SURPASS-3
SURPASS-3 (Lancet 2021, PMID 34370970) compared tirzepatide (5mg, 10mg, 15mg once-weekly) with titrated insulin degludec (once-daily) as add-on to metformin with or without SGLT2 inhibitors in 1444 insulin-naive adults with T2D. The 52-week, open-label trial was conducted at 122 sites across 13 countries. The population reflects a clinically relevant group — patients who have progressed beyond oral agents but have not yet started insulin. Baseline HbA1c averaged 8.17% and mean body weight was 94.3 kg.
Primary endpoint: Tirzepatide 10mg and 15mg were noninferior to degludec in HbA1c reduction — and both met superiority. HbA1c reductions were −1.93% (5mg), −2.20% (10mg), and −2.37% (15mg) versus −1.34% for degludec. Estimated treatment differences vs degludec were −0.59%, −0.86%, and −1.04% for the three doses respectively (all P<0.0001 for superiority). All three doses of tirzepatide achieved HbA1c <7.0% in 82–93% of patients versus 61% on degludec.
Weight: The contrast with insulin degludec was striking on body weight. All three tirzepatide doses produced weight loss (−7.5 kg, estimated −10.7 kg, −12.9 kg from baseline of 94.3 kg) versus a weight gain of +2.3 kg with insulin degludec. Estimated treatment differences were −9.8 to −15.2 kg (all P<0.0001). This weight divergence has important long-term implications for cardiovascular and metabolic risk in T2D patients transitioning from oral agents to injectable therapy.
Hypoglycaemia: A critical advantage for tirzepatide: hypoglycaemia (<54 mg/dL or severe) occurred in only 1–2% of tirzepatide-treated patients versus 7% of degludec-treated patients. This positions tirzepatide as a much safer option than basal insulin for patients who experience frequent hypoglycaemia on insulin, or who are at high hypoglycaemia risk (elderly patients, renally impaired, those with hypoglycaemia unawareness).
GI side effects: The main disadvantage vs basal insulin — nausea (12–24% tirzepatide vs 2% degludec), diarrhoea (15–17% vs 4%), vomiting (6–10% vs 1%) — predominantly mild to moderate and decreasing over time.
Substudies: The SURPASS-3 MRI substudy demonstrated significant reductions in liver fat content and visceral adipose tissue with tirzepatide versus insulin degludec — important for patients with metabolic-associated steatotic liver disease (MASLD/NAFLD). The CGM substudy showed superior time-in-range and 24h glucose profiles with tirzepatide.
Vault context: SURPASS-3 fills the clinical niche of insulin initiation vs escalation. A T2D patient inadequately controlled on metformin ± SGLT2i who needs injectable therapy faces a choice: start basal insulin (standard) or tirzepatide (more effective but costlier and with more GI side effects). SURPASS-3 makes a strong case for tirzepatide in motivated patients willing to manage GI side effects, particularly when weight gain with insulin is a concern.
Ontology SURPASS-3 [evidence_for] Tirzepatide SURPASS-3 [relates] Insulin Degludec SURPASS-3 [relates] SURPASS Clinical Trial Programme SURPASS-3 [relates] HbA1c Reduction SURPASS-3 [relates] Weight Loss in T2D
Numbers
| Arm | HbA1c Δ | Weight Δ | HbA1c <7.0% | Hypoglycaemia |
|---|---|---|---|---|
| Tirzepatide 5mg | −1.93% | −7.5 kg | 82% | 1% |
| Tirzepatide 10mg | −2.20% | ~−10.7 kg | 90% | 1% |
| Tirzepatide 15mg | −2.37% | −12.9 kg | 93% | 2% |
| Insulin degludec | −1.34% | +2.3 kg | 61% | 7% |
Baseline HbA1c: 8.17% | Baseline weight: 94.3 kg | Duration: 52 weeks
Practical Interpretation
- Who this applies to: insulin-naive T2D patients on metformin ± SGLT2i who need injectable therapy escalation
- Key advantage over basal insulin: weight loss vs weight gain; markedly lower hypoglycaemia; superior HbA1c
- Key disadvantage: GI side effects 6–15× higher than basal insulin; cost; injection pen vs insulin
- MRI substudy: reduced liver fat — potentially important for MASLD co-management
Connections
- Tirzepatide — evidence_for (T2D insulin-naive escalation)
- Insulin Degludec — compared_to (active comparator, SURPASS-3)
- SURPASS Clinical Trial Programme — part_of
- HbA1c Reduction — evidence_from
- Weight Loss in T2D — evidence_from
- Hypoglycaemia Risk — evidence_from (1–2% vs 7%)