 (2-2,2-2) arrange gtable[layout]
## TableGrob (2 x 2) "arrange": 4 grobs
## z cells name grob
## 1 1 (1-1,1-1) arrange gtable[layout]
## 2 2 (1-1,2-2) arrange gtable[layout]
## 3 3 (2-2,1-1) arrange gtable[layout]
## 4 4 (2-2,2-2) arrange gtable[layout]With these results, we both replicate the Sperling findings of higher tau for a given layer of amyloid as well as demonstrate an in vivo age x sex x amyloid effect on tau tangles in the frontal part of the brain.
Below is a visualization of this 3 way effect. Note that the axes for amyloid and tau are Z scores, not raw values of amyloid and tau, so the range of values described by the plane is +/- one standard deviation from the mean amount of amyloid or tau.
```