What SURPASS-3 Actually Changes in Clinical Practice

At a glance

Detail	Value
Trial	SURPASS-3 (NCT03882970)
N	1,444 randomized
Intervention	Tirzepatide 5 mg, 10 mg, or 15 mg SC once weekly
Comparator	Insulin degludec, titrated to fasting glucose <90 mg/dL
Duration	52 weeks
Primary endpoint	Change in A1C from baseline
Key result	All tirzepatide doses were superior to degludec for A1C reduction (up to −2.37% vs −1.34%) and produced weight loss (up to −12.9 kg) vs weight gain (+2.3 kg)
Publication	Ludvik et al., Lancet 2021

Why This Trial Exists

Before SURPASS-3, clinicians treating type 2 diabetes inadequately controlled on metformin faced a familiar fork. Add a GLP-1 receptor agonist, or start basal insulin? Guidelines from the American Diabetes Association (ADA) had long placed both options on the same decision tier. The problem was that no large trial had pitted a dual GIP/GLP-1 receptor agonist directly against optimally titrated basal insulin.

SURPASS-3 filled that gap. Published in The Lancet in 2021, the trial randomized 1,444 adults across 122 sites in 13 countries. Each participant had a baseline A1C between 7.0% and 10.5% on metformin (with or without an SGLT2 inhibitor). The comparator, insulin degludec, was not capped at a token dose. Investigators titrated it weekly to a fasting glucose target below 90 mg/dL, giving insulin the best possible shot.

What the Methodology Actually Looked Like

The trial used an open-label design. Blinding was not feasible because tirzepatide is a fixed weekly injection while degludec requires daily dose adjustments. This is a reasonable concession, though it means patient-reported outcomes (nausea, satisfaction) carry some bias risk.

Tirzepatide doses escalated in 2.5 mg increments every four weeks to reduce GI side effects. Patients assigned to the 5 mg arm stayed at 5 mg after one escalation step. The 10 mg and 15 mg arms continued climbing. Degludec started at 10 units daily and was titrated based on three consecutive fasting glucose readings, following the manufacturer's label recommendations.

Randomization was stratified by baseline A1C, country, and concomitant SGLT2 inhibitor use. About 32% of participants were on an SGLT2 inhibitor at entry, a detail that matters because it approximates what real prescribers see today: many patients arrive already on empagliflozin or dapagliflozin before the insulin-vs-incretin decision is made.

The primary estimand used a treatment-policy approach, meaning all data were included regardless of whether patients discontinued study drug. A secondary estimand (efficacy) censored data after discontinuation. Both told the same story.

The Numbers in Context

Results at 52 weeks, as reported in the primary publication:

The A1C separation is notable on its own. But the weight delta is what changes practice. A 15 kg swing between the 15 mg tirzepatide arm and degludec over one year is not a secondary curiosity; it is a primary reason patients and clinicians will choose differently.

Clinically significant hypoglycemia (glucose below 54 mg/dL) occurred roughly 10 to 40 times more frequently with degludec, depending on the tirzepatide dose comparison. That disparity held even though degludec was titrated conservatively by trained investigators, not by patients adjusting doses alone at home.

What the Guidelines Have Done Since

The 2023 and 2024 ADA Standards of Care now explicitly recommend GLP-1 receptor agonists (including dual agonists) over basal insulin for most patients with type 2 diabetes who need injectable therapy, particularly those with established cardiovascular disease or obesity. This was a direct shift. Before the SURPASS program and the SUSTAIN trials for semaglutide, insulin and GLP-1 agonists sat side by side on the algorithm. Now the incretin class leads.

The FDA approved tirzepatide (Mounjaro) in May 2022 for type 2 diabetes in adults, with SURPASS-3 as a key registration trial. The label includes the head-to-head insulin data, which gave formulary committees a concrete reason to position tirzepatide ahead of basal insulin initiation in step therapy.

The European Association for the Study of Diabetes (EASD) and ADA joint consensus updated in 2022 reinforces this hierarchy. For patients with type 2 diabetes and obesity (BMI ≥30), the algorithm now routes through GLP-1 RA or dual GIP/GLP-1 RA before considering insulin at all.

Who in Your Practice This Actually Applies To

SURPASS-3 enrolled adults with a mean age of 57, mean BMI of 33.5, and mean diabetes duration of 8.4 years. Baseline A1C averaged 8.17%. These are the patients most primary care physicians and endocrinologists see every day: middle-aged, overweight, on metformin for years, and drifting above target.

The trial excluded patients with an eGFR below 45 mL/min, a history of pancreatitis, or personal/family history of medullary thyroid carcinoma. Those exclusions matter. The tirzepatide prescribing information carries a boxed warning about thyroid C-cell tumors based on rodent data, and renal dosing data remain limited below eGFR 45.

Patients who differ from the trial population in important ways:

• A1C above 10.5%. SURPASS-3 capped enrollment at 10.5%. Patients with A1C of 12% or higher, especially with symptoms of glucotoxicity, may still need insulin as a first injectable to break the hyperglycemic cycle rapidly. Guidelines from the ADA continue to recommend insulin for symptomatic hyperglycemia regardless of other therapy options.
• Type 1 diabetes or late autoimmune diabetes. Tirzepatide has no role here. The trial enrolled only confirmed type 2 diabetes.
• Patients already on prandial insulin. SURPASS-3 compared against basal insulin alone. Patients on basal-bolus regimens were not studied.
• Older adults (>75). Only 5.2% of participants were 75 or older, per the supplementary appendix. GI tolerability and weight loss in frail elderly patients may behave differently than what the trial captured.

Limitations the Authors Acknowledged

The open-label design is the most discussed limitation. Patients who know they are on a "newer" drug may report side effects and satisfaction differently. Investigators titrating degludec knew the fasting glucose targets, but real-world insulin titration is often less aggressive, which means the trial may actually underestimate the practical advantage of tirzepatide (since real-world insulin use tends to produce less A1C reduction than trial insulin use).

The 52-week duration is long enough to establish efficacy but short for durability questions. Will the A1C advantage persist at three or five years? The SURPASS-CVOT (SURPASS-4 extension and dedicated CV outcomes trial) provides some longer-term safety data, but true durability data for tirzepatide vs. insulin over multiple years remain thin.

GI adverse events were higher with tirzepatide. Nausea occurred in 12% to 24% of tirzepatide patients vs. 2% with degludec. Diarrhea and decreased appetite followed a similar pattern. Most GI events were mild to moderate and clustered during dose escalation. Still, for patients who cannot tolerate GI side effects, insulin remains the straightforward alternative.

What This Means for Prescribing Decisions Today

Three practical shifts have emerged since SURPASS-3 published:

Basal insulin is no longer the default first injectable. For the typical patient on metformin (with or without an SGLT2 inhibitor) whose A1C is between 7% and 10.5%, tirzepatide or semaglutide should be discussed before insulin degludec or glargine. The data support superior glycemic control, meaningful weight reduction, and far less hypoglycemia.

Insurance remains the bottleneck. The wholesale acquisition cost for Mounjaro exceeds $1,000/month without coverage, while insulin degludec generics and biosimilars continue to drop in price. Prior authorization requirements vary by insurer. For patients whose plans deny tirzepatide, basal insulin is still effective therapy, just not the better option when both are accessible.

The weight effect reframes the conversation. Clinicians who previously told patients, "You need to start insulin, and you may gain some weight," can now say, "There is an injectable option that controls blood sugar better than insulin and also reduces weight." That reframing changes adherence. Patients who feared insulin-associated weight gain now have a quantified alternative. The SURMOUNT-1 trial in obesity (without diabetes) showed even larger weight reductions with tirzepatide, reinforcing that the weight effect seen in SURPASS-3 is pharmacologically real, not an artifact of the diabetic population.

The Bottom Line for Clinicians

SURPASS-3 did not merely add another option. It reordered the algorithm. For the large population of adults with type 2 diabetes inadequately controlled on metformin, tirzepatide demonstrated superiority over optimally titrated basal insulin on both the outcome clinicians track most closely (A1C) and the outcome patients care about most (weight). Guidelines caught up within two years. The remaining barrier is cost and coverage, not evidence.

Frequently asked questions

›What was the primary endpoint of SURPASS-3?

The primary endpoint was change in A1C from baseline at 52 weeks. All three tirzepatide doses (5 mg, 10 mg, 15 mg) were statistically superior to insulin degludec on this measure.

›How much weight did patients lose on tirzepatide vs insulin degludec?

Tirzepatide patients lost 7.5 kg (5 mg), 10.7 kg (10 mg), or 12.9 kg (15 mg) on average, while degludec patients gained 2.3 kg over 52 weeks.

›Was SURPASS-3 double-blinded?

No. The trial was open-label because tirzepatide is a fixed weekly dose while degludec requires daily titration. Blinding was not feasible given the different dosing regimens.

›Did SURPASS-3 include patients on SGLT2 inhibitors?

Yes. About 32% of participants were on a concomitant SGLT2 inhibitor at baseline, reflecting common real-world prescribing patterns.

›How does tirzepatide compare to semaglutide?

SURPASS-3 did not compare tirzepatide to semaglutide directly. The SURPASS-2 trial compared tirzepatide to semaglutide 1 mg, and tirzepatide showed greater A1C and weight reduction at higher doses.

›Is tirzepatide approved by the FDA for type 2 diabetes?

Yes. The FDA approved tirzepatide (brand name Mounjaro) in May 2022 for glycemic control in adults with type 2 diabetes.

›What were the most common side effects of tirzepatide in SURPASS-3?

Nausea (12% to 24%), diarrhea (12% to 17%), and decreased appetite (6% to 11%) were the most common adverse events. Most were mild to moderate and occurred during dose escalation.

›Can tirzepatide replace insulin for all type 2 diabetes patients?

No. Patients with very high A1C (above 10.5%), symptomatic hyperglycemia, or those needing basal-bolus regimens may still require insulin. The ADA guidelines recommend insulin for symptomatic presentations regardless of other options.

›How was insulin degludec dosed in SURPASS-3?

Degludec started at 10 units daily and was titrated weekly to a fasting glucose target below 90 mg/dL, per the prescribing label. The mean dose at 52 weeks was approximately 49 units/day.

›Did SURPASS-3 measure cardiovascular outcomes?

No. SURPASS-3 was not powered for cardiovascular events. The dedicated cardiovascular outcomes trial for tirzepatide is SURPASS-CVOT.

References

1. Ludvik B, Giorgino F, Jodar E, et al. Once-weekly tirzepatide versus once-daily insulin degludec as add-on to metformin with or without SGLT2 inhibitors in patients with type 2 diabetes (SURPASS-3): a randomised, open-label, parallel-group, phase 3 trial. Lancet. 2021;398(10300):583-598. PubMed
2. ElSayed NA, Aleppo G, Aroda VR, et al. 9. Pharmacologic approaches to glycemic treatment: Standards of Care in Diabetes, 2023. Diabetes Care. 2023;46(Suppl 1):S140-S157. Diabetes Care
3. U.S. Food and Drug Administration. Mounjaro (tirzepatide) prescribing information. 2022. FDA Label
4. Davies MJ, Aroda VR, Collins BS, et al. Management of hyperglycemia in type 2 diabetes, 2022. A consensus report by the ADA and EASD. Diabetes Care. 2022;45(11):2753-2786. PubMed
5. Jastreboff AM, Aronne LJ, Ahmad NN, et al. Tirzepatide once weekly for the treatment of obesity. N Engl J Med. 2022;387(3):205-216. PubMed
6. U.S. Food and Drug Administration. Tresiba (insulin degludec) prescribing information. 2023. FDA Label