Hyperglycemia caused by rapamycin

[cl-user]

Glucose hyperglycemia caused by rapamycin is not related to insulin resistance but to hepatic glucose production, at least in my case study.

First here is my 24h average glucose level measured by CGM calibrated with finger glucose.
I was off rapamycin for a few months and restarted 12mg every other week in May.
The flat part are when I didn’t wear a CGM.

Clearly not good for sure.
BTW I’m taking a SGLT2i (empagliflozin 12.5mg/day then 25mg/day from May), Tirzepatide 4mg/week and Acarbose 100mg/meal.
All that did not prevent the glucose rise.

Then I did an Oral Glucose Tolerance Test (OGTT) which is ingesting a bolus of 75g of pure glucose in 300ml of water. That was a horrifying experience for somebody eating low carbs!

I was really surprised to see that it was better after rapamycin. I think the theory here is that as I eat low carbs, my pancreas is not used to such a large bolus of glucose and so the first OGTT primed it for the second one. In fact it’s recommended that low carbs [people eat at least 150g of glucose for 2 days before an OGTT.
My fasting insulin was 5.8 before rapamycin and 6.8 after which a rather mild increase.
Fasting glucose at the same time (7:30AM) was 99 then 97.
HOMA-IR went from 1.4 to 1.6 which is still OK.
Clearly rapamycin did not cause a high insulin resistance.

Let’s look at the liver glucose production below:

As it’s rather noisy with the peaks due to eating (noon and 6) and exercising (around 9~10) I used a low pass filter to smooth it:

The green one is after 3 months without Rapamycin and stays below 100mg/dl on average which is rather good.
The green one is 3 days after rapamycin and clearly it’s very bad. It even stays above 105 mg/dl at night.
2 weeks after rapamycin there is some improvement but it’s still very bad.

We know there is no issue with the insulin resistance because the OGTT is completely OK so the glucose could go down. It’s really an issue with regulation here where the liver generates all that glucose independently of the meals.

Now after 3 weeks my hepatic glucose is around 105 during the day which is still not good.

All that with 25mg/day of Empagliflozin which makes me pee glucose like crazy (Checked with urine strips).

[CronosTempi]

@cl-user

BTW I’m taking a SGLT2i (empagliflozin 12.5mg/day then 25mg/day from May), Tirzepatide 4mg/week and Acarbose 100mg/meal.
All that did not prevent the glucose rise.

If you suspect hepatic glucose production elevation, then the only agent which addresses hepatic gluconeogenesis is pioglitazone. Metformin does not.

All that with 25mg/day of Empagliflozin which makes me pee glucose like crazy (Checked with urine strips).

Yes, there are studies showing that empagliflozin use increases neoglucogenesis in a compensatory manner, so it’s not shocking that if hepaticneogenesis is your issue, empagliflozin doesn’t have much effect. I’m in the same boat, which is why I’m exploring low dose pioglitazone (7.5mg/day), perhaps sometime next year.

[RapamycinCurious]

Thanks for sharing your data! I think it’s premature to conclude that your rapa-induced hyperglycemia is not due to insulin resistance on the basis that your OGTTs are normal or even improved on rapa is premature. First, you’re on empagliflozin, which blunts your OGTT regardless. And second, excess hepatic glucose output is a normal result of hepatic insulin resistance: that’s why metformin works. So a reasonable reading is that rapa may be causing you hepatic but not muscular insulin resistance.

[qBx123Yk]

Interesting experiment, thanks for sharing your data!

[CronosTempi]

Btw. here’s the study establishing that pioglitazone inhibits hepatic gluconeogenesis, while metformin does not (I’ve posted this study before). The study is in DMT2, but no reason to suspect that it’s any different in pre-diabetic or normal glucose individuals.

Comparison of the Effects of Pioglitazone and Metformin on Hepatic and Extra-Hepatic Insulin Action in People With Type 2 Diabetes

https://diabetesjournals.org/diabetes/article/57/1/24/40558/Comparison-of-the-Effects-of-Pioglitazone-and

“[…]pioglitazone enhanced (P < 0.01) insulin-induced suppression of both glucose production (6.0 ± 1.0 vs. 0.2 ± 1.6 μmol · kg−1 · min−1) and gluconeogenesis (n = 11; 4.5 ± 0.9 vs. 0.8 ± 1.2 μmol · kg−1 · min−1). Metformin did not alter either suppression of glucose production (5.8 ± 1.0 vs. 5.0 ± 0.8 μmol · kg−1 · min−1) or gluconeogenesis (n = 9; 3.7 ± 0.8 vs. 2.6 ± 0.7 μmol · kg−1 · min−1).”

Meanwhile here’s a study showing that empagliflozin causes compensatory increase in gluconeogenesis:

Inhibition of Sodium-Glucose Cotransporter-2 during Serum Deprivation Increases Hepatic Gluconeogenesis via the AMPK/AKT/FOXO Signaling Pathway

Inhibition of Sodium-Glucose Cotransporter-2 during Serum Deprivation...

“These data show that SGLT2 mediates glucose uptake in hepatocytes and that SGLT2 inhibition during serum deprivation increases gluconeogenesis via the AMPK/AKT/FOXO1 signaling pathway.”

Low dose pioglitazone (7.5mg/day) is as about as effective at lowering A1C and FBG as full dose or high dose, but with fewer side effects:

The Low-Dose (7.5 mg/day) Pioglitazone Therapy

The Low-Dose (7.5 mg/day) Pioglitazone Therapy

“The low-dose pioglitazone therapy may show the same degree of improvements in glucose and lipid metabolism, fatty liver, insulin resistance, and adiponectin as the standard- and high-dose pioglitazone therapy. Furthermore, the low-dose pioglitazone therapy may also show less adverse effects on weight gain, edema and heart failure as compared with the standard- and high-dose pioglitazone therapy.”

[Rapan]

@cl-user

We know there is no issue with the insulin resistance because the OGTT is completely OK so the glucose could go down. It’s really an issue with regulation here where the liver generates all that glucose independently of the meals.

I think you cannot say this without measuring at least insulin levels during OGTT. Your insulin levels might be in a prediabetic range postprandial indicating insulin resistance with rapamaycin.
Some might say even dynamic insulin level measurement is not sufficient and it requieres hyperinsulinemic euglycemic clamp to measure insulin sensivitiy or resistance. It has been shown by Defonzo that e.g. healthy kids have normal glucose repsonse in OGTT, but are insulin resistant.

[John_Hemming]

I take a high dose (higher than you) of Rapamycin, but quite infrequently. I posted the results from one here:

Second High Dose of Rapamycin topic for results

On September 23rd I took 1 packet of biocon and 2 packets of zydus with GFJ and a bit of pomegranate. The Grapefruit was half eaten a while before the rapamycin and half with the rapamycin. This time I had been wearing a CGM (One+) for a few days before taking the rapamycin. It fell off and I got another, I then used a third on schedule, but it stopped uploading on Sunday and that got fixed part way through Tuesday. It will take some work to produce useful reports from this, but I have already done an average glucose chart. I have excluded the days with part data as they will not have useful averages. The results are skewed a bit by alcohol consumption. Alcohol takes down the average glucose. The results, however, still show a useful pattern of Rapamycin initially driving up the average glucose figure. Interestingly it appears that it is the baseline that moves up moreso than the peaks. The body adjusts to produce more insulin, but the rapamycin fades and so at a point when…

What I see happening from a single dose is
a) A development of hepatic insulin resistance
b) An over compensation by producing more insulin.
c) The fading of insulin resistance
d) An overswing where glucose goes low
e) back to normal.

With 12mg fortnightly your trough is still going to have some mTOR inhibition which I think is the mechanism for the hepatic insulin resistance. I don’t think I would want to take that much that frequently.

With my large, but intermittent dosing I get back to normal.

[cl-user]

@CronosTempi

If you suspect hepatic glucose production elevation, then the only agent which addresses hepatic gluconeogenesis is pioglitazone. Metformin does not.

Yes indeed. I’ve currently added 1000mg Metformin and I don’t see any improvement.

[cl-user]

@RapamycinCurious

And second, excess hepatic glucose output is a normal result of hepatic insulin resistance: that’s why metformin works. So a reasonable reading is that rapa may be causing you hepatic but not muscular insulin resistance.

Metformin never worked for me. I tried it several times in the past with various doses without any change in glucose. I’m currently re-trying it to see it it could work in addition to the SGLT2i and the GLP1-RA but no change either.

Also the OGTT shows that I can clear 75g of glucose in less than 2h so, whatever amount of insulin used for that, it was enough. My issue is that, even fasted, I’m at 120mg/dl with rapamycin vs less than 100 with rapamycin.
If we look at the daily CGM plots we see that with rapamycin, the fasted baseline is around 120 and that the postprandial spikes after the 2 meals went back down to that baseline (below it even).
Basically the glucose regulation works but with a fasted baseline of 120 instead of less than 100.

[cl-user]

@CronosTempi

here’s the study establishing that pioglitazone inhibits hepatic gluconeogenesis, while metformin does not (I’ve posted this study before).

Thanks for the references. Very interesting studies for sure. I will try low-dose pioglitazone next.

[cl-user]

@John_Hemming

With 12mg fortnightly your trough is still going to have some mTOR inhibition which I think is the mechanism for the hepatic insulin resistance. I don’t think I would want to take that much that frequently.

With my latest blood rapamycin measurements I get a trough at 0.3 ng/ml after 2 weeks and 0.05ng/ml after 3 weeks when things start to normalize again.
Also at 12mg the max at equilibrium is around 8 ng/ml so maybe it’s too high and cause mTOR2 inhibition as we know that this can be caused by staying too long at some level but also with a level too high.

[John_Hemming]

A lot of what is called “normal” is worth having. So my plan is to be in that range a lot of the time.

[SteveRoedde1]

A very interesting post. Thank you.
I have steroid-induced DM. 10 years of 15-50 mg of prednisone. I take pioglitazone, empagliflozin and acarbose (with carbs). Lean (11%fat on Dexa), aerobically fit (6+ h per week plus weights), and also carb restricted. Gluconeogenisis is a huge factor here, and even fasting/exercise had zero effect before the meds, and improved but still present after.
8 mg rapamycin q 2 weeks had similar effects. Only detected via random blood pricks.
I ended up buying a CGM (off rapa) and did two 75 g GTT’s. One before daily prednisone (nearly normal), and one during peak steroid effect (not normal!).

I will stay the course for now, but once the neurosis-inducing effects of the CGM have worn off, I’ll repeat those trials on rapa.
Again, gratitude for your post.

[SteveRoedde1]

I used CHBT to calculate AUC and predicted insulin requirements in my two OGTT’s see post below for why I did them). Big increase in both.
Again, grateful for your post.

[SteveRoedde1]

This is purely speculative on my part, but I suspect that the body interprets rapamycin as a “stressor”, with HRV, HR, impacts and (in those not adrenally suppressed), a cortisol response.
I wonder if it is cortisol, not MTORC2 that is driving this short term response.

[cl-user]

@SteveRoedde1

I wonder if it is cortisol, not MTORC2 that is driving this short term response.

Looks like it’s a direct effect of Rapamycin suppressing mTORC1 and not a side effect (or mTOTC2 inhibition) as I thought.
Here is a detailed explanation from ChatGPT 4.5:

1. Rapamycin’s Effects on mTORC1:

• Rapamycin primarily inhibits mTORC1, a central nutrient-sensing pathway.
• mTORC1 inhibition promotes a fasting/starvation-like metabolic state, signaling tissues (particularly the liver) to increase hepatic glucose production (gluconeogenesis and glycogenolysis).

2. Increased Hepatic Glucose Output:

• Inhibition of mTORC1 reduces hepatic insulin signaling efficiency, especially regarding insulin’s action on suppressing gluconeogenesis.
• Even though peripheral insulin sensitivity (muscle/adipose tissue glucose uptake) may remain intact or even improve, hepatic insulin resistance specifically in the context of glucose production occurs due to impaired insulin action on the liver under rapamycin exposure.
• This selective “hepatic insulin resistance” is a hallmark of rapamycin’s metabolic effect.

3. Role of FOXO and Gluconeogenesis:

• mTORC1 inhibition results in reduced Akt-mediated phosphorylation of FOXO transcription factors.
• Reduced phosphorylation allows FOXO transcription factors to migrate into the nucleus, upregulating gluconeogenic genes like PEPCK and G6Pase, thus enhancing hepatic glucose production.
• Consequently, baseline hepatic glucose output increases, leading to elevated fasting glucose levels.

4. Dissociation between Hepatic and Peripheral Insulin Sensitivity:

• As demonstrated by the normal OGTT result, the patient’s peripheral tissues (muscle/adipose) effectively clear glucose after glucose administration, showing preserved or even enhanced peripheral insulin sensitivity.
• However, fasting hyperglycemia reflects specifically impaired insulin signaling in liver tissues during fasting conditions, making the liver insensitive to insulin’s gluconeogenesis-suppressive effects.

Then I got it to explain Why is rapamycin-induced glucose elevation termed as hepatic insulin resistance?

• Under normal conditions, insulin suppresses gluconeogenesis via the Akt pathway, leading to phosphorylation (and inactivation) of the transcription factors FOXO1 and FOXO3, reducing expression of gluconeogenic enzymes (PEPCK, G6Pase).

• If this signaling is disrupted, the liver produces glucose excessively—even in the presence of insulin—leading to elevated fasting glucose.|

Rapamycin inhibits mTORC1, a key nutrient-sensing kinase downstream of insulin and other growth signals:

• Normally, insulin signaling through the liver activates Akt → mTORC1, contributing to suppression of gluconeogenesis.
• Rapamycin blocks mTORC1 downstream of insulin signaling, thus removing an essential step in insulin-mediated gluconeogenesis suppression.
• The net result: despite normal insulin levels and insulin-sensitive muscle/adipose tissues (as confirmed by a normal OGTT), the liver “behaves” resistant to insulin’s glucose-suppressing effects.

Thus, this situation functionally appears identical to insulin resistance, even though the primary issue isn’t a defect in insulin receptors or insulin availability per se—but rather inhibition of downstream insulin signaling (specifically mTORC1) within hepatic cells.

Functional Differences:

Aspect	Hepatic Insulin Resistance	Peripheral Insulin Resistance
Main issue	Excessive liver glucose output	Reduced glucose uptake by muscle/fat
Predominantly affects	Fasting glucose	Postprandial glucose
OGTT result	Often normal (quick return to baseline)	Impaired (slow return to baseline)
Baseline glucose	Elevated fasting glucose	Elevated after meals; fasting glucose often normal initially
Insulin secretion	Usually adequate or increased	Usually adequate or increased initially

[cl-user]

@CronosTempi

here’s the study establishing that pioglitazone inhibits hepatic gluconeogenesis, while metformin does not

Here is it explained by ChatGPT 4.5

Why Metformin May Have Been Ineffective:

• Metformin primarily suppresses hepatic glucose production through AMPK activation and reduction of gluconeogenic enzymes.
• However, rapamycin-induced hepatic insulin resistance results largely from disrupted insulin signaling downstream of mTORC1, via FOXO and Akt pathways, rather than AMPK-dependent mechanisms.
• Therefore, metformin’s AMPK-mediated mechanism may not be sufficient to counteract the rapamycin-induced hepatic glucose output effectively in certain individuals.

Potential Advantages of Pioglitazone:

• Pioglitazone is a PPAR-γ agonist (thiazolidinedione), primarily enhancing insulin sensitivity by improving hepatic and peripheral insulin signaling.
• Pioglitazone’s insulin-sensitizing effects are independent of AMPK activation, instead relying on PPAR-γ mediated transcriptional regulation, reducing hepatic gluconeogenesis indirectly through improved insulin action.
• In addition, pioglitazone shifts metabolic balance by:
  • Reducing hepatic fat content, which further improves hepatic insulin sensitivity.
  • Reducing systemic inflammation, which may be relevant since rapamycin also influences inflammatory signaling.
  • Increasing adipose tissue insulin sensitivity, reducing systemic free fatty acid levels, indirectly reducing hepatic glucose production.

Clinical Relevance and Evidence:

• Studies have shown pioglitazone effectively lowers fasting glucose levels and hepatic glucose production, particularly in conditions characterized by hepatic insulin resistance, such as NAFLD (nonalcoholic fatty liver disease) or type 2 diabetes.
• Given the patient’s normal peripheral insulin sensitivity but elevated fasting glucose due specifically to rapamycin-induced hepatic insulin resistance, pioglitazone is a logical therapeutic strategy.

[cl-user]

Update: Started to take Pioglitazone 7.5mg/day (PZ)
Looks like that’s finally something that works for me. If you look at the magenta curve (3 days after rapamycin+PZ) it’s not too bad especially compared to the red line (3 days after rapamycin) or even 2 weeks after rapamycin without PZ.
BTW it’s the tri-therapy pushed by Ralf DeFronzo: Pioglitazone + GLP1-RA (Tirzepatide) + SGLT2i (Empgliflozin). It’s even a quadri-therapy when I take acarbose with meals.

[CronosTempi]

That’s depressing. I guess I should congratulate, but I’m rather enervated by these graphs. A freakin’ quadri-therapy and that’s all it gets you. I would’ve hoped for absolutely stellar results. What chance do I stand, when I’ll be limited to only a dual-therapy - I am on SGLT2i (empagliflozin) and am contemplating pioglitazone, that will make it TWO, possibly, possibly incorporate acarbose. I have no current plans for GLP-1RA. I really was hoping for more from the addition of pioglitazone. I suppose I will suspend judgment until you have a reading of A1c. If pioglitazone brings A1c to some civilized level, I would take that as a cautious win.

Right now that green curve looks reasonable, although it starts trending up later toward 26+ hours, where it crosses the magenta line (plus toward the beginning, same situation). I would’ve wanted pio to look like the best parts of green and magenta curves

[cl-user]

@CronosTempi

I’ll be limited to only a dual-therapy - I am on SGLT2i (empagliflozin) and am contemplating pioglitazone

I think it might work if like me the issue is with excessive hepatic glucose production. In that case PZ is more useful than a GLP1-RA.
That said the idea behind the try-therapy is that the GLP1-RA will block the possible weight gain from PZ. So far I my weight has not changed either way.

@CronosTempi

I would’ve wanted pio to look like the best parts of green and magenta curves

Me too but the magenta is 3 days after rapamycin (12mg). I will post how it will be 3 weeks after rapa. Hopefully it will be good as there is not much I can add to that.