The stakes of this comparison are enormous. Type 2 diabetes now affects an estimated 589 million adults worldwide as of 2024, a figure projected to climb to 853 million by 2050. Because the disease drives complications through both chronic hyperglycemia and the twin burdens of obesity and cardiovascular disease, contemporary treatment guidelines no longer ask simply whether a drug lowers glucose. They demand sustained reductions in hemoglobin A1c, meaningful weight reduction, cardiovascular safety, and tolerability that patients can live with for decades. Both tirzepatide and dulaglutide are once-weekly subcutaneous injections, but they work differently at the molecular level, and until now no trial had ever compared all clinically relevant doses of the two drugs simultaneously.

Dulaglutide is a selective glucagon-like peptide-1 receptor agonist, a class of drugs that mimics an intestinal hormone to boost glucose-dependent insulin secretion, suppress glucagon, slow gastric emptying, and reduce appetite. It carries hard cardiovascular outcome evidence: in the landmark REWIND trial, dulaglutide reduced major adverse cardiovascular events compared with placebo in people with type 2 diabetes. Tirzepatide, by contrast, is a first-in-class dual agonist that engages both the GIP and GLP-1 receptors, amplifying the incretin effect through two complementary hormonal pathways. Head-to-head trials had already hinted that tirzepatide’s dual mechanism translated into greater reductions in HbA1c and body weight, but individual studies compared only selected dose pairs, leaving clinicians without a comprehensive map of which specific dose delivers which specific benefit.

The research team, following Cochrane Handbook methodology and PRISMA reporting guidelines with a prospectively registered protocol, searched four major databases through the end of 2025 and identified four eligible randomized trials. Three of these were judged to carry low risk of bias across all assessed domains; the fourth, an open-label switching study, raised some concerns because participants and clinicians knew which treatment was being given. The largest trial by far was the SURPASS-CVOT cardiovascular outcomes study, which enrolled 13,165 patients with established atherosclerotic cardiovascular disease and demonstrated that tirzepatide titrated up to 15 milligrams was noninferior to dulaglutide 1.5 milligrams for the composite of cardiovascular death, myocardial infarction, or stroke over a median follow-up of roughly four years.

Using a frequentist random-effects network model, the analysts computed treatment effects and SUCRA rankings, a statistical measure of each dose’s cumulative probability of being among the most effective options. The results formed a clean dose-response gradient. For glycemic control, tirzepatide 15 milligrams ranked highest at both 12 and 24 weeks, achieving SUCRA values of 0.95 and 0.99 respectively, while dulaglutide 1.5 milligrams sat near the bottom of the efficacy rankings. At 24 weeks, tirzepatide 15 milligrams reduced HbA1c significantly more than tirzepatide 5 milligrams, and both dulaglutide 0.75 and 1.5 milligrams were associated with meaningfully smaller reductions. The differences of 0.2 to 0.4 percentage points between higher and lower tirzepatide doses are clinically meaningful, because shifts of that magnitude can move patients across standard treatment targets and delay the need for therapy intensification.

The weight findings followed the same pattern with even sharper separation. By week 16, tirzepatide 15 milligrams outperformed tirzepatide 5 milligrams by an additional 2.68 kilograms of weight loss, and the contrast against the weakest doses was dramatic: dulaglutide 0.75 milligrams and tirzepatide 1 milligram trailed the top dose by roughly 5 to 8 kilograms. Tirzepatide 15 milligrams earned a SUCRA of 0.95 for weight reduction, with the 10 milligram dose close behind, while tirzepatide 1 milligram ranked nearly last. Notably, the early weight outcomes showed substantial statistical heterogeneity, which the authors attribute to the sensitivity of short-term weight trajectories to titration schedules, baseline body mass index, diabetes duration, and transient gastrointestinal symptoms that can temporarily suppress food intake during dose escalation.

Safety told the mirror-image story. Tirzepatide 1 milligram carried the lowest risk of any treatment-emergent adverse event and of nausea, while dulaglutide 0.75 milligrams showed the lowest risks of diarrhea and vomiting. At the other end of the spectrum, both tirzepatide 15 milligrams and dulaglutide 1.5 milligrams were associated with significantly higher nausea risk compared with tirzepatide 5 milligrams. This gradient is biologically coherent: higher incretin receptor engagement intensifies slowed gastric emptying and central satiety signaling, which are precisely the mechanisms that drive both the metabolic benefits and the gastrointestinal complaints. Encouragingly, no significant differences emerged between any dose pairs for all-cause death, severe hypoglycemia, serious adverse events, pancreatitis, treatment discontinuation, or the composite cardiovascular endpoint, although the authors caution that wide confidence intervals for these rare outcomes mean the absence of statistical differences cannot be read as proof of equivalent safety.

The team stress-tested their findings with a battery of sensitivity analyses, including one that removed the dominant SURPASS-CVOT trial entirely to check whether its enormous statistical weight was distorting the network. The core conclusions held. Some outcomes, notably HbA1c at 12 weeks, early body weight change, and nausea and vomiting, proved sensitive to the inclusion of the Japanese SURPASS J-mono trial, which compared tirzepatide against the 0.75 milligram dulaglutide dose approved in Japan rather than the 1.5 milligram standard used elsewhere. Inconsistency testing using both design-by-treatment interaction models and node-splitting found no meaningful disagreement between direct and indirect evidence for most outcomes, lending credibility to the network’s internal structure.

The clinical implications are refreshingly practical. For patients whose primary goals are maximal HbA1c reduction and weight loss, the data support escalating to tirzepatide 10 or 15 milligrams when tolerability permits, with an expected stepwise advantage over lower doses and over any dulaglutide regimen. For patients whose priority is gastrointestinal tolerability, starting at lower doses such as dulaglutide 0.75 milligrams or tirzepatide 1 milligram remains a reasonable strategy, accepting more modest metabolic gains and possible later intensification. The findings also reinforce the results of the SURPASS-SWITCH trial, which showed that patients inadequately controlled on submaximal dulaglutide achieved greater HbA1c and weight improvements by switching to tirzepatide than by pushing the dulaglutide dose higher. An important interpretive caveat applies to the early time points: many participants assigned to the highest tirzepatide doses were still in protocol-mandated dose escalation at weeks 12 and 16, meaning the full maintenance-dose effects may be even larger than these figures suggest.

The authors are candid about the limitations. Only four trials informed the network, several dose nodes rested on sparse comparisons, and long-term durability, late adverse events, and sustained discontinuation could not be assessed because later time points were not consistently reported across trials. Some data were digitized from published figures, introducing the possibility of minor extraction error despite independent double-checking. Still, within those constraints, the analysis delivers something no single trial could: a unified, dose-level ranking of the two most widely used once-weekly injectable therapies for type 2 diabetes. The message for the millions of patients and their clinicians navigating this decision is one of individualized titration, balancing the escalating metabolic rewards of higher tirzepatide doses against the dose-related gastrointestinal costs, with the reassurance that cardiovascular safety appears preserved across the spectrum.

Subject of Research: Comparative efficacy and safety of tirzepatide versus dulaglutide doses in type 2 diabetes

Article Title: Efficacy and Safety of Tirzepatide Versus Dulaglutide in Type 2 Diabetes With or Without Established Atherosclerotic Cardiovascular Disease: A Network Meta‐Analysis of Randomized Clinical Trials

Article References: Hageen, A. W., Gadelmawla, A. F., Saleh, A. O., Bahnasy, S., Eladawi, S., Abdelaziz, M., Iyad, K., Kandil, A. H., Zinhom, K., Mohamed, M. R., Abdulhay, H., Turkman, M., Abdelazeem, B., & Fonarow, G. C. (2026). Efficacy and Safety of Tirzepatide Versus Dulaglutide in Type 2 Diabetes With or Without Established Atherosclerotic Cardiovascular Disease: A Network Meta‐Analysis of Randomized Clinical Trials. Endocrinology, Diabetes & Metabolism, 9(5), Article e70313. https://doi.org/10.1002/edm2.70313

Image Credits: AI Generated

DOI: 10.1002/edm2.70313

Keywords: tirzepatide, dulaglutide, type 2 diabetes, network meta-analysis, HbA1c, weight loss, GLP-1 receptor agonist, GIP, cardiovascular outcomes, gastrointestinal adverse events, dose-response, incretin therapy