What does ORIGIN stand for?

ORIGIN stands for Outcome Reduction with an Initial Glargine Intervention. The name reflects the hypothesis that starting insulin early might reduce cardiovascular events.

Did insulin glargine reduce heart attacks in the ORIGIN trial?

No. The rate of nonfatal myocardial infarction was nearly identical between the glargine and standard-care groups (HR 1.06 to 95% CI 0.93 to 1.20, p = 0.38). The trial was neutral for all cardiovascular endpoints.

Did ORIGIN show that insulin causes cancer?

No. Cancer incidence was 7.7 per 1,000 patient-years with glargine and 7.2 per 1,000 patient-years with standard care, a difference that was not statistically significant. This was an important finding given prior concern about insulin's mitogenic properties.

Can insulin glargine prevent type 2 diabetes?

During active treatment, glargine reduced new diabetes diagnoses by 28% among participants with pre-diabetes. However, the effect largely disappeared three months after stopping glargine, suggesting it suppressed hyperglycemia rather than altering the underlying disease process.

How long did the ORIGIN trial last?

The median follow-up was 6.2 years, making it one of the longest cardiovascular outcome trials in diabetes research.

What dose of insulin glargine was used in ORIGIN?

Participants started at 2 to 6 units per day, titrated to a fasting glucose target below 95 mg/dL. By the end of the trial, the median dose was approximately 0.40 units/kg/day.

How much weight gain did insulin glargine cause in ORIGIN?

Participants in the glargine group gained a median of 1.6 kg from baseline, while the standard-care group lost 0.5 kg. The net difference was about 2 kg over the course of the trial.

Is ORIGIN still relevant now that GLP-1 drugs exist?

ORIGIN remains relevant as safety evidence. It confirmed that basal insulin does not cause cardiovascular harm. However, its clinical influence has been partly superseded by CVOTs showing that GLP-1 receptor agonists and SGLT2 inhibitors provide active cardiovascular and renal benefit, which moved those drug classes ahead of insulin in treatment guidelines.

Was the ORIGIN trial blinded?

The insulin-versus-standard-care comparison was open-label, meaning participants knew their assignment. Endpoint adjudication was blinded. The omega-3-versus-placebo comparison within the same factorial design was double-blinded.

What was the fasting glucose target in the ORIGIN insulin arm?

The target was a fasting plasma glucose below 95 mg/dL (5.3 mmol/L), which is lower than many current clinical targets and reflects the trial's goal of testing aggressive early glucose normalization.

ORIGIN Trial: A Plain-English Overview of What It Established

By HealthRX Medical Team

Published May 25, 2026Updated May 25, 2026Last reviewed May 25, 2026

At a glance

| Field | Detail | |---|---| | Full title | Outcome Reduction with an Initial Glargine Intervention (ORIGIN) | | N | 12,537 | | Intervention | Insulin glargine (Lantus), titrated to a fasting glucose <95 mg/dL | | Comparator | Standard care (oral agents, lifestyle, or no pharmacotherapy) | | Median follow-up | 6.2 years | | Primary endpoint | Co-primary: (1) CV death, nonfatal MI, or nonfatal stroke; (2) composite #1 plus revascularization or hospitalization for heart failure | | Key result | Neutral on both co-primary endpoints (HR 1.02 and 1.04, both non-significant) | | Publication | NEJM, June 2012 |

The Question ORIGIN Set Out to Answer

By the mid-2000s, clinicians knew that people with impaired fasting glucose (IFG), impaired glucose tolerance (IGT), or early type 2 diabetes carry elevated cardiovascular risk. What nobody had tested in a large, randomized way was whether starting basal insulin early, before patients clearly "needed" it, could cut that risk. There was also a lingering worry in the opposite direction: could exogenous insulin, with its growth-factor-like signaling, actually cause cardiovascular harm or promote cancer? The ORIGIN trial was designed to answer both sides of that question.

A secondary aim was practical. If insulin glargine could normalize fasting glucose in people who were still in the pre-diabetes or early-diabetes window, would that delay or prevent frank type 2 diabetes?

Who Was Enrolled

ORIGIN recruited participants aged 50 or older from 573 sites in 40 countries. Eligibility required at least one cardiovascular risk factor plus one of the following glycemic categories:

Impaired fasting glucose (fasting plasma glucose 110 to 125 mg/dL)
Impaired glucose tolerance (2-hour post-load glucose 140 to 199 mg/dL)
Newly diagnosed type 2 diabetes (treated with no more than one oral agent)
Established type 2 diabetes on zero or one oral glucose-lowering drug

Roughly 12% of participants had IFG or IGT only, meaning they did not yet have diabetes at randomization. About 82% had established cardiovascular disease or additional risk factors beyond age. The population was intentionally broad, spanning the full spectrum from pre-diabetes to early diabetes, which made the trial's reach unusual compared to later cardiovascular outcome trials (CVOTs) that enrolled only people with established disease.

What Participants Received

The ORIGIN 2x2 Factorial Design

ORIGIN used a 2x2 factorial design, randomizing participants to two independent comparisons simultaneously:

Insulin glargine (titrated to fasting glucose <95 mg/dL) vs. standard care
Omega-3 fatty acids (1 g/day) vs. placebo

This means each participant was assigned to one of four groups: glargine + omega-3, glargine + placebo, standard care + omega-3, or standard care + placebo. The factorial structure allowed two questions to be answered in a single trial without doubling enrollment. The omega-3 comparison was reported separately and was also neutral.

The insulin glargine arm started with a bedtime injection of 2 to 6 units, titrated upward every week using a patient-driven algorithm. The target was a fasting glucose below 95 mg/dL. Median dose at end of follow-up was approximately 0.40 units/kg/day.

The standard care arm continued whatever glucose management their local physician recommended, which could include metformin, sulfonylureas, lifestyle modification, or no pharmacotherapy at all. The key constraint was that standard-care participants could not use insulin glargine unless it became medically necessary, though other insulins were permitted if needed over time.

This open-label design (participants knew which group they were in) is a legitimate concern, though endpoint adjudication was blinded.

What Was Measured

The co-primary endpoints were:

A three-component composite: cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke
An expanded five-component composite: the above three plus revascularization or hospitalization for heart failure

Secondary endpoints included all-cause mortality, each component of the composites individually, new-onset diabetes (in those without it at baseline), hypoglycemia rates, cancer incidence, weight change, and microvascular outcomes.

What ORIGIN Found

Cardiovascular Outcomes: No Signal in Either Direction

| Endpoint | Glargine (rate/100 py) | Standard care (rate/100 py) | HR (95% CI) | p | |---|---|---|---|---| | CV death, MI, or stroke | 2.94 | 2.85 | 1.02 (0.94 to 1.11) | 0.63 | | Expanded 5-component composite | 5.52 | 5.28 | 1.04 (0.97 to 1.11) | 0.27 | | All-cause mortality | 2.57 | 2.60 | 0.98 (0.90 to 1.08) | 0.70 | | Nonfatal MI | 1.18 | 1.11 | 1.06 (0.93 to 1.20) | 0.38 | | Nonfatal stroke | 0.65 | 0.61 | 1.03 (0.87 to 1.22) | 0.72 |

The confidence intervals for the primary composite comfortably included 1.0, and the point estimates hovered within 2 to 4 percent of unity. With over 12,000 participants and a median 6.2 years of follow-up, the trial had strong statistical power to detect even modest harm or benefit. The finding was genuinely neutral.

Diabetes Prevention: A Real but Complicated Win

Among the 1,456 participants who did not have diabetes at baseline, insulin glargine reduced the rate of new diabetes diagnosis by 28% (odds ratio 0.72 to 95% CI 0.58 to 0.91) during the active treatment period. However, three months after stopping glargine, the difference largely disappeared. This suggested that glargine was suppressing hyperglycemia rather than altering the underlying disease trajectory, a pattern consistent with findings from the earlier Diabetes Prevention Program with metformin, where stopping the drug also erased much of the benefit.

Safety: Hypoglycemia and Weight

| Outcome | Glargine | Standard care | |---|---|---| | Severe hypoglycemia (rate/100 py) | 1.00 | 0.31 | | Non-severe confirmed hypoglycemia | More frequent | Less frequent | | Weight change from baseline (median) | +1.6 kg | −0.5 kg | | Any cancer incidence | 7.7 per 1000 py | 7.2 per 1000 py (NS) |

Severe hypoglycemia was roughly three times more common with glargine, though the absolute rate remained low (about 1 episode per 100 patient-years). Weight gain averaged about 2 kg more than the standard-care group. Cancer rates were essentially identical, which was reassuring given that animal and observational studies had raised concern about insulin's mitogenic potential.

What These Results Actually Mean

The Good News

ORIGIN provided the first large-scale, randomized evidence that basal insulin does not increase cardiovascular risk in people with dysglycemia. Before ORIGIN, the safety debate was driven by observational data and biological speculation. After ORIGIN, clinicians could prescribe insulin glargine in early type 2 diabetes without worrying that they were trading glucose control for heart attacks.

This finding was important enough that the FDA's 2008 guidance requiring CVOTs for new diabetes drugs was partly motivated by the same safety uncertainty ORIGIN addressed. Every subsequent CVOT (EMPA-REG, LEADER, SUSTAIN-6) owed its regulatory framework to concerns that ORIGIN helped resolve.

The Less Useful News

ORIGIN did not show that early insulin prevents heart disease. The neutral result means clinicians cannot justify starting insulin glargine primarily for cardiovascular protection. This matters because the trial enrolled a population with substantial CV risk, so if early fasting-glucose normalization were going to help, this was a fair test.

The diabetes-prevention finding is interesting but clinically limited. Since the effect faded after stopping glargine, and since the intervention required daily injections with more hypoglycemia and weight gain, no guideline recommends insulin glargine for diabetes prevention. The ADA Standards of Care continues to recommend metformin, lifestyle modification, or (more recently) GLP-1 receptor agonists for this purpose.

Design Strengths and Honest Limitations

Strengths:

One of the largest and longest diabetes CVOTs ever conducted
Enrolled a broad dysglycemia spectrum, not just established T2D
Factorial design efficiently tested two hypotheses
Blinded endpoint adjudication reduced bias

Limitations:

Open-label for the insulin comparison, so behavioral differences between arms (more frequent glucose monitoring in the glargine group) could have influenced some secondary outcomes
The standard-care arm was heterogeneous across 40 countries, making it harder to isolate the specific effect of glargine versus a uniform comparator
Median HbA1c difference between arms was only about 0.3% by end of study, so the glycemic separation was modest. A larger glucose gap might have revealed a CV signal, positive or negative
The trial predated SGLT2 inhibitors and GLP-1 receptor agonists, so the standard-care arm does not reflect current treatment patterns

Where ORIGIN Fits in the Broader Evidence

ORIGIN was neutral. The subsequent DEVOTE trial (2017) confirmed cardiovascular safety for the newer insulin degludec as well. Neither trial showed CV benefit.

By contrast, the GLP-1 receptor agonist and SGLT2 inhibitor CVOTs (LEADER, EMPA-REG OUTCOME, DAPA-HF) showed actual cardiovascular and renal benefit, which reshaped prescribing hierarchies. Today, basal insulin has moved further down the ADA/EASD treatment algorithm for type 2 diabetes, used when oral agents and injectables with proven CV or renal benefit are insufficient. ORIGIN's legacy is less "use glargine early" and more "glargine is safe when you do need it."

Frequently asked questions

›

References

ORIGIN Trial Investigators. "Basal insulin and cardiovascular and other outcomes in dysglycemia." New England Journal of Medicine 367, no. 4 (2012): 319-328. PubMed
Marso SP, et al. "Efficacy and safety of degludec vs glargine in type 2 diabetes (DEVOTE)." New England Journal of Medicine 377, no. 8 (2017): 723-732. PubMed
Diabetes Prevention Program Research Group. "Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin." New England Journal of Medicine 346, no. 6 (2002): 393-403. PubMed
FDA Guidance for Industry. "Diabetes mellitus: evaluating cardiovascular risk in new antidiabetic therapies to treat type 2 diabetes." (2008). FDA.gov
American Diabetes Association. "Standards of Care in Diabetes, 2024." Diabetes Care 47, Supplement 1 (2024). Diabetes Care
Davies MJ, et al. "Management of hyperglycemia in type 2 diabetes, 2022: a consensus report by ADA and EASD." Diabetes Care 45, no. 11 (2022): 2753-2786. Diabetes Care