Ethics for testing healthy people for research

[PharmacistCatherine]

What are your opinions on testing for the APO E4 alleles in a healthy population?

When Stavia and I caught up, we discussed the concerns about recommending LCHF diets that do not restrict saturated fats without first testing for the ApoE4 alleles. For example, in the research setting, looking at whether LCHF is superior to HCLF for improving glycaemic control in people with type 2 diabetes.

I'm trying to get my head around the ethics of this when performing research in a relatively large population. Yes, we can do the gene testing fairly easily, but what about reporting the results to the participants? If we were to collect data for a prospective study, looking at risk markers for long-term health outcomes in a young, currently healthy population, should we collect gene data?

Then, what do we do about reporting the results? Should participants be informed about their gene results? I read one paper on this today and it suggested that genetic counselling be given to anyone who wants to have gene testing, both when there are significant risk makers (e.g. strong family history of AD) or out of curiosity (or research). Having a positive result can come as a shock to the system. But is genetic counselling practical when you are trying to collect a cohort of 1000 people?

I'd appreciate your thoughts and comments.

[Julie G]

This is a hot topic in the genetic counseling arena today, Catherine. Many have been traumatized by “accidentally” learning of their APOE status. You are very astute to be alert to this. We’re currently working with Dr. Doris Zallen at Virginia Tech in a a study to examine the effect of this knowledge. Dr. Zallen is developing guidelines to be used prior to revealing APOE status. I’m sure she’d be happy to discuss this with you: dtzallen@vt.edu.

[Harrison]

Hi Catherine,

I'm glad you raised this point. If you are conducting a prospective study, I think obtaining genotype is very useful information to identify the exact type of subgroup interactions you describe. Things invariably country by country on this, but some IRBs will allow you to collect the genotype data in a blinded manner for a specific purpose without disclosing to the participant if genetic counseling is not practical. If the participant is going to be allowed to request the data, I think genetic counseling is a must for the reasons Julie outlined.

The short version of this is, if you have any ability to look at apoE genotype effects you should, whether or not this is disclosed to the participant.

[RichardS]

The short version of this is, if you have any ability to look at apoE genotype effects you should, whether or not this is disclosed to the participant.

From first hand knowledge, I would say presently IRB's are generally reluctant to allow genetic testing without explicit permission from subjects. By that I don't mean permission gene-by-gene, but researchers need to declare they are doing broad screening or identify specific genes they will be testing. There used to be more latitude, but things are getting more strict. There is definitely variability from one IRB to another. YMMV That said, what IRB's allow and what researchers end up doing does not always overlap 100%.

As to actually reporting those results to participants, no question that it has to be explicitly permitted by the IRB.

[Harrison]

Thanks for the clarifications Richard. There is no question that informed consent for genetic testing must be obtained, and IRB permission to give results to the participants must be obtained. What I should have said was this:

If you have any ability to look at apoE genotype effects you should, whether or not the participant is told their individual genotype (Participants will always be informed if genetic information is going to be collected).

This side steps the ethical issue Catherine brought up, but I would much rather have study done with apoE as a factor than not.

[circular]

I agree that including apoe information would be really important in such a study if possible. I hope it would be possible to disclose to patients (and have approved by the IRB) that their apoe gene will be genotyped but that no patient will receive the result; ie, they would have to have it tested elsewhere if they're interested. That removes the burden of arranging and ensuring genetic counseling from the researchers. I'm guessing, anyway, that the apoe gene testing within the study could be handled like some PET and MRI scans are, where patients agree to have it done but also not to receive their individual results. I'm really not well informed in this area and agree Doris Zallen would be a great contact, but I think one of the big issues becomes use of the genetic data by other researchers once it's made available, even in aggregate, and whether and how patients are informed up front that their anonymized result could be used in subsequent research (eg, a review study or another study by the same researchers). But those nuances will come up when you talk to someone like Doris about it. I love that you're considering including apoe information.

[MarcR]

concerns about recommending LCHF diets that do not restrict saturated fats without first testing for the ApoE4 alleles.

This is your issue, right Catherine? You're concerned about negatively affecting the health of ApoE4 carriers?

In my opinion, for this to be an ethical concern for any study, one would need scientifically convincing evidence that

1. LCHF diets with ad libitum saturated fat consumption affect ApoE4 carriers differently, and
2. those differential effects across the study duration significantly damage health in those ApoE4 carriers.

I think both A and B are open questions with credible conflicting evidence. Also, that you're not imposing a saturated fat requirement - you're simply electing not to restrict it - relieves any potential ethical burden even more.

For a recent discussion of LDL and all cause mortality, I recommend this topic. And this post within that topic is where I came to grips with an unwarranted assumption on my part regarding A.

[Stavia]

Catherine, heaps of facets here. And as genetic testing becomes more of a player in scientific research, this question will become more common.
Apoe4 and saturated fat? There are no RCTs at this stage. But it's a question that IMO needs to be answered especially as HFLC becomes more mainstream. We have seen LDLp in some of our homozygous members rise with increased sat fat and conversely drop. It's a surrogate marker.....

Two issues here - the science and the ethics.
Excellent questions.

[Tincup]

Catherine, for what it is worth, we consult with Steven Gundry who tests all his patients for ApoE4. He told me about 30% are E4 positive in his population. What he has observed in testing his patients is E4's spike sdLDL when they eat fat from land animals. Copies of our tests are in the post after this one. The linked post is a transcript of our (my E4/4 wife and my) first consult with him.

Interestingly, on the ethics side, Gundry tests all for ApoE status and tells it to them without any drama. His attitude is "it is not the hand you are dealt, it is how you play it." He has told us he has many older high functioning E4's that are in his patient population. I'm guessing he believes his approach strongly enough that he doesn't think it would be traumatic for his patients to learn of their status. Corresponding online with many of his patients, this seems to be true. My wife and I found him after we learned of our status.

[Julie G]

Ah, the conversation has taken an interesting turn! We APOE-ε4 carriers currently find ourselves in a no man’s land with the LDL particle number vs. size debate which is very relevant for those of us applying a HFLC diet. Catherine (a la Stavia) seems to be erring on the side of caution (and rightly so!) given the scrutiny an IRB board would apply. We desperately need well-designed long term studies looking at multiple end points -both surrogate biomarkers & real CVD events- to determine any pathology that might result from a HFLC (high SFA) diet for our APOE genotype. Until we have that data, any researcher would have to err on the side of safety given the correlation between elevated LDL-P and CVD...right?