Long-Term T2D Outcomes Beyond 2 Years
The original SURPASS trials were 40–104 weeks — adequate for regulatory approval but insufficient to characterise the durability of tirzepatide’s glycaemic effects across the multi-decade trajectory of T2D. This gap was partially addressed by a 104-week post-hoc extension analysis of SURPASS-4, which provides the longest T2D-specific metabolic outcome data currently available. However, no published data examines tirzepatide’s T2D effects beyond two years.
SURPASS-4 104-week post-hoc (Diabetes Obes Metab 2025, PMC 12515763): An extension analysis of SURPASS-4 (tirzepatide vs insulin glargine in high-CV-risk T2D, n=2002) followed patients through Week 104. Key findings:
- HbA1c sustained at 2 years: tirzepatide 5mg −2.3%, 10mg −2.5%, 15mg −2.6% vs glargine −1.0% (all doses maintained superiority with no signal of attenuation vs Week 40/52 data)
- Insulin avoidance maintained: tirzepatide-treated patients continued to avoid insulin initiation at 104 weeks, consistent with the 52-week primary analysis
- Kidney endpoints: eGFR slope preservation vs glargine maintained at 104 weeks (the 3.7 mL/min/1.73m²/year between-group benefit reported in the CKD post-hoc)
- Safety: no new safety signals emerging at 104 weeks vs 52-week data; GI adverse events largely resolved after escalation phase
This is a post-hoc analysis, not a pre-specified extension. The cohort available at Week 104 is a survivor cohort (those who did not discontinue or die), introducing attrition bias — the sustained HbA1c result may overestimate real-world durability.
Ontology Long-Term T2D Outcomes [evidence_from] SURPASS-4 Long-Term T2D Outcomes [relates] HbA1c Reduction Long-Term T2D Outcomes [relates] Tirzepatide
SURMOUNT-1 3-year data (NEJM 2025, PMID 39536238): A pre-specified extension of SURMOUNT-1 followed 1,032 participants with obesity AND prediabetes to 176 weeks (3 years). Weight loss at 176 weeks: −12.3% (5mg), −18.7% (10mg), −19.7% (15mg) vs −1.3% placebo. T2D incidence during the 176-week on-treatment period: 1.3% (tirzepatide) vs 13.3% (placebo) — HR 0.07 (95% CI 0.0–0.1), a 93% relative risk reduction. This is not T2D evidence — it is obesity+prediabetes data — but it documents two important facts: (1) weight loss is maintained without significant plateau across 3 years; (2) the glucose-normalising effect in prediabetes is durable on-treatment. Both are plausible proxies for sustained T2D benefit, but cannot be directly extrapolated.
SURPASS-CVOT (5 years): This trial followed 13,299 T2D+ASCVD patients for a median 4 years and collected MACE endpoints. It is NOT a source of 4-year HbA1c or weight data — metabolic outcomes were not systematically collected beyond baseline. The CV results confirm safety and non-inferiority at 5 years but say nothing about metabolic durability.
Ontology Long-Term T2D Outcomes [relates] SURPASS-CVOT Long-Term T2D Outcomes [relates] Beta-Cell Function Long-Term T2D Outcomes [relates] Stopping Effects and Treatment Dependence
What Is Known vs Unknown
| Aspect | Status |
|---|---|
| HbA1c sustained at 2 years (T2D) | Confirmed — SURPASS-4 104-week post-hoc |
| HbA1c beyond 2 years (T2D) | Not available — no data |
| Weight loss sustained at 3 years (obesity) | Confirmed — SURMOUNT-1 extension |
| Beta-cell function preservation at 2+ years | Not reported — HOMA2-B not tracked at 104 weeks |
| Insulin dose reduction durability | Partial — SURPASS-4 post-hoc shows maintained insulin avoidance |
| MACE over 5 years | Confirmed (non-inferiority vs dulaglutide) — SURPASS-CVOT |
| HbA1c >2 years | Confirmed absent — not available; SURMOUNT-1 prediabetes is closest proxy |
Connections
- SURPASS-4 — primary evidence source for 104-week T2D data
- HbA1c Reduction — durability confirmed at 2 years
- Beta-Cell Function — long-term beta-cell preservation status unknown
- Stopping Effects and Treatment Dependence — reversal on cessation confirms pharmacological (not structural) mechanism