• Please Remember: Members are only permitted to share their own experiences. Members are not qualified to give medical advice. Additionally, everyone manages their health differently. Please be respectful of other people's opinions about their own diabetes management.
  • We seem to be having technical difficulties with new user accounts. If you are trying to register please check your Spam or Junk folder for your confirmation email. If you still haven't received a confirmation email, please reach out to our support inbox: support.forum@diabetes.org.uk

Aetiology of type 2 diabetes: an experimental medicine odyssey

Eddy Edson

Eddy Edson

Well-Known Member
Relationship to Diabetes
In remission from Type 2
Just-published Roy Taylor review of the "prolonged research endeavour to test the twin cycle hypothesis that type 2 diabetes is caused by fat-induced dysfunction of the liver and pancreas, guided by the happenstance of clinical practice."

A great resource!

link.springer.com

Aetiology of type 2 diabetes: an experimental medicine odyssey - Diabetologia

This review describes a prolonged research endeavour to test the twin cycle hypothesis that type 2 diabetes is caused by fat-induced dysfunction of the liver and pancreas, guided by the happenstance of clinical practice. Testing of the personal fat threshold hypothesis, that individuals exhibit...
link.springer.com link.springer.com



1746470722483.png
 
Thanks for this! Quite a lot to get though, here.

One of my biggest worries is that due to having higher levels for quite a while, I would've lost enough beta cell function to never be able to be fully cured. And while that is still a major concern, I am somewhat buoyed by this section:

Screenshot 2025-05-06 at 12.40.35.png
 
An interesting if very long read. It certainly gives hope for remission for a lot of type 2s.

Unfortunately I don't seem to fit the profile. I was overweight post covid and led a mostly sedentary lifestyle so I did diet and increased exercise, and BMI dropped from 29 to 26. Stopped dieting over Christmas and noticed in Feb that I was still losing weight although my calorie intake was back to around 2000 per day. Over the next 12 months with undiagnosed diabetes my BMI had dropped to 18 without dieting. I had little if any visceral fat by then and still have a slim waist/tummy now. Prior to diagnosis I had gradually increased my calorie intake to over 4000 per day and I felt permanently hungry and thirsty and was still losing weight.
Since diagnosis and medication plus a reduced carb diet of 2000 calorie per day I have slowly regained weight and muscle mass to a healthy BMI of 21 as my blood glucose has reduced. I am now struggling to continue reducing my BG and keep my weight at its current level.
 
The low level of carbs I need to keep my numbers normal has been remarked on, but for decades I was protesting about how carbs made me feel so ill, but I was still encouraged to eat high carb foods, low fat - and ten years before diagnosis I had a flagged high glucose level in some tests - that was dealt with by not repeating the test or letting me know about it - I still feel betrayed by the very people who should have been paying attention and helping me deal with my inability to cope with carbs - many others in my family had the same problem, but I am now the oldest survivor.
 
silver minion said:
An interesting if very long read. It certainly gives hope for remission for a lot of type 2s.

Unfortunately I don't seem to fit the profile. I was overweight post covid and led a mostly sedentary lifestyle so I did diet and increased exercise, and BMI dropped from 29 to 26. Stopped dieting over Christmas and noticed in Feb that I was still losing weight although my calorie intake was back to around 2000 per day. Over the next 12 months with undiagnosed diabetes my BMI had dropped to 18 without dieting. I had little if any visceral fat by then and still have a slim waist/tummy now. Prior to diagnosis I had gradually increased my calorie intake to over 4000 per day and I felt permanently hungry and thirsty and was still losing weight.
Since diagnosis and medication plus a reduced carb diet of 2000 calorie per day I have slowly regained weight and muscle mass to a healthy BMI of 21 as my blood glucose has reduced. I am now struggling to continue reducing my BG and keep my weight at its current level.
Click to expand...
Just checking that you & yr doc are sure you're T2? From your profile, I might have guessed at something T1-ish - FWIW.

Anyway, one of the things I don't like about Taylor's messaging is the implication that *every* T2 case is explained by the twin cycles model. I'm pretty sure it's true in most cases, but also pretty sure not every case.

(On the other hand, I don't believe in the usefulness of the common story you see about dozens or whatever of different species of T2. It's like hypertension - lots & lots of genetic combinations which can lead to it but that doesn't translate into lots & lots of different types from a clinical perspective.)
 
Eddy Edson said:
Just checking that you & yr doc are sure you're T2? From your profile, I might have guessed at something T1-ish - FWIW.

Anyway, one of the things I don't like about Taylor's messaging is the implication that *every* T2 case is explained by the twin cycles model. I'm pretty sure it's true in most cases, but also pretty sure not every case.

(On the other hand, I don't believe in the usefulness of the common story you see about dozens or whatever of different species of T2. It's like hypertension - lots & lots of genetic combinations which can lead to it but that doesn't translate into lots & lots of different types from a clinical perspective.)
Click to expand...
I have queried it but they seem to think I am T2 although AFAIK I haven't had any T1 blood tests. As my BG has dropped from 165 to low 70s they say I am still producing insulin.
On max dose of metformin and gliclazide now so keen to find out what my HbA1c will be this month. Hard to predict from FP testing as the results are still quite variable.
 
Eddy Edson said:
Anyway, one of the things I don't like about Taylor's messaging is the implication that *every* T2 case is explained by the twin cycles model. I'm pretty sure it's true in most cases, but also pretty sure not every case.
Click to expand...

Agree wholeheartedly with you there.
 
Docb said:
Agree wholeheartedly with you there.
Click to expand...
Obviously, some cases of T2 as diagnosed by doctors are (wholly or partially) caused by disease, drugs and who knows what else. ICD-11 provides a way of qualifying the diagnoses with 'stem codes' and 'extension codes', see link below. That approach should help to clarify the messaging between senders and receivers.

 
I've been testing out ChatGPT for various things. Have to say it's not bad at medical stuff,as far as I can tell - quite a lot better than Grok, FWIW. Anyway, here's what it gives me for applicability of the twin cycles model as an explanation of initially diagnosed T2D, by final diagnosis subtype:

1747029012211.png

So about 80%. FWIW, that's consistent with the view I had put together from reading a bunch of studies.

Of course most of the "misses" here are simply misdiagnoses, except for that top category where things are murky. So if you restrict the discussion to validated T2D diagnoses, the twin cycles proportion would be well above 80%, according to ChatGPT.

Note this does't speak to reversability/remission potential. Just because the twin cycles model explains the origin of the diabetes it doesn't mean that weight loss will necessarily reverse it - eg the "Advanced t2d with beta-cell failure" group, where the beta cells are toasted beyond recovery. And obviously nothing guarantees weight loss sustainability!

It also doesn't necessarily fit with people's views of themselves. Eg the common situation of people saying they don't fit this model because they are "not overweight" and "healthy", and then when you dig they have a waist-to-height ratio >0.50 (which is the important metric for predicting excess visceral fat, much more so than BMI), and signs of insulin resistance as suggested by a simple tool like the Triglyceride Glucose (TyG) Index.

Even though the numbers here are just ChatGPT's estimates, IMO it's fair to say that the actual number of cases of T2D not amenable to a twin cycles explanation is quite small, in relative terms.
 
Last edited:
Eddy Edson said:
I've been testing out ChatGPT for various things. Have to say it's not bad at medical stuff,as far as I can tell - quite a lot better than Grok, FWIW. Anyway, here's what it gives me for applicability of the twin cycles model as an explanation of initially diagnosed T2D, by final diagnosis subtype:

View attachment 35340
Click to expand...

@Eddy Edson
That's an interesting set of results. Haven't had time to get to grips with AI at all yet. Would you be good enough to outline what you had to do to get that information from ChatGPT? Was it a single question or a set of questions? Did you have to specify constraints or provide some form of model? Etc?

Thanks.
 
Last edited:
JITR said:
@Eddy Edson
That's an interesting set of results. Haven't had time to get to grips with AI at all yet. Would you be good enough to outline what you had to do to get that information from ChatGPT? Was it a single question or a set of questions? Did you have to specify constraints or provide some form of model? Etc?

Thanks.
Click to expand...

Hopefully this works as a link to the session this occurred in: https://chatgpt.com/share/6825df1d-2224-800f-b4ec-a60c62f4972a

Bunch of other stuff in that session as well; the twin cycles piece is towards the end. You'll notice that ChatGPT tends to enthusiastically agree with & adopt critiques of its initial responses, which makes me a bit nervous that it's tuned to be a yes-man 🙂
 
Eddy Edson said:
Hopefully this works as a link to the session this occurred in: https://chatgpt.com/share/6825df1d-2224-800f-b4ec-a60c62f4972a

Bunch of other stuff in that session as well; the twin cycles piece is towards the end. You'll notice that ChatGPT tends to enthusiastically agree with & adopt critiques of its initial responses, which makes me a bit nervous that it's tuned to be a yes-man 🙂
Click to expand...

Brilliant. Yes it works, yes-man or not.

Thank you.
 
I think the lean onset w/o ectopic fat group is an interesting one and it is possible that it too may be impacted by the Taylor twin cycle model.

As someone who did not fit the "typical" type 2 diabetic patient (not overweight, regularly exercised and eat freshly made meals with) on diagnosis and who had it confirmed that my liver and pancreas were in good health with no excess fat, within three months of diagnosis, the twin cycle model, in theory does not apply to my T2D.

However, when I look back at my lifestyle over the decades I know that when I first moved to London back in early 80s and through until my early 30s, I did not eat particularly well with lots of processed food and definitely an excess of bread and other high carb foods. During my 40s I started to exercise more regularly, frequently and increased exertion but in my late 40s and early 50s I fell back to old habits due to working away a lot plus I started eating to many sweet snacks whilst driving on long journeys. At sixty my exercise level was up but my diet was again heavy with carbs and sugars due to working away from home.

From my perspective, I think I had undiagnosed diabetes for potentially 12 months before the diagnosis and I do wonder if the "lean onset w/o ectopic fat group" are actually people whose diagnosis occurs after the issue with fatty livers and pancreas has been addressed through other factors such as taking up cycling, running or other activity which many people did in and during the lockdowns/covid period or for some other reason and thus the delayed diagnosis means that those issues are addressed but an individual has yet to reach remission? Of course the fact that my HbA1c at diagnosis was 86 suggests that possibly the twin cycles does not apply as I would have expected that removing the fat from the organs would have taken me to remission.

I think therefore that there is still more research to be done to fully understand T2D in all its forms, although it appears that for a specific class or classes of patient, the twin cycles model seems to be a very good fit. Time has shown that clinically there are a number of forms of diabetes and while causes for T1 including some variations is well understood and also type 3c, type 2 still has some mysteries.
 
Strangely, my weight is not far off what it was when I was diagnosed, but clothes sizes are still the same as they were when I lost weight initially and there has been no change in my body shape.

I suspect that I've gained muscle mass and maybe glycogen mass as well. Perhaps due to improved metabolism - muscle mass improving and more glycogen getting stored, both due to improved insulin sensitivity.
 
JimmyBlue said:
As someone who did not fit the "typical" type 2 diabetic patient (not overweight, regularly exercised and eat freshly made meals with) on diagnosis and who had it confirmed that my liver and pancreas were in good health with no excess fat, within three months of diagnosis, the twin cycle model, in theory does not apply to my T2D.
Click to expand...

A couple of questions, plus one, if I may ask:
  1. How long after diagnosis was 'it confirmed that my liver and pancreas were in good health with no excess fat'?
  2. Did you lose any weight at all during that time around your waist?
  3. How much more exercise were you doing?
A fibroscan confirmed my liver was back to normal after four months and an HbA1c of 32 after six months indicated my pancreas was doing its job. I could get back into my old trousers again.
 
@JITR the examination was done in the December so about 2-3 months after diagnosis. I had lost weight but this was down to not enough carbs and ketosis kicking in, my HBa1c only fell to 47 at the start of Jan this year.
Exercise wise I went from 2x6K runs to 3x6k.

I went from 13 stone 4lbs to 12 stone 8lbs. I had also had a scan in the September to check my liver as I originally thought my issues were down to prostate or cystitis. That review raised no concerns about liver fat either.
 
@JimmyBlue
Your HbA1c of 86 last September suggests all was not well with your liver and pancreas, and HbA1c of 47 in January indicates whatever you did in in October to December corrected that.

Have you seen details of Prof Taylor's ReTUNE study of T2Ds of normal BMI in the paper in the OP and ref [67] of the paper?

Why not write to Roy Taylor at Newcastle University with a more detailed chronology? He would be able to tell you how your experience could fit the twin cycle hypothesis, or yours, as the case may be.
 
Last edited:
You must log in or register to reply here.
Back
Top