SUSTAIN OPTIMIZE: Semaglutide Trial

SUSTAIN OPTIMIZE found that adding once-weekly semaglutide to a planned reduction in insulin glargine lowered HbA1c and body weight more than continued insulin glargine titration over 40 weeks. The phase 3b trial randomized 573 adults with type 2 diabetes and overweight. Its comparison tested two treatment strategies, not semaglutide against an unchanged insulin dose^[1].

The estimated difference in HbA1c change was -0.74 percentage points, with a 95% confidence interval from -0.90 to -0.59. Body weight differed by -8.5 kg. Gastrointestinal events were more common in the semaglutide strategy, while severe hypoglycaemia occurred less often. These group results do not supply a self-directed method for changing insulin^[2].

How SUSTAIN OPTIMIZE tested semaglutide

The open-label trial ran at 113 outpatient sites in nine countries. Adults were eligible if they had type 2 diabetes, a body mass index of at least 25 kg/m2 and HbA1c from 7% to 10%. They were already using no more than 40 units of basal insulin per day, plus stable metformin with or without an SGLT2 inhibitor. The mean age was 61.1 years and mean HbA1c was 7.9%^[2].

Participants were assigned equally to one of two strategies. One added subcutaneous semaglutide while insulin glargine was reduced according to the trial protocol. The other continued insulin glargine with glucose-guided titration and no semaglutide. Different titration procedures made masking impractical, according to the authors. The primary endpoint was HbA1c change at week 40, first tested against a prespecified non-inferiority margin of 0.3 percentage points^[2].

The dose schedule belongs to a monitored research protocol. It should not be copied into personal care. Changing insulin while starting another glucose-lowering medicine can alter hypoglycaemia risk, and a qualified clinician needs to assess glucose records, other medicines and the reason for treatment.

HbA1c and weight favored the combined strategy

Mean HbA1c fell by 1.4 percentage points in the semaglutide plus reduced-insulin group and by 0.6 points in the titrated-insulin group. Mean values at week 40 were 6.5% and 7.3%, respectively. The trial met its non-inferiority objective and then its hierarchical superiority test for HbA1c. The 95% confidence interval around the -0.74-point treatment difference did not cross zero^[2].

Mean body weight changed by -8.0 kg with the semaglutide strategy and +0.5 kg with insulin titration. The estimated difference was -8.5 kg, with a 95% confidence interval from -9.5 to -7.4. At week 40, 73.2% of participants in the semaglutide group and 9.1% in the comparator group had lost at least 5% of baseline weight. That threshold analysis was exploratory rather than part of the confirmatory hierarchy^[2].

The semaglutide evidence overview places this result within a larger research program. SUSTAIN OPTIMIZE answers a narrower question: what happened in adults already taking limited-dose basal insulin when a protocol shifted emphasis toward semaglutide instead of further insulin titration. It does not compare the strategy with every other intensification option.

Insulin use fell, with an important data caveat

Mean relative daily insulin use changed by -17.8% in the semaglutide group and +104.1% with titrated insulin. The estimated between-group difference was -121.9 percentage points. In an exploratory absolute analysis, the groups differed by -25.3 units per day. These figures describe the protocol's contrasting directions and should not be read as an expected reduction for an individual patient^[2].

The paper reports a recording problem for this endpoint. Some participants reached zero insulin, but those zero values were entered as missing. Among 288 participants assigned to the semaglutide strategy, 106 had missing insulin-dose data after this issue and were handled through the trial's missing-data method. The authors ran a sensitivity analysis, yet the amount and cause of missingness still matter when interpreting the exact insulin estimate^[2].

Safety results moved in different directions

Adverse events were reported by 60.1% of participants assigned to semaglutide plus reduced insulin and 53.5% assigned to titrated insulin. Gastrointestinal disorders affected 106 versus 19 participants. Ten participants in the semaglutide group and one in the comparator group discontinued treatment because of adverse events. Serious adverse events occurred in 5.9% and 5.3%, respectively^[2].

Severe hypoglycaemia was reported in 4 participants, or 1.4%, in the semaglutide group and 11 participants, or 3.9%, with titrated insulin. The combined rate for clinically significant or severe episodes produced a rate ratio of 0.45, with a 95% confidence interval from 0.23 to 0.87. The authors caution that the trial was not powered specifically for a between-group hypoglycaemia comparison^[2].

The paper states that no new safety concern was identified, but that wording does not mean no harms occurred. Gastrointestinal events and treatment discontinuations remain part of the result. Our semaglutide safety section covers established adverse-effect context beyond this single trial. Product-specific prescribing information remains the appropriate reference for clinical use.

Open-label design and sponsorship limit interpretation

The trial's open-label design may have influenced treatment-satisfaction reporting because participants knew which strategy they received. The authors also limit generalization to people resembling the enrolled group: basal insulin at 40 units per day or less, HbA1c between 7% and 10%, and body mass index at least 25 kg/m2. Most participants were White and 70% were enrolled in Europe^[2].

Novo Nordisk funded the study, employed three authors and supplied the statistical analysis group. Other authors reported consulting, speaking or research relationships with Novo Nordisk and other pharmaceutical companies. Sponsorship does not erase a randomized result, but it belongs in the evidence assessment alongside the protocol, missing data and declared conflicts^[2].

What we don't yet know

Forty weeks cannot establish whether the differences persist over years or what happens after semaglutide is stopped. The trial does not answer the same question for people using more than 40 units of basal insulin, those with HbA1c above 10%, or patients excluded because of particular clinical risks. It also does not test unsupervised insulin reduction.

Continuous glucose monitoring was not used for the primary safety comparison, so asymptomatic and overnight glucose patterns were not characterized as fully as they might be in a future trial. A blinded design was not feasible for the two titration strategies, and treatment satisfaction therefore carries more uncertainty than the laboratory HbA1c result.

SUSTAIN OPTIMIZE provides randomized evidence that the combined protocol improved glycaemia and weight relative to further basal-insulin titration in its selected population. The result supports clinical discussion, not a universal treatment switch. Anyone considering a change to insulin or semaglutide should consult the clinician managing their diabetes.

Medical disclaimer: This article is for educational and journalistic purposes only and does not constitute medical advice. Peptides discussed may be classified as prescription medicines or research chemicals depending on your jurisdiction. Always consult a qualified healthcare professional before using any peptide product. PeptideMethods.com does not sell, distribute, or facilitate the sale of any peptide product.