I recently posted an introduction to Maturity Onset Diabetes of the Young (MODY). This is of personal interest to me as I’ve had a suspicion that I might have MODY for more than five years. This last April I was shocked to actually be referred for the genetic tests for MODY. So being able to attend the 75th ADA Scientific Sessions and learn more about MODY and have a chance to hear the worlds experts and even ask some questions was a great opportunity. In this post I’ll be talking about my attendance at an ask the expert session on MODY by Dr. Rochelle Naylor of Kovler. In subsequent posts I’ll talk about a presentation by the infamous Professor Andrew Hattersley of Exeter as well as my viewing of the film “Journey to a Miracle” a movie about MODY.
I’m reading and learning, Brian. Good luck with your continuing discovery.
Great stuff, though I might have been tempted to keep it all in one thread.
On the very reasonable question of why elevated A1cs in MODY-2 (GCK-MODY) patients may not be a problem when the same levels in the general population, including more typical T1s and T2s, seem to be associated with various complications, I guess it hinges on what is causal.
In the general population including various degrees of what gets called T2, elevated A1cs are a signal of a host of other, underlying things going wrong, as part of what people call “metabolic syndrome,” that are harder to measure directly, e.g. low levels of systemic inflammation. So elevated blood sugars may be markers of those other things rather than causing harm in their own right.
For those with GCK-MODY mutations, elevated blood sugars have no relationship to those other underlying conditions–they’re just elevated due to a gene defect that establishes a higher blood sugar set point. So if blood sugars are more signal than cause, they those elevated GCK-MODY sugars may not matter in the way the same sugars in a more typical person would.
If the same levels of elevated sugars in both Type 1 and Type 2 correlate with the same complications, that would suggest the argument I’m making here is wrong, because the Type 1 blood sugars are elevated for wholly different reasons (lack of insulin). And there’s some evidence for just that. But I do think these issues are central to the argument the MODY researchers are making, even if I’m not totally sold on that argument at this point (or at least still trying to understand it better).
There’s another wrinkle. Most T2s, and almost all T1s, get worse over time. GCK-MODYers effectively stay the same over time (slight progression over time). So it’s hard to compare the static situation on the one hand with a dynamic disease progression on the other. So that makes the apples to oranges comparison yet more complicated.
I actually was going to have a single post but it is clear that it was just too much. These are pretty esoteric topics and I wasn’t sure if most readers could make it through in one bite.
And while I understand your argument that maybe the issue in T2 is that there is underlying metabolic syndrome, I think there are reasons to think that isn’t the case. Studies (such as this one here) seem to show that HbA1c is a CVD risk factor in non-diabetics independent of BMI, blood pressure and cholesterol. This suggests that it is high blood sugars themselves (or some other unobserved defect) that is associated with CVD. Until I see other evidence I’m inclined to think it is high blood sugars and thus I just remain skeptical.
ps. So the good news is that I actually was able to speak with Dr. Naylor after her session and she was agreeable to an interview of some form.
Great! I’ve been engaging the Kovler folks on these issues also. My impression is that elevated sugars are a marker of systemic inflammation, on a spectrum from non-diabetic to pre-diabetic through T2. The fact that there’s some evidence that modestly elevated sugars in T1s also correlate with complications does complicate things, though.
This article, which we’ve discussed before, seems to be the main piece of evidence the Kovler folks cite for the assertion that having GCK-MODY level blood sugars for long periods doesn’t appear to be associated with additional complications relative to similar individuals who lack the GCK-MODY mutation (two links below in case one doesn’t work). It’s not a huge sample, but not a tiny one, either.
I consider this a tiny sample. Here is my reasoning. The DCCT had 1,441 participants and basically found that for A1cs above 7% there was an increased CVD risk. The only risks that were quantifiable at A1cs of 7% or less seemed to be retinopathy and that is how the 7% is ok was established. If even the DCCT could not resolve CVD risk at A1cs below 7% we can’t expect to get strong evidence from a MODY study with 100 participants in that range. I just want stronger evidence before I place all my bets on running high blood sugars for the rest of my life.
I hear you. I’m looking forward to your engaging with the Kovler researchers further, because I’m curious to hear their responses to your very reasonable points, since they and the Exeter folks both seem very confident in their assessments.
You’d think that 1441 participants would shed light on all but the most subtle effects, though I suppose the key question is the much smaller participant subsets at various A1c levels, including relatively low ones.
In the case of the study in question, I guess the key question is what are the odds that 99 GCK mutation carriers show effectively identical complication rates to 91 non-GCK mutation carriers if it is actually the case that the GCK mutation causes meaningfully higher complications than not having it. Put differently, what are the odds that a weighted die rolled 100 times will yield the same results as an unweighted die rolled 100 times, i.e. won’t skew toward the number toward which it is biased over that many rolls? In addition to the sample sizes, obviously the magnitude of the possible weighting/complications is key, i.e. more subtle weighting/complications are harder to discern. But my stats knowledge is pretty rusty these days.
Maybe you can help me formulate some questions to pose to Dr. Naylor? I’m going to talk with @EmilyC about a live interview, but I may just email her some interview questions and then write up the interview.
Happily! This is really helpful for my own grappling with these issues.
Re the study we’re discussing, here’s what I take away from it:
-
A study cohort with an average of almost 50 years of modestly elevated blood sugars due to GCK-MODY gene mutations (which by definition they had from birth) demonstrated very few complications
-
Complication rates for GCK-MODYers were similar to those for family members without the mutation and with consequently lower blood sugar levels, but the sample size was too small to be confident that the two groups were identical–but if they were not identical, the difference must be very subtle
-
Complications rates for GCK-MODYers were much lower than for a control group diagnosed with T2 and with meaningfully higher blood sugar levels, but by definition for much shorter periods of time (the controls developed T2 later, they haven’t had it since infancy)
As I have posted before, many people with MODY-2 diagnoses have family histories of early heart attack. My family, which carries a different insulin sensitive diabetes gene does too.
When you see your thin, fit, active relatives dropping dead in their 50s from heart attacks as I did, you realize these doctors are missing something.
MODY-2 may not cause traditional diabetic complications but lifetime exposure to just “prediabetic” blood sugars cause heart disease, vertebral disc deterioration, and fragile tendons, all of which greatly decrease quality of life,
Rephrase, “many people [with MODY-2 diagnoses] have family histories of early heart attack.” This is where the scientific method comes in. Human beings are notoriously bad at diagnosing patterns, or their absence, from anecdote. It is quite plausible that MODY-2 carriers might have elevated risks of heart attack, or other complications. But that’s an impulse to try to study the topic more rigorously.
In the study we’re discussing, which seems to be the best evidence we have on this topic (though there are other studies on it, too), the rates of heart disease among a pool of 99 MODY-2 carriers were effectively identical, actually slightly lower than, the rates of heart disease among a pool of 91 family members who do not carry the MODY-2 gene mutation. This despite the fact that the GCK-MODY carriers have meaningfully higher average blood sugars, as expected, than their relates who do not carry the gene mutation, and by definition they’ve had those higher blood sugars their entire lives, since before birth.
Yes, these folks, both the gene mutation carriers and the non-gene mutation carriers, have a spectrum of blood sugar levels, some of which we’d categorize as pre-diabetic. But that’s the point. What the study is doing is a “natural experiment” where family members who are presumably otherwise similar (other genes, lifestyle, diet, etc.) but who differ on one crucial variable (whether they have the MODY-2 gene mutation or not) are being compared. And the study strongly suggests that having the MODY-2 gene mutation doesn’t increase heart disease risks, despite the life-long elevated blood sugars that gene mutation entails. In fact, if there is a difference in the population, if anything the gene mutation may lower the risk of heart disease. The prevalence of heart disease is too low for the comparison to be statistically significant at these sample sizes, i.e. we can’t be sure that gene mutation carriers have a lower, or an identical, risk of heart disease as non-gene mutation carriers, controlling (via natural experiment) for other relevant factors, but it’s hard not to be at least somewhat reassured by the results.
And for what it’s worth, I really welcome @Jenny’s engagement, because it’s a chance to push myself to try to articulate what I think, and where we may agree and disagree. So I hope we can continue spirited dialogue without it becoming personal, because I think that serves us and the broader community on here.
Niccolo could you point me to that study?
There is just too much other data connecting increased risk of heart attack with prediabetic post-meal blood sugars for me to buy into this without a whole lot more supporting data. So because I apply Pascal’s Wager to questions involving risk, I ask, “What happens if I’m wrong?”
If I’m wrong and it is a waste of time and effort to achieve completely normal blood sugars, then I have missed out on a lot of fine pastry and tested my blood sugar a lot, sad, but not fatal.
If I’m wrong about it being safe to ignore those post-meal numbers, on the other hand, I die young of heart disease.
I made the decision to ignore what doctors told me back in 1998–that I should ignore my very high post-meal numbers because my fasting numbers were normal (they redefined normal fasting a few years later making mine prediabetic.) It’s 17 years later and I have zero diabetic complications except for the disc and tendon problems which only specialists seem tor realize are blood sugar related. I have missed out on a lot of pastry and done a lot of testing, injecting, etc. I don’t know what would have happened to me if I had made a different decision at age 50.
And since I am too old to die young, I’m eating a bit more pastry now. What I’ve seen of people living into their 90s hasn’t motivated me to want to do that.
There are two different links to it a few posts up.
I hear you on Pascal’s Wager. But it’s more complicated. If you have GCK-MODY, most therapies don’t seem to do much for your blood sugars, with the possible exception of low carb, which might affect them a little.
Based on the data, we can’t be sure that failing to pursue that has zero consequences. But we can be fairly confident that at most it results in a very modest increase in the risk of heart disease, not a binary choice between no heart disease and surely dying of heart disease!
Here is the study:
I would by no means interpret the results either as being statistically significant or strong and given that 30% of GCK patients had retinopathy I would hardly claim that GCK provides invulnerabiity. I would also note that a study of GCK patients that have “survived” for 50 years does not give you any measure of how many GCK patients died from CVD before the study occured.
We’re talking about heart disease, where I suppose you might say the fact that those lacking the GCK-MODY mutation’s apparent higher risk of heart disease has troubling implications for those not fortunate to have one of these gene mutations. 
But obviously a larger study might detect a subtle difference between the GCK-MODY and non-mutated populations, and maybe the direction of the difference would even reverse and the GCK-MODYers would be at higher risk of heart disease.
I’m not sure who claimed “invulnerability.” The higher prevalence of mild retinopathy is notable, and suggests that those elevated blood sugars do appear to have some effect (which hints at possible other, albeit more subtle, effects).
It still raises the question of what one proposes to do about it, though. My hunch, based on my own experiences and the literature, is that medications, including insulin, really won’t do much, but low carb can have some effect on curbing post-prandial excursions, and modestly affecting average blood sugar levels. And ultimately I suspect I’ll end up eating a moderately low-carb diet, which isn’t so far off my current practice.
Also, the study did explicitly address the potential survival bias issue. How effectively it did that pushes the limits of my current methodological skills. I take it you found those parts unsatisfying?
My real concern is prematurely giving MODY patients the message that they can just not manage blood sugars and everything will be ok is not appropriate. This is one my criticisms of the movie I saw at ADA. We do need better evidence.
And I still have major concerns over the bias issues.
I hear you. And your “invulnerability” comment was appropriate, because that is the broader message that both Kovler and Exeter, the two most well-informed research centers regarding monogenetic diabetes in the world, are both sending to patients and to their caregivers.
Perhaps I have an establishment bias, and I get that that’s in tension with what is arguably an anti-establisment bias among lots of other folks, but if I have to choose between the hypothesis that Kovler and Exeter are both misreading the data, whether due to ineptness and/or some sort of bias, or that they’re reading it correctly, my own bias is toward the former.
But I don’t fully understand why they reach the conclusions they do, so at a minimum I’m hoping to understand that better. And I don’t mind reaching conclusions in tension with theirs, either, but I’d want to have a pretty compelling explanation for where I thought they went astray.