SURPASS-6
Tirzepatide vs Insulin Lispro Added to Basal Insulin in Type 2 Diabetes: The SURPASS-6 Randomized Clinical Trial.
Rosenstock J, Frías JP, Rodbard HW, Tofé S, Sears E, Huh R, Fernández Landó L, Patel H. · JAMA
Study record
- Population
1,428 adults randomized with basal-insulin-treated type 2 diabetes; 1,425 treated.
Read source · Full-text reportResults — Baseline Characteristics- Formulation
The protocol specifies tirzepatide single-dose autoinjector pens, insulin lispro U100 prefilled pens with 3 mL cartridges, and background insulin glargine U100 pens. The table allows sourcing, packaging and labeling to vary by country or during the study; this is not batch-level verification.
Read source · Study protocolProtocol I8F-MC-GPHD(c), Section 6.1, Table 6, printed page 61 (PDF page 62)- Route
Subcutaneous.
Read source · Full-text reportMethods — Randomization and Blinding- Dose & frequency
Tirzepatide: 2.5 mg weekly, increasing 2.5 mg every four weeks toward 5/10/15 mg. All groups reduced glargine 30% at randomization under supervision.
Read source · Full-text reportMethods — Interventions- Study design
Open-label randomized comparison with mealtime insulin lispro; continued glargine with or without metformin.
Read source · Full-text reportMethods- Duration
52 weeks; four-week safety follow-up.
Read source · Full-text reportMethods — Randomization and Blinding- Reported findings
Treatment-regimen analysis: pooled HbA1c change −2.1 versus −1.1 percentage points; difference −0.98 (95% CI −1.17 to −0.79).
Read source · Full-text reportResults — Primary Outcome- Adverse events
Serious events: 44/717 versus 77/708. Hypoglycemia below 54 mg/dL or severe: 0.41 versus 4.38 events/patient-year; gastrointestinal events were frequent.
Read source · Full-text reportTables 3–4- Monitoring
Glucose-guided insulin titration, vital signs, adverse-event surveillance and pancreatitis adjudication.
Read source · Full-text reportMethods — Interventions; Outcomes
What it supports
Compares two additions to basal insulin.
What it does not establish
Not unsupervised pairing guidance. Safety rates reflect protocol-managed insulin. Corrected Tables 2–3 distinguish dose groups and study withdrawal from treatment discontinuation.
Funding
Eli Lilly (Article Information).
Publication and registry context
The registry primary analysis excludes data after rescue medication or early treatment discontinuation: HbA1c change −2.26 versus −1.16 percentage points, rather than the treatment-regimen estimates above. Its safety table through week 56 reports 53/717 pooled tirzepatide participants with serious events versus 81/708 with insulin lispro; the publication reports 44/717 versus 77/708. The sources differ; the cause of the safety-count difference has not been established here, so the counts remain separately attributed.
Inspect NCT04537923 resultsLiterature search cutoff: 2026-09-08 · Not independently clinically reviewed
Browse study library