The mixed meal tolerance test (MMTT) measures pancreatic beta-cell function by tracking C-peptide and glucose after a standardized liquid meal — typically Boost or Sustacal containing about 50 grams of carbohydrate plus protein and fat. The test is the gold standard for assessing residual beta-cell function in type 1 diabetes clinical trials, pre-transplant evaluations, and ambiguous diabetes classification cases. It is more physiologic than the OGTT and better tolerated than glucagon stimulation, with AUC C-peptide over 2 hours as the standard endpoint.
What the MMTT Is
The MMTT is a dynamic test of how your pancreas responds to a meal. Instead of measuring only fasting beta-cell tone (as with fasting C-peptide), the MMTT challenges the pancreas with a defined caloric load containing all three macronutrients — carbohydrate, protein, and fat — then samples blood at multiple time points to capture the timing and magnitude of insulin and C-peptide release.
Because C-peptide is co-secreted with insulin in a 1:1 molar ratio but cleared more slowly (half-life about 30 minutes versus 5 minutes for insulin), it gives a more stable measure of pancreatic output. The integrated C-peptide response over 2 hours — area under the curve, or AUC C-peptide — is the primary endpoint of the test.
Standard Protocol
| Step | Detail |
|---|---|
| Preparation | Overnight fast (at least 8 hours, water allowed); hold short-acting insulin overnight per protocol |
| Baseline draw | Time 0: blood for C-peptide, glucose, sometimes insulin |
| Test meal | Boost or Sustacal liquid: 6 mL/kg body weight, up to maximum 360 mL — consumed within 5 to 10 minutes |
| Composition (typical 360 mL Boost) | About 360 to 400 kcal: 50 g carbohydrate, 14 g protein, 9 g fat |
| Sampling timepoints | 15, 30, 60, 90, 120 minutes (some protocols add 180 and 240) |
| Total clinic time | Approximately 3 hours including setup |
| Assays | Plasma C-peptide and glucose at every timepoint |
Interpretation: AUC C-Peptide
The classic readouts from a 2-hour MMTT are:
- Peak C-peptide: highest single value during the test (usually at 60 or 90 minutes)
- AUC C-peptide (2-hour): integrated area under the C-peptide curve — total pancreatic output
- Glucose response curve: glucose at each timepoint; reflects the combined glucose tolerance
| 2-hour AUC C-peptide (nmol/L × min) | Interpretation |
|---|---|
| Less than 0.2 | Severe beta-cell failure (typical advanced T1D) |
| 0.2 to 0.6 | Limited reserve (long-standing T1D or honeymoon) |
| 0.6 to 1.5 | Partial reserve (LADA, early T1D, advanced T2D) |
| 1.5 to 3.0 | Normal range (T2D or non-diabetic) |
| Greater than 3.0 | Hyperresponsive (insulin resistance with intact beta cells) |
For TrialNet clinical trial protocols, peak C-peptide greater than 0.2 nmol/L is often used as a binary indicator of preserved beta-cell function.
When the MMTT Is Used
| Clinical or research setting | Why MMTT helps |
|---|---|
| T1D prevention or intervention clinical trials | Gold standard endpoint for beta-cell preservation |
| TrialNet at-risk relative monitoring | Tracks beta-cell function over time |
| Pre-pancreas or islet transplant evaluation | Documents residual function before procedure |
| Post-transplant graft function | Monitors transplant beta-cell output |
| Diabetes classification when C-peptide ambiguous | Dynamic response clarifies T1D vs T2D vs LADA |
| Pre-bariatric surgery evaluation | Beta-cell baseline before metabolic surgery |
| Post-pancreatectomy assessment | Confirms beta-cell deficit |
| Suspected MODY or rare diabetes | Phenotypes unusual beta-cell patterns |
MMTT vs OGTT vs Glucagon Stimulation Test
| Feature | MMTT | OGTT | Glucagon Test |
|---|---|---|---|
| Test stimulus | Liquid mixed meal (50 g CHO + protein + fat) | 75 g pure glucose drink | 1 mg IV glucagon |
| Duration | 2 to 4 hours | 2 hours (occasionally 3) | 6 minutes |
| Tolerability | Good (real food taste) | Poor (nausea common) | Poor (nausea, vomiting) |
| Sampling | 5 to 7 timepoints | 2 to 3 timepoints | 2 timepoints |
| Primary readout | AUC C-peptide | 2h glucose | 6-min C-peptide |
| Best for diabetes diagnosis | No | Yes | No |
| Best for beta-cell function | Yes (gold standard) | Possible (with insulin curve) | Quick alternative |
| Physiologic | Most physiologic | Non-physiologic glucose load | Pharmacologic stimulus |
MMTT in T1D Clinical Trials
The MMTT became standard in T1D research after the DCCT and subsequent trials demonstrated that residual C-peptide secretion — even small amounts — correlates with better glycemic control and fewer complications. Modern T1D prevention and disease-modifying therapy trials (teplizumab, low-dose ATG, alefacept, GLP-1 agonists in T1D) all use MMTT-derived AUC C-peptide as the primary efficacy endpoint.
The protocol standardization — fixed dose (6 mL/kg up to 360 mL Boost), fixed sampling schedule, fixed AUC calculation — makes results comparable across trials and centers. TrialNet maintains the canonical protocol and reference data for the field.
MMTT Sample Output (Hypothetical)
| Time (min) | Glucose (mg/dL) | C-peptide (ng/mL) | Notes |
|---|---|---|---|
| 0 (baseline) | 110 | 1.2 | Fasting |
| 15 | 145 | 2.8 | Early response |
| 30 | 170 | 3.5 | Approaching peak |
| 60 | 180 | 4.1 | Peak C-peptide |
| 90 | 165 | 3.6 | Plateau |
| 120 | 140 | 2.9 | Declining |
This pattern — peak around 60 minutes, gradual decline by 120 minutes — is typical of preserved beta-cell function. Patients with severe T1D show essentially flat C-peptide curves throughout the test.
Limitations and Pitfalls
- Logistically demanding: 3-hour clinic visit, multiple blood draws, sometimes IV line placement.
- Not for diabetes diagnosis: the OGTT and A1C are the diagnostic standards.
- Beta-cell suppression by acute hyperglycemia: if baseline glucose is very high, the test may underestimate true beta-cell capacity (glucotoxicity).
- Insulin therapy: patients on insulin should hold short-acting insulin per protocol; basal insulin handling varies by study.
- Test meal palatability and tolerance: some patients experience nausea or early satiety with the full 360 mL bolus.
- Lactose intolerance: Boost and Sustacal contain milk-based protein; alternatives may be needed.
- Assay variability: central laboratory C-peptide assays are standardized but local labs may vary.
- Day-to-day variability: repeat MMTT results can differ by 10 to 20 percent in stable subjects.
What to Expect Practically
- You will be scheduled for a morning clinic visit, fasted 8 to 10 hours overnight.
- An IV line is placed in your arm for easier sequential blood draws.
- Baseline blood is drawn for C-peptide and glucose.
- You drink the Boost or Sustacal within 5 to 10 minutes.
- Blood is drawn at 15, 30, 60, 90, and 120 minutes.
- You may feel mildly full or slightly nauseated during the test but it is generally well tolerated.
- Results take 5 to 10 days; your endocrinologist will review them with you.
Related Reading
See our broader guides on detection of prediabetes, C-peptide testing, postprandial glucose, GAD-65 antibody testing, and A1C levels.
The Bottom Line
The mixed meal tolerance test is the gold standard for assessing pancreatic beta-cell function with sequential C-peptide measurements after a standardized liquid meal. It is the primary endpoint of T1D prevention and intervention trials, an essential tool in pre-transplant and pre-bariatric evaluations, and a useful adjunct when fasting C-peptide leaves diabetes classification ambiguous. The protocol — Boost or Sustacal at 6 mL/kg up to 360 mL, with sampling at 0, 15, 30, 60, 90, and 120 minutes — is well standardized through TrialNet. AUC C-peptide over 2 hours is the headline number; higher AUC means more preserved beta-cell function. The MMTT is not used for diabetes diagnosis or screening. Talk to your endocrinologist about whether the test fits your specific clinical or research situation.
