Oral orforglipron vs dapagliflozin on metformin: ACHIEVE-2 HbA1c results posted

The study

ACHIEVE-2 is a phase 3, randomised, open-label trial. It compared once-daily oral orforglipron with dapagliflozin in adults with type 2 diabetes and inadequate glycaemic control on metformin. The sponsor is Eli Lilly and Company. The registry record (NCT06192108) lists an enrolment of 962 participants. The study is described as lasting approximately 46 weeks.

The stated main purpose is to assess the safety and efficacy of orforglipron compared with dapagliflozin for blood sugar control. The primary outcome is the change in HbA1c from baseline to week 40. It is labelled a noninferiority analysis.

These data come from the results section of a ClinicalTrials.gov record. They are not from a journal article.

What they found

The record reports least squares mean changes in HbA1c (percentage units) from baseline to week 40. All four values are falls in HbA1c:

• Orforglipron 3 mg: -1.32 (spread 0.074)
• Orforglipron 12 mg: -1.74 (spread 0.062)
• Orforglipron 36 mg: -1.73 (spread 0.079)
• Dapagliflozin 10 mg: -0.76 (spread 0.076)

The falls were numerically larger in all three orforglipron groups than in the dapagliflozin group. The record does not say what the "spread" figure is, for example standard error. The text available here does not give the noninferiority margin, confidence intervals or p values. It does not report whether noninferiority or superiority was formally shown.

The available text does not report safety results, side effects, weight change, treatment discontinuation or secondary outcomes.

How much weight to give it

The design is a randomised phase 3 trial with an active comparator. That is a useful design for comparing HbA1c effects. Several limits remain.

• The source is a registry results entry from the sponsor. It is not a full peer-reviewed paper, and details may differ from a later publication.
• The trial is open-label.
• HbA1c is a surrogate outcome. The record gives no data on cardiovascular, kidney or other hard outcomes.
• Without the formal test results and safety data, the balance of benefit and harm cannot be judged.
• The record does not say how many participants completed the trial or how missing data were handled.

For your clinic

The least squares mean HbA1c falls at week 40 were numerically larger with all three orforglipron doses (-1.32, -1.74, -1.73) than with dapagliflozin 10 mg (-0.76). The record does not report whether noninferiority or superiority was formally shown. This is a general signal only. It is not a basis for choosing between drugs for an individual patient.

Before this changes practice, read the full publication or regulatory review. Look for the noninferiority and superiority results, safety and tolerability, and the characteristics of the participants. Clinicians should also consider the other effects and the cost of each option, which this record does not cover. For now, current guidelines and the availability of drugs in India remain the practical guide.