chillier
Senior Member (Voting Rights)
Really interesting and thought provoking read, thank you!
The buspirone challenge test clearly distinguishes ME/CFS patients from healthy controls: why is it not being developed and deployed?
- Thread starter Hip
- Start date
Really interesting and thought provoking read, thank you!
Jonathan Edwards
Senior Member (Voting Rights)
A practical question. @forestglip and others.
Does it look as if one might be able to do the buspirone challenge with a smaller dose? From what I have seen of the results even from 20 years back the responses looked easy to detect. A lower dose might even give a more qualitative difference?
I think further work should make sure it is using the minimum intevention to get good data.
Does it look as if one might be able to do the buspirone challenge with a smaller dose? From what I have seen of the results even from 20 years back the responses looked easy to detect. A lower dose might even give a more qualitative difference?
I think further work should make sure it is using the minimum intevention to get good data.
What we have in terms of existing dosage evidence:
Bakheit 1992 and the Majeed/Behan 1996 studies (grouping the latter two since they shared a control group) used 60 mg. Richardson 1995 used 50 mg. Sharpe 1996 used 0.5 mg/kg up to 45 mg (avg. weight 80 kg, so avg. dosage probably less than 40 mg). They all found significant effects.
There was also the Sharma 2001 case report that only used 30 mg of buspirone, in which the patient had a very large prolactin response while symptomatic.
Two studies of migraine without aura, from the same group, found substantially increased prolactin response with only 30 mg of buspirone (Cassidy 2003a, Cassidy 2003b). (No one followed up or showed negative results for buspirone challenge in migraine, yet that line of research stopped as quickly as it started in 2003.)
30 mg of buspirone also led to significant differences in studies of IBS (O'Mahony 2008) and non-ulcer dyspepsia (Dinan 2001). (Some other studies of dyspepsia used 60 mg and also found significant differences, eg. Dinan 1990 and Dinan 1993.) Note that Ted Dinan was an author on all the mentioned migraine, IBS, and dyspepsia papers, as well as on the Bakheit 1992 ME/CFS paper.
Thus, I can't be sure, but I suspect that 30 mg would still be sufficient to show group differences in ME/CFS. Maybe less.
Edit: I have a slight worry about a new study trying too low of a dose, or testing a new drug, and not getting a significant response, and this dampening enthusiasm for prolactin response research if it gets taken as evidence that there's nothing different at all. So I'm not sure, but maybe it would be good to at least include an arm which is 50 or 60 mg buspirone, along with whatever new thing is tried.
Kitty
Senior Member (Voting Rights)
Maybe it would be worth building in a dose response element if enough participants can be found / enough money is available for the lab work? It doesn't strike me as a particularly expensive project.
Here's one interesting finding I just noticed from a dyspepsia paper, showing increased nausea response in the patient group, to go along with the reports of increased side effects to buspirone in ME/CFS. This one also found a correlation between side effects and prolactin response, unlike Sharpe et al's study of ME/CFS.
A double-blind placebo-controlled study of buspirone-stimulated prolactin release in non-ulcer dyspepsia–are central serotoninergic responses enhanced? (Dinan et al., 2001)
A double-blind placebo-controlled study of buspirone-stimulated prolactin release in non-ulcer dyspepsia–are central serotoninergic responses enhanced? (Dinan et al., 2001)
(Line breaks added)
Kitty
Senior Member (Voting Rights)
I wonder if that's tangentially connected to domperidone (dopamine antagonist / anti-nausea drug) causing vomiting in some people?
Jonathan Edwards
Senior Member (Voting Rights)
Yes, I think the best plan would be to do a small stratified dose-ranging study. It might be possible to do that starting at 30mg, with a diagnosis-blinded assessment of the range of responses, followed by either or both higher or lower doses depending on the range of response. That might be problematic but having several time points for sampling and a profile makes the result a lot more robust than just a single post-challenge measurement. If I can get interest we should discuss further.
hotblack
Senior Member (Voting Rights)
Just looking at the bnf pages and it seems lower is commonly used.
The mention of CYP3A4 inhibitors is interesting too. Could there be something relevant there to increased effect in some people? I guess we’d have seen it in other medications if so, so perhaps not.
I guess there’s a balance of wanting to be sure of an effect with wanting to limit potential side effects. Some risk tolerance may be dependent upon study group severity.
I wonder what rationale the single doses for these studies were based upon? Common safe daily dosages in use at the time? Were there other reasons?
The mention of CYP3A4 inhibitors is interesting too. Could there be something relevant there to increased effect in some people? I guess we’d have seen it in other medications if so, so perhaps not.
I guess there’s a balance of wanting to be sure of an effect with wanting to limit potential side effects. Some risk tolerance may be dependent upon study group severity.
Good question. I found two sources that say the max dosage in the US is 60 mg: PDF from FDA in 2001 and paper from 2012.
Maybe they wanted to have the best chance of seeing an effect, so used the maximum dose regarded as safe.
Jonathan Edwards
Senior Member (Voting Rights)
Dose choices are often poorly justified in the context of therapy. For a test, I agree that in the old days a top whack dose might well be chosen in the hope of getting clearest results. It seems that the test never became a routine service procedure. If it had, more attention to dose might have occurred but even then there is no knowing.
hotblack
Senior Member (Voting Rights)
As someone who has actually had to make these decisions do you have any thoughts on what the approach should be here? Is there any informed approach we can base this decision on? Is it a bit of an educated guess? Or is it really just a case of doing a well designed experiment to find out?
Jonathan Edwards
Senior Member (Voting Rights)
The standard protocol is to establish minimum effective dose for whatever your purpose is. In the clinic you tend to want maximum efficacy which is the top of the sigma of the dose response curve but for a test you want minimum reliable discrimination. That is likely to be nearer tthe bottom of the sigma. The ethical thing to do if the test is explored more is to find that minimum reliable discrimination dose by titrating down, either from the previously used dose or a step lower. There is a complicated trade off in terms of risks of wasting effort and exposing people to tests without getting useful data and I think how you best handle that is going to be an educated guess.
hotblack
Senior Member (Voting Rights)
Thanks @Jonathan Edwards that’s really useful. Even talking about it here with something comparatively innocuous feels like quite a responsibility. I don’t envy anyone having to weigh some of these factors up.
I’m really torn on this. On the one hand I want clear, indisputable evidence. I want to give this the best chance of helping us understand what’s happening. And I’ve argued that studying, including and understanding what is happening with severely affected people is key and possibly the best place to start for some studies. But the idea of giving severe people a higher dose seems unethical. I wouldn’t want to try it. So do we only include more mildly affected people? How do we understand both dosage and severity impacts here? It’s a really tough one.
I assume most of the patients in the older studies were closer to the mild/moderate end of the spectrum if they could make it to the clinic, and the differences were already pretty clear. So I think the bulk of the research can probably continue to be on more mildly affected patients.
Maybe eventually there will be a reason to test in severe patients, but a lot can probably still be learned without doing that.
Utsikt
Senior Member (Voting Rights)
Purely from a safety perspective, would it not make sense to start with the mildly affected and at a low dose?
Then gradually ramp up the dose to see what happens, possibly with the same participants.
Then try to replicate the findings in the more severely affected.
In terms of subgroups, it would be interesting to see if there’s a difference between genders, age of onset, and infectious, non-infectious sudden and non-infectious gradual onset. At least if nothing else pops up along the way. Checking if the menstrual cycle impacts the results is also be worthwhile.
I imagine the primary bottleneck would be having people that can do the testing. Recruitment shouldn’t be an issue if the S4ME hive promotes it in their networks, and it looks like we’d be able to find funding somewhere for preliminary studies.
Then gradually ramp up the dose to see what happens, possibly with the same participants.
Then try to replicate the findings in the more severely affected.
In terms of subgroups, it would be interesting to see if there’s a difference between genders, age of onset, and infectious, non-infectious sudden and non-infectious gradual onset. At least if nothing else pops up along the way. Checking if the menstrual cycle impacts the results is also be worthwhile.
I imagine the primary bottleneck would be having people that can do the testing. Recruitment shouldn’t be an issue if the S4ME hive promotes it in their networks, and it looks like we’d be able to find funding somewhere for preliminary studies.
V.R.T.
Senior Member (Voting Rights)
Have we had any indication of long term harm on patients in the original studies? If they drop the dose and then it doesn't replicate we're left with a lot of uncertainty.
Kitty
Senior Member (Voting Rights)
Might that involve more clinic visits and lab work, though? If you start on a moderate dose (perhaps the lowest used in past studies) and titrate down, you might be able to cut some of that out. But I agree initial trials should probably be on mild and moderately affected people.
I don't know if it's allowable, but it might even make sense to test moderate dose against placebo, then plan the next phase when the results are known. If there's a response in most of those who had the drug, you could potentially split them to test a range of lower doses. After all it isn't a treatment trial where blinding has to be maintained throughout the follow-up period, it's a replication study asking some additional questions.
What stands out to me is that pretty much the only specific mechanistic pathway known to increase prolactin response to various drugs that I've seen is increased estradiol. (Though there was also some less fleshed-out research finding increased prolactin response after giving animals neurotoxic drugs, but I don't think they identified the specific way it did that - could be estrogen-related too, for all I know.) And estradiol appears to at least partly do this through changing sensitivity or density of dopamine D2 receptors on lactotrophs. (e.g. Labrie 1980, Bression 1985, Pasqualini 1986).
I think there's a decent chance that the increases in prolactin response are downstream of changes in dopamine D2 receptor density on the pituitary cells. And I think one of the most straightforward things to do is check whether the ME/CFS findings relate to this fairly well-characterized mechanism.
If this is right, PET scans of these receptors might be even more sensitive and explanatory than buspirone challenges, and could be done without requiring taking a drug. [Edit: Well, it does require injecting the PET tracer.]
I think there's a decent chance that the increases in prolactin response are downstream of changes in dopamine D2 receptor density on the pituitary cells. And I think one of the most straightforward things to do is check whether the ME/CFS findings relate to this fairly well-characterized mechanism.
If this is right, PET scans of these receptors might be even more sensitive and explanatory than buspirone challenges, and could be done without requiring taking a drug. [Edit: Well, it does require injecting the PET tracer.]
Last edited:
hotblack
Senior Member (Voting Rights)
Yeah that’s why I’m so interested in if there’s a correlation with severity and response. I want to see nice graphs of both dose and severity responses. That information could be really useful. Good point about there potentially being other, maybe even better ways of determining that if we can get over the first hurdle.