6  Application to Drugs

As we saw in Chapter 2, physiological GLP-1 and GIP have a major weakness: they are destroyed by DPP-4 within minutes. The history of incretin-based drugs is the history of how to overcome this weakness.

flowchart TB
  Problem["Problem: GLP-1 half-life is 1–2 min"]
  Problem --> A1["Approach 1<br/>Block the destroying enzyme"]
  Problem --> A2["Approach 2<br/>Make analogues that are not destroyed"]
  A1 --> DPP4i["DPP-4 inhibitors<br/>(oral, endogenous incretins↑)"]
  A2 --> RA["GLP-1 receptor agonists<br/>(injection, pharmacological doses)"]
  RA --> Dual["GIP/GLP-1 receptor agonist<br/>(tirzepatide)"]
  Dual --> Next["Next generation<br/>triple agonists, oral small molecules,<br/>amylin combinations, GIPR antagonism"]

6.1 DPP-4 inhibitors: blocking the enzyme

Sitagliptin, vildagliptin, alogliptin, linagliptin, etc.

  • Oral drugs. They inhibit DPP-4 and raise the concentrations of endogenous active GLP-1 and GIP by about 2–3-fold
  • The rise is strictly within the physiological range → the glucose-lowering effect is mild, and body weight hardly changes
  • Rarely cause hypoglycemia (because of the glucose-dependent action)
  • Very widely used for the treatment of type 2 diabetes in Japan

6.2 GLP-1 receptor agonists: analogues that are not destroyed

6.2.1 Design tricks

Drug Trick for DPP-4 resistance Trick for prolonged action Half-life Administration
Native GLP-1 None None 1–2 min (not a drug)
Exenatide Exendin-4 from Gila monster saliva. Position 2 is Gly None about 2.4 h Twice daily, subcutaneous
Liraglutide (somewhat resistant to DPP-4) C16 fatty acid → albumin binding about 13 h Once daily, subcutaneous
Dulaglutide Position 2 replaced with Gly IgG4 Fc fusion → larger molecular weight, less renal excretion about 5 days Once weekly, subcutaneous
Semaglutide Position 2 replaced with Aib (a non-natural amino acid) C18 fatty diacid → strong albumin binding about 1 week Once weekly subcutaneous / once daily oral
Native GLP-1(7-37):  H-[A]-E-G-T-F-T-S-D-V-S-S-Y-L-E-G-Q-A-A-[K]-E-F-I-A-W-L-V-[K]-G-R-G
Semaglutide:         H-[Aib]-E-G-T-F-T-S-D-V-S-S-Y-L-E-G-Q-A-A-[K*]-E-F-I-A-W-L-V-[R]-G-R-G

  Position 8   A → Aib : DPP-4 can no longer recognize it (DPP-4 resistance)
  Position 26  K → K*  : C18 fatty diacid attached via a linker (albumin binding)
  Position 34  K → R   : the other Lys replaced with Arg so that the fatty acid attaches only at position 26
TipThe invention of albumin binding

When a fatty-acid chain is attached, the drug binds reversibly to albumin in the blood.

  • While bound to albumin, it is less easily filtered by the kidney and less easily degraded by enzymes such as DPP-4
  • Albumin itself has a long half-life of about 19 days
  • Through repeated binding and dissociation, the free form acts on the receptor little by little

This strategy of “using albumin as a carrier” made it possible to go from once daily (liraglutide) → once weekly (semaglutide).

6.2.2 Oral semaglutide

Peptides are normally degraded in the stomach and are barely absorbed from the intestine. Oral semaglutide (Rybelsus) is formulated together with the absorption enhancer SNAC (salcaprozate sodium).

  • SNAC raises the local pH around the tablet, preventing degradation by gastric enzymes, and temporarily promotes absorption through the gastric mucosa
  • Even so, the absorption rate (bioavailability) is low, about 1% → a much larger amount than the weekly injected dose must be taken every day
  • Because absorption varies greatly with food and water, there are dosing conditions such as take on an empty stomach with a small amount of water, and do not eat or drink for 30 minutes after taking it

6.3 GIP/GLP-1 receptor agonist: tirzepatide

6.3.1 “GIP doesn’t work” — or so it was thought

As we saw in Chapter 1, because the response to GIP is reduced in type 2 diabetes, it was long thought that GIP could not be made into a drug. However, findings such as

  • the response to GIP partly recovers when blood glucose is lowered
  • synergistic effects can be expected when it acts together with GLP-1
  • the possibility of favorable extrapancreatic actions, such as reduced nausea in the brain

led to the development of a drug that stimulates both GIPR and GLP-1R with a single molecule. That is tirzepatide (Mounjaro/Zepbound).

6.3.2 Molecular features

  • A 39-amino-acid peptide based on the GIP sequence, also incorporating elements of the GLP-1 and exenatide sequences
  • Aib at positions 2 and 13 → DPP-4 resistance
  • C20 fatty diacid attached → half-life about 5 days via albumin binding, once-weekly dosing
  • “Imbalanced” agonist:
    • At GIPR, affinity comparable to native GIP
    • At GLP-1R, affinity about 5-fold weaker than native GLP-1
    • → a GIPR-leaning agonist
  • At GLP-1R, a cAMP-biased agonist (weak β-arrestin recruitment, less internalization; Chapter 3)

6.3.3 Clinical effects

Trial Population Comparison Main result
STEP 1 Obesity (no diabetes) Semaglutide 2.4 mg vs placebo Body weight −14.9% vs −2.4% at 68 weeks
SURMOUNT-1 Obesity (no diabetes) Tirzepatide 15 mg vs placebo Body weight −20.9% vs −3.1% at 72 weeks
SURPASS-2 Type 2 diabetes Tirzepatide vs semaglutide 1 mg Tirzepatide superior in both HbA1c and body weight
SURMOUNT-5 Obesity (no diabetes) Tirzepatide vs semaglutide 2.4 mg Body weight −20.2% vs −13.7% at 72 weeks

(Values are representative figures from each trial’s primary analysis; they vary with dose and analysis method.)

6.4 The GIP paradox

Now let’s return to the question raised earlier.

  • GIPR agonism: tirzepatide (GIPR + GLP-1R agonism) reduces body weight more than a GLP-1RA alone
  • GIPR antagonism: GIPR-deficient mice resist obesity. Moreover, a drug that conjugates a GIPR-antagonist antibody with GLP-1 peptides (maridebart cafraglutide; MariTide) has also shown substantial weight loss in clinical trials

Weight falls whether GIPR is activated or blocked —— this is the “GIP paradox”. Leading hypotheses at present include:

  1. Sustained agonism → receptor desensitization/down-regulation → effectively the same as antagonism
  2. Agonism and antagonism act in different tissues (the agonist reduces nausea in the brain and improves lipid handling in adipose tissue; the antagonist acts via other pathways)
  3. Both are effects in combination with GLP-1R agonism, and must be considered separately from the action of GIP alone
NoteA topic still under investigation

This paradox has not yet been resolved and is one of the central questions of incretin research in the 2020s. When covering it in a lecture, it is appropriate to present it as “a comparison of hypotheses” rather than “an established mechanism”.

6.5 Next-generation drugs

Type Example Key point
GIP/GLP-1/glucagon triple agonist Retatrutide Aims for energy expenditure↑, liver fat↓ through glucagon-receptor agonism. About −24% at 48 weeks in phase 2
Oral small-molecule GLP-1RA Orforglipron A small molecule, not a peptide. Can be taken without food/water restrictions and is easy to manufacture
Amylin + GLP-1 CagriSema (cagrilintide + semaglutide) Combines an analogue of amylin, another satiety hormone released from pancreatic β-cells
GIPR antagonism + GLP-1R agonism Maridebart cafraglutide Antibody-based. Aims for dosing about once a month
CautionApproval status changes quickly

The development and approval status of next-generation drugs changes from year to year. Check the latest information when using this in a lecture.

6.6 Side effects and precautions

Side effect / precaution Mechanism / background
Nausea, vomiting, diarrhea, constipation The most common. Action on the brainstem (area postrema) and delayed gastric emptying. Reduced by starting at a low dose and escalating gradually
Gallstones, cholecystitis Rapid weight loss, reduced gallbladder motility
Acute pancreatitis An association has been debated. Use with caution in patients with a history
Hypoglycemia Rare on its own. Caution when combined with SU drugs or insulin (Chapter 3)
Thyroid C-cell tumors Reported in rodents. Significance in humans is unknown, but not used in people with a personal or family history of medullary thyroid carcinoma (e.g., US package insert)
Worsening of diabetic retinopathy Transient worsening associated with rapid lowering of blood glucose (noted in SUSTAIN-6)
Loss of lean mass (muscle) Part of the weight loss is lean mass. Adequate protein intake and exercise are recommended
Weight regain after discontinuation In the STEP 1 extension, about 2/3 of the lost weight returned within 1 year after stopping
Perioperative aspiration Due to delayed gastric emptying. Management before anesthesia and endoscopy is being discussed

6.7 Summary

  • DPP-4 inhibitors: block the enzyme to increase endogenous incretins within the physiological range. Mild, weight-neutral
  • GLP-1RAs: prolonged action through DPP-4 resistance + albumin binding/Fc fusion. Continuously stimulate GLP-1R throughout the body at pharmacological doses, with strong glucose lowering and weight loss
  • Tirzepatide: a GIPR-leaning, imbalanced dual agonist + cAMP bias at GLP-1R. Weight loss exceeding that of GLP-1RAs alone
  • GIP paradox: body weight falls with both agonism and antagonism. The mechanism is still under investigation