Plasmalogens- exciting new evidence

Insights and discussion from the cutting edge with reference to journal articles and other research papers.
antimatter37
Contributor
Contributor
Posts: 56
Joined: Fri Feb 02, 2018 11:52 am

Re: Plasmalogens- exciting new evidence

Post by antimatter37 »

I agree that this line of research presents a very interesting and positive potential intervention therapy for APOE4's. The article itself is a bit difficult to immediately relate to the real world as the authors seem to really like taking the "normalized base ten log" of many standard lipid profile numbers (and others) prior to presenting them in the text or in charts. For example, their very important "PBV" value is an amalgam of 4 or 5 serum plasmalogen numbers, normalized to an APOE3 baseline then transformed by taking the base 10 log. For me, giving up on trying to relate these types of numbers to the real world and instead just looking at relative values proved most helpful. For example:

1. Each 1 standard deviation (SD) increase in serum PBV is associated with a 30% reduction in the odds ratio for developing MCI and AD in the APOE4 community.

2. Each 1 SD increase in HDL-R (HDL to total cholesterol ratio) is associated with a 10% decrease in odds ratio for MCI/AD in the APOE4 community.

Using the APOE3/3 community as a baseline, the odds ratio for the APOE 3/4 4/4 (the authors combine these) community for MCI/AD is 1.996, or about a factor of two (the authors numbers). I interpret this as meaning that an increase in PBV level for the APOE4 community of somewhere between 1 and 2 standard deviations would put our odds ratio for MCI/AD on a par with that of the APOE3/3 community. This would be very nice indeed!

Even better is the fact that changes of serum PBV values of this magnitude are already seen in the various APOE genotype populations. This means that a 1 or 2 (or more) standard deviation increase in PBV might be relatively achievable as a therapy.

With regard to HDL-R values, the higher the better. However, obtaining the same "bang for the buck" in attempting a many multiple standard deviation increase in HDL-R to realize the same decrease in odds ratio for MCI/AD as seen with the 1 or 2 SD increase in PBV would seem to be pretty much impossible.

In searching the web for means to increase serum plasmalogens (and thus PBV) it seems that some natural foods or supplements such as shark liver and krill contain very small amounts. There are also a handful of chemical precursors that come up, especially alkyl-glycerol (I don't know what this is or if it can be ingested safely). Is anyone in this community taking supplements or precursors to increase their plasmalogen level?
antimatter37
Contributor
Contributor
Posts: 56
Joined: Fri Feb 02, 2018 11:52 am

Re: Plasmalogens- exciting new evidence

Post by antimatter37 »

Here is more information on trying to reverse engineer some of the numbers presented in Fig. 1 of this newest paper (pet peeve - we should not need to do this to see some real-world numbers). Also, I might not have this right, so please feel free to analyze what I have done. Basically it involves eyeballing values from the figure, inverse log transforming them, and then inverse normalizing them.

For figure 1 (D), the percentage chance for dementia for the three groups of APOE carriers as a function of HDL-R value, the reverse engineered x-axis of HDL-R values has new values of approximately 0.623 on the far left, 0.335 where the zero value currently is, and 0.144 on the far right. These new values should represent the real world HDL-R values for the study group from largest value (0.623) to lowest (0.144) and a sample mean or average value of 0.335.

For figure 1 (E), the percentage chance for dementia for the three groups of APOE carriers as a function of PBV value, the reverse engineered x-axis of PBV values has new values of approximately -0.34 on the far left, -0.086 where the zero value currently is, and -0.014 on the far right. These new values should represent the real world PBV values for the study group from largest negative value (-0.34) to lowest (-0.014) and a sample mean or average value of -0.086.

As an example of dementia probability comparison, from fig 1 (E), we can inverse transform the x-axis point where E3's cross the 0.15 y-axis probability (since the x-axis is a log scale, you need to look at each point separately). Doing this inverse transform, this 15% probability point occurs at an HDL-R value of 0.271. For an E4 to have that same probability of dementia, 15%, the HDL-R ratio would need to be about 0.623 as seen on the chart where it crosses the 0.15 line. If the standard deviation in HDL-R values for this study group is 0.09 as reported, that is a 3.9 standard deviation improvement in HDL-R needed to equalize E4 with E3 dementia probabilities.

Doing the same calculations for the PBV values shows only a approximate 1/2 standard deviation improvement needed to equalize the dementia probabilities between the E3 and E4 groups (I think I guessed a value of 2 previously, it appears to be much better!). In the paper abstract and body, the authors use 1 SD for the equalization factor, so there may be some errors in my approximations, but the overall results are similar.

I looked at my own values here too. With an HDL of 47 and TC of 182, that gives me an HDL-R of 0.26, slightly worse that the sample norm here. Reverse normalizing this value to the study sample mean of .335, I get a normailzed HDL-R of 0.78. Taking the base 10 log of this, I get a value of -0.11. Looking at figure 1 (D) in the paper, I can see that I lie just to the left of the -0.12 mark on the chart, very close to the mean (the magic of logs). If I could somehow raise my HDL to 113 while keeping my total cholesterol at 182, I would be about the same as an E3 for probability of dementia.
aphorist
Senior Contributor
Senior Contributor
Posts: 233
Joined: Sat Apr 19, 2014 3:47 am

Re: Plasmalogens- exciting new evidence

Post by aphorist »

antimatter37 wrote: For figure 1 (D), the percentage chance for dementia for the three groups of APOE carriers as a function of HDL-R value, the reverse engineered x-axis of HDL-R values has new values of approximately 0.623 on the far left, 0.335 where the zero value currently is, and 0.144 on the far right. These new values should represent the real world HDL-R values for the study group from largest value (0.623) to lowest (0.144) and a sample mean or average value of 0.335.
I don't follow how you are getting to an HDL-R of 0.335 or why you are trying to extract it off of figure 1D. Why not look at Table 1. where the mean value is 0.338 for Female and 0.331 for Male and use that as a normalizing constant?
antimatter37
Contributor
Contributor
Posts: 56
Joined: Fri Feb 02, 2018 11:52 am

Re: Plasmalogens- exciting new evidence

Post by antimatter37 »

Hi Aphorist! In the charts, the authors appear to not have used either the male of female HDL-R values, but instead a weighted average of the two. To calculate the weighted average, begin with the percentage of females in the study (921 of 1205 total persons, or 76.4%). Do the same for men (284 out of 1205 or 23.6%). The weighted average is then the female HDL-R times .764 plus the male HDL-R times .236, or (.338*.764)+(.331*.236) = .335.

An interesting calculation is if someone had the exact mean value of HDL=R, or 0.335. You can do the transforms on this just as the authors did. First normalize to the mean value, which of course just gives the number 1. Then, take the base 10 log of that. The log of 1 is zero. If you look at the two left columns of charts in figure 1, the lower, or x, axis all have a point marked of zero. That point on these 6 charts is where the sample mean HDL-R (leftmost 3 charts) or sample mean PBV (middle three vertical charts) lie.

Because the authors chose to transform all of their participants data in this way, readers cannot simply calculate their HDL-R number and go to the charts to see where they might be. They need to do this needlessly complicated transform process!
aphorist
Senior Contributor
Senior Contributor
Posts: 233
Joined: Sat Apr 19, 2014 3:47 am

Re: Plasmalogens- exciting new evidence

Post by aphorist »

antimatter37 wrote: Because the authors chose to transform all of their participants data in this way, readers cannot simply calculate their HDL-R number and go to the charts to see where they might be. They need to do this needlessly complicated transform process!
I think they did this to create normally distributed variables for an OLS regression, otherwise there would be non-constant variance and an invalid regression. I see what you mean by weighting female / male participants to get a weighted normalization constant, but the real world people are not 70% female and 30% male. For individual analysis it's probably better to just take the value from Table 1 and then understand the regression is just a regression -- a best fit of the parameters and there is some margin of error at any given point estimate.
User avatar
Julie G
Mod
Mod
Posts: 9187
Joined: Sat Oct 26, 2013 6:36 pm

Re: Plasmalogens- exciting new evidence

Post by Julie G »

Once again, apologies for posting this twice. I didn't want it to be lost in our biomarker study opportunity thread. In a new YouTube video, Dr. Goodenow explains his most recent paper. I found it very enlightening! To refresh your memory, see the paper below:

Relation of Serum Plasmalogens and APOE Genotype to Cognition and Dementia in Older Persons in
a Cross-Sectional Study/b]
Abstract: Using a community sample of 1205 elderly persons, we investigated the associations and potential interactions between Apolipoprotein E (APOE) genotype and serum phosphatidylethanolamine (PlsEtn) on cognition and dementia. For each person, APOE genotype, PlsEtn Biosynthesis value (PBV, the combination of three key PlsEtn species), cognition (the combination of five specific cognitive domains), and diagnosis of dementia was determined. APOE genotype and PBV were observed to be non-interacting (p > 0.05) and independently associated with cognition: APOE (relative to ε3ε3:ε2ε3 (Coef = 0.14, p = 4.2 × 10−2); ε3ε4/ε4ε4 (Coef = −0.22, p = 6.2 × 10−5); PBV (Coef = 0.12, p = 1.7 × 10−7) and dementia: APOE (relative to ε3ε3:ε2ε3 (Odds Ratio OR = 0.44, p = 3.0 × 10−2); ε3ε4/ε4ε4 (OR = 2.1, p = 2.2 × 10−4)); PBV (OR = 0.61, p = 3.3 × 10−6). Associations are expressed per standard deviation (SD) and adjusted for serum lipids and demographics. Due to the independent and non-interacting nature of the APOE and PBV associations, the prevalence of dementia in APOE ε3ε4/ε4ε4 persons with high PBV values (>1 SD from mean) was observed to be the same as APOE ε3ε3 persons (14.3% versus 14.0%). Similarly, the prevalence of dementia in APOE ε3ε3 persons with high PBV values was only 5.7% versus 6.7% for APOE ε2ε3 persons. The results of these analyses indicate that the net effect of APOE genotype on cognition and the prevalence of dementia is dependent upon the plasmalogen status of the person.
goodenowe_2019_relation of serum plasmalogens and apoe to cognition and dementia.pdf
You do not have the required permissions to view the files attached to this post.
mike
Senior Contributor
Senior Contributor
Posts: 849
Joined: Fri Mar 09, 2018 4:55 pm
Location: CA - Sonoma County

Re: Plasmalogens- exciting new evidence

Post by mike »

Thanks for being this back. Great video. Now if only we could get the straight dope on phospholipids in DHA...
Sonoma Mike
4/4
User avatar
Julie G
Mod
Mod
Posts: 9187
Joined: Sat Oct 26, 2013 6:36 pm

Re: Plasmalogens- exciting new evidence

Post by Julie G »

Thanks for being this back. Great video. Now if only we could get the straight dope on phospholipids in DHA...
Yeah, there appears to be some overlap between Rhonda's hypotheses and Dr. Goodenowe's. I shared her paper with him. He seemed enthused and mentioned that our proposed biomarker study would be measuring the phosphatidylcholine-DHA that Rhonda is targeting.

I hope others get a chance to listen to this video. I found one statement to be intriguing. Dr. Goodenowe asserts that HIGH levels of cholesterol in the brain are associated with an increased risk of AD. (Of course many papers assert that the opposite is true; that lower levels of cholesterol in the brain are associated with AD.) If Dr. Goodenowe's assertion is true, could a lipophilic statin that could potentially cross the BBB be helpful to counteract this? This is basically the opposite of what Dr. Dayspring recommends to protect the cholesterol pool in the brain. Head spinning.

FWIW, based on Dr. Goodenowe's plasmalogen/cholesterol correlations, I've decided my plasmalogen levels are high: LDL 79, HDL 99, TG 50. I'll consider this my own personal "get out of jail free card"... for today. :roll:
chrissyr
Senior Contributor
Senior Contributor
Posts: 184
Joined: Sat Sep 02, 2017 4:38 am

Re: Plasmalogens- exciting new evidence

Post by chrissyr »

Thank you Julie! Great video and I appreciate that it was explained simply. I wonder if Dr. Goodenowe would share his thoughts on the lipophilic stain question? Congrats on your get out of jail free card-- those are some gorgeous numbers!
My HDL has always been below normal with LDL high, just like my mom's who suffers from AD, so to me it might be worth the gamble!
Daily exercise and higher fat diet does inch it slightly higher, but still pretty abysmal.
I'll continue to keep by eye on this as how to raise both HDL and plasmalogens!
User avatar
Rainbow
Senior Contributor
Senior Contributor
Posts: 132
Joined: Fri Apr 27, 2018 9:34 pm

Re: Plasmalogens- exciting new evidence

Post by Rainbow »

Awesome video! Will be sharing with my parents and sister. Thank you!
4/4
Post Reply