Tirzepatide

At week 72, average body weight had fallen 12.8% with tirzepatide 10 mg and 14.7% with 15 mg, compared with 3.2% on placebo, in the journal's headline analysis (differences from placebo of 9.6 and 11.6 percentage points). The registry's posted figures were 13.4%, 15.7% and 3.3%. In the registry, 81.55% (10 mg) and 86.41% (15 mg) of people lost at least 5% of their weight, compared with 30.55% on placebo; the journal reports 79% to 83% versus 32%. Blood sugar fell by 2.14 percentage points (10 mg) and 2.22 (15 mg) in HbA1c, versus 0.16 on placebo, and about 90% of people on tirzepatide reached an HbA1c under 7% versus 29% on placebo.

• The two sources report slightly different headline weight figures, and the registry does not say which analysis method it used.
• Only the PubMed abstract of the journal report was reviewed, not the full paper, so details such as subgroup results are not covered.
• The trial lasted 72 weeks and measured weight and blood-sugar markers; it was not designed to show effects on heart attacks, strokes or death.
• Participants were adults with type 2 diabetes (HbA1c 7% to 10%) and BMI of 27 or higher who were not taking GLP-1 medicines or DPP-4 inhibitors, and 60% were Hispanic or Latino, so results may not apply to other groups.
• Not everyone finished the study (34 people on placebo, 16 on 10 mg and 29 on 15 mg did not), so some outcomes depend on how missing data were handled.

• Average body-weight change at week 72: Lancet abstract (treatment-regimen analysis) reports -12.8% (10 mg), -14.7% (15 mg), -3.2% (placebo); ClinicalTrials.gov posted results reports -13.4% (10 mg), -15.7% (15 mg), -3.3% (placebo). The registry figures are about 0.1 to 1.0 percentage point larger in the tirzepatide groups. The reason for the difference is not established from the sources reviewed.

At week 72 the journal abstract reports average weight changes of -15.0% (5 mg), -19.5% (10 mg) and -20.9% (15 mg) versus -3.1% with placebo (P<0.001 for all), under an estimand that counts everyone regardless of whether they stayed on treatment. The registry, which excludes data collected after people stopped the study drug, reports -16.0%, -21.4% and -22.5% versus -2.4%. Weight loss of at least 5% was reached by 85%, 89% and 91% of people on the three doses versus 35% on placebo (journal), and loss of 20% or more by 50% (10 mg) and 57% (15 mg) versus 3% (journal). In the prediabetes extension (to week 176), type 2 diabetes was diagnosed in 1.2% of people on tirzepatide (pooled doses) versus 12.6% on placebo.

• The registry and the journal report different weight-loss averages because they handle people who stopped treatment differently (excluding their later data versus counting everyone).
• More people left the placebo group early: 134 of 643 did not complete the 72-week period, compared with 55 of 630 on 15 mg.
• Participants did not have diabetes; two thirds were women (1,714 of 2,539), and results may not apply to other groups.
• The extension to week 176 and the diabetes-onset results cover only the 1,032 people with prediabetes at the start (901 continued), are secondary analyses, and were not designed to test the drug as a diabetes-prevention treatment.
• Percentages reaching 5% to 20% weight loss are model-based estimates, and the registry's estimates differ from the journal's.
• The trial was funded by the drug's developer, and only the journal abstract was reviewed, not the full paper.

• Average change in body weight at week 72 with tirzepatide 15 mg: ClinicalTrials.gov results (data after stopping the drug excluded) reports -22.5%; NEJM abstract (treatment-regimen estimand) reports -20.9%. 1.6 percentage points; the placebo averages also differ (-2.4% vs -3.1%). The registry states that its analysis excludes data after people prematurely stopped the study drug, while the abstract states that its estimand assessed effects regardless of treatment discontinuation.

At week 40, HbA1c had fallen by 2.01, 2.24 and 2.30 percentage points with tirzepatide 5, 10 and 15 mg, compared with 1.86 with semaglutide 1 mg (journal values). The differences from semaglutide were 0.15, 0.39 and 0.45 points, and the paper reports all three doses were noninferior and superior to semaglutide. Body weight fell more with tirzepatide at every dose (differences from semaglutide of 1.9, 3.6 and 5.5 kg; registry averages of 7.8, 10.3 and 12.4 kg for tirzepatide versus 6.2 kg for semaglutide). Stomach-related side effects were the most common events.

• An open-label trial: participants and doctors knew which medicine was used, which can affect reported symptoms and side effects.
• The comparison is with one dose of semaglutide (1 mg), so it does not describe how tirzepatide compares with other semaglutide doses or other medicines.
• The journal and the registry give different difference-from-semaglutide figures for HbA1c and weight, and the sources do not explain why.
• Only the PubMed abstract of the journal report was reviewed, not the full paper.
• The trial lasted 40 weeks and measured blood-sugar and weight markers; it was not designed to show effects on heart attacks, strokes or death.
• Participants were adults with HbA1c 7.0% to 10.5% on stable metformin, about 70% Hispanic or Latino, so results may not apply to other groups.

• HbA1c difference between each tirzepatide dose and semaglutide at week 40: NEJM abstract reports -0.15 (5 mg), -0.39 (10 mg), -0.45 (15 mg) percentage points; ClinicalTrials.gov posted results reports -0.23 (5 mg), -0.51 (10 mg), -0.60 (15 mg) percentage points. The registry differences are 0.08 to 0.15 percentage points larger for every dose. The reason for the difference is not established from the sources reviewed.
• Weight difference between each tirzepatide dose and semaglutide at week 40: NEJM abstract reports -1.9 kg (5 mg), -3.6 kg (10 mg), -5.5 kg (15 mg); ClinicalTrials.gov posted results reports -1.7 kg (5 mg), -4.1 kg (10 mg), -6.2 kg (15 mg). The differences vary by up to 0.7 kg between the sources. The reason for the difference is not established from the sources reviewed.