ORIGIN Extension Data and What Happened After the Trial Ended

At a glance

Parameter	Detail
N	12,537 (main trial); ~8,000 in extension
Intervention	Insulin glargine titrated to FPG <95 mg/dL
Comparator	Standard care
Duration	Median 6.2 years (main); ~2.5 additional years (extension)
Primary endpoint	Composite CV death, nonfatal MI, nonfatal stroke
Key result	HR 1.02 (95% CI 0.94-1.11) for CV composite; 28% reduction in new-onset diabetes during active treatment; benefit lost post-cessation

Why Extension Data Matters Here

The ORIGIN trial was designed to answer whether early insulin glargine could prevent cardiovascular events in people with dysglycemia (impaired fasting glucose, impaired glucose tolerance, or early type 2 diabetes) plus additional CV risk factors. The main trial ran from 2003 to 2011, reporting neutral CV outcomes and a 28% reduction in progression to overt diabetes among those with prediabetes at baseline.

But neutrality at 6 years does not guarantee neutrality at 10. And a 28% diabetes reduction during active treatment raises the question: does early insulin preserve beta-cell function, or does it merely suppress diagnostic glucose thresholds while the drug is on board?

The ORIGIN and Legacy Effects (ORIGINALE) extension study, published in 2014, attempted to answer these questions by following participants after trial completion.

The ORIGINALE Extension Design

After the main ORIGIN trial ended in February 2012, investigators invited all surviving participants to enter a passive observational follow-up. Key design elements:

• No continued randomized treatment. Participants returned to their local physicians for standard diabetes care. Insulin glargine was discontinued in the intervention arm unless clinically indicated.
• Endpoints tracked. The same composite CV endpoint (CV death, nonfatal MI, nonfatal stroke), all-cause mortality, new-onset diabetes, and cancer incidence.
• Follow-up duration. Approximately 2.5 additional years of observation beyond the main trial, for a total of roughly 8.5-9 years from randomization.
• Retention. Approximately 8,000 of the original 12,537 participants consented to extension follow-up. Loss was primarily due to death during the main trial, withdrawal, and site closures.

The extension was not powered independently. It was designed to detect legacy effects (persistent benefit or harm from the in-trial intervention period) rather than to test a new hypothesis.

What the Extension Showed

Cardiovascular Outcomes Remained Neutral

Extended follow-up confirmed the main trial finding. The original ORIGIN publication reported a hazard ratio of 1.02 (95% CI 0.94-1.11) for the primary CV composite. Through the extension period, no divergence of the Kaplan-Meier curves emerged. There was no late cardiovascular benefit and no late cardiovascular harm.

This finding is clinically important for two reasons. First, it rules out the theoretical concern that years of mild hyperinsulinemia from exogenous glargine might accelerate atherosclerosis over longer time horizons. Second, it confirms there is no delayed cardioprotective "legacy effect" analogous to what was seen with metformin in UKPDS post-trial monitoring.

Diabetes Prevention Did Not Persist

During the active treatment phase of ORIGIN, participants randomized to glargine who entered with prediabetes (IFG or IGT) had a 28% lower rate of progression to type 2 diabetes compared to standard care. This was a secondary outcome in the main trial.

In the extension, after glargine discontinuation, rates of new diabetes diagnosis converged between groups. The separation in cumulative diabetes incidence that had developed during the trial period did not widen further. Former glargine-treated participants developed diabetes at approximately the same rate as former standard-care participants once the intervention stopped.

The Regression Framework for Interpreting This Result:

This pattern is consistent with what was observed in the DPP/DPPOS trial with troglitazone (benefit lost after drug withdrawal) and contrasts sharply with the durable effects of lifestyle intervention in DPP, where weight loss and exercise continued to reduce diabetes risk years after the active intervention ended.

Cancer and Mortality

The main ORIGIN trial was partially motivated by the 2009 controversy over whether insulin glargine might increase cancer risk. The original publication found no signal (HR 1.00 for any cancer), and the extension confirmed this null finding with additional person-years of follow-up. No site-specific cancer showed excess incidence in the glargine arm through 8+ years of observation.

All-cause mortality remained balanced between arms through the extension period. This is the longest controlled safety dataset for any basal insulin analog in a population with early dysglycemia.

Weight and Hypoglycemia: What Carried Forward

During the main trial, the glargine arm gained approximately 1.6 kg more than standard care and experienced more hypoglycemia (1.00 vs 0.31 events per 100 person-years for severe episodes). In the extension, after glargine discontinuation, weight differences narrowed and hypoglycemia rates equalized. Neither the weight gain nor the hypoglycemia risk produced lasting metabolic consequences detectable in the extension period.

Methodological Considerations

Strengths of the Extension

The ORIGINALE study had several methodological strengths worth noting:

1. Large sample with long total follow-up. Even with attrition, ~8,000 participants over 8-9 total years provides substantial statistical power for major CV events.
2. Clean washout. Because glargine was stopped in most participants, the extension provides a true test of legacy effects rather than ongoing drug exposure.
3. Diverse population. ORIGIN enrolled across 40 countries, and the extension maintained geographic diversity.

Limitations

The extension also had important weaknesses:

• Non-randomized observation. Post-trial treatment was at physician discretion. Participants in the former glargine arm may have been managed differently due to familiarity with insulin.
• Attrition bias. The ~4,500 participants who did not enter the extension were older, sicker, and had more events during the trial. Their exclusion may bias toward null findings.
• Insufficient power for subgroups. The extension could not reliably test whether specific subgroups (e.g., those with IFG only vs. early T2D) experienced differential legacy effects.
• No mechanistic data. Beta-cell function (HOMA-B, C-peptide, disposition index) was not systematically measured in the extension, so the statement "no disease modification" rests on clinical glucose outcomes rather than direct physiological measurement.

Clinical Translation

What This Means for Prescribers

The ORIGIN extension data supports three practical conclusions:

First, basal insulin is cardiovascularly safe in early dysglycemia over long time horizons. Clinicians who initiate glargine in patients with early T2D and high CV risk are not adding cardiovascular hazard. This aligns with the FDA label for insulin glargine, which does not carry a CV warning.

Second, early basal insulin does not "cure" or durably modify prediabetes. The 28% diabetes reduction seen during active treatment was a pharmacological suppression of glucose, not a restoration of beta-cell capacity. This distinguishes insulin from lifestyle intervention and from certain pharmacological approaches (metformin in DPPOS showed modest persistent benefit).

Third, the weight gain and hypoglycemia from insulin glargine do not create lasting metabolic damage. While these are valid reasons to prefer other agents for diabetes prevention, they are reversible upon discontinuation.

Positioning Within Current Guidelines

The ADA Standards of Care (2024) does not recommend basal insulin for diabetes prevention. ORIGIN's extension data is one reason: the benefit does not persist. For established T2D, the ADA positions basal insulin as appropriate when other agents fail to achieve glycemic targets, and the ORIGIN CV safety data supports this positioning without the concern that early insulin exposure creates downstream harm.

The 2022 ADA/EASD consensus report on hyperglycemia management in T2D emphasizes cardioprotective agents (GLP-1 RAs, SGLT2 inhibitors) over insulin for patients with established CV disease. ORIGIN's neutral CV finding, while reassuring, means insulin glargine does not compete with agents that actively reduce CV events.

Comparison to Other Insulin CV Trials

ORIGIN remains unique in testing insulin in prediabetes/early diabetes rather than established T2D, and in having long-term post-trial follow-up data.

The Bigger Question ORIGIN Raised

ORIGIN was conceived in an era when the "early insulin hypothesis" proposed that resting beta cells with exogenous insulin might preserve their function. The trial and extension together effectively refuted this hypothesis at the clinical level. Six years of maintaining fasting glucose below 95 mg/dL with glargine did not produce any detectable lasting benefit on glycemic trajectory.

This has shaped how the field thinks about disease modification in T2D. Glucose-lowering alone is necessary but insufficient. Agents that show durable benefits (tirzepatide in SURMOUNT, semaglutide in SELECT) appear to work through mechanisms beyond glucose suppression, including weight loss, inflammation reduction, and possibly direct effects on beta-cell stress.

Frequently asked questions

›Did the ORIGIN extension show any late cardiovascular harm from insulin glargine?

No. Through approximately 8.5-9 total years from randomization (including ~2.5 years of post-trial observation), there was no divergence in cardiovascular event rates between the former glargine and standard care arms. The extended follow-up actually strengthens the cardiovascular safety case for basal insulin in early dysglycemia.

›Was the diabetes prevention effect of insulin glargine durable after stopping the drug?

No. The 28% reduction in new-onset diabetes observed during active treatment did not persist after glargine discontinuation. Rates of diabetes diagnosis converged between groups in the extension period, indicating pharmacological glucose suppression rather than true disease modification.

›How does ORIGIN's extension compare to the UKPDS legacy effect?

They differ substantially. UKPDS showed a persistent cardiovascular benefit from early metformin and sulfonylurea/insulin treatment in newly diagnosed T2D that emerged years after the trial ended. ORIGIN showed no such legacy effect. The populations differed (new T2D vs. prediabetes/early dysglycemia), and the mechanisms likely differ as well.

›Did ORIGIN resolve the insulin-cancer controversy?

Largely yes. Through 8+ years of follow-up including the extension period, there was no excess cancer incidence in the glargine arm. This provided the longest controlled safety observation for insulin glargine and effectively addressed the epidemiological concerns raised in 2009.

›Why doesn't early insulin preserve beta-cell function?

The ORIGIN extension did not include direct beta-cell measurements, so the mechanism is inferred. The most likely explanation is that exogenous insulin suppresses glucose but does not address the underlying pathophysiology driving beta-cell failure: lipotoxicity, inflammation, amyloid deposition, and progressive loss of beta-cell mass. Simply resting the cells is insufficient to reverse these processes.

›How many participants were in the ORIGIN extension?

Approximately 8,000 of the original 12,537 randomized participants consented to the observational extension (ORIGINALE). Attrition was due to death during the main trial, withdrawal, and closure of some study sites.

›Does ORIGIN's data support using basal insulin for prediabetes?

No. While glargine reduced diabetes incidence during active treatment, the benefit did not persist after stopping. Combined with weight gain and hypoglycemia risk, current guidelines (ADA, EASD) do not recommend basal insulin for diabetes prevention. Lifestyle intervention and metformin remain preferred.

›What is the longest cardiovascular safety data available for insulin glargine?

ORIGIN provides the longest controlled dataset, with approximately 8.5-9 years of total follow-up from randomization through the extension period. This exceeds DEVOTE (2 years) and other insulin cardiovascular outcome trials in both duration and population size.

›Did weight gain from insulin glargine in ORIGIN cause lasting metabolic harm?

No. The ~1.6 kg weight difference between arms during the trial narrowed after glargine discontinuation in the extension. There was no evidence that trial-period weight gain produced persistent metabolic consequences, and cardiovascular outcomes remained neutral throughout.

›How does ORIGIN inform current treatment algorithms for early type 2 diabetes?

ORIGIN confirms that basal insulin is safe but not disease-modifying in early dysglycemia. Current algorithms prioritize agents with proven CV benefit (GLP-1 RAs, SGLT2 inhibitors) or disease-modifying potential over insulin in early disease. Insulin is reserved for later-stage glycemic failure when other agents prove insufficient.

References

1. ORIGIN Trial Investigators. Basal insulin and cardiovascular and other outcomes in dysglycemia. N Engl J Med. 2012;367(4):319-328. https://pubmed.ncbi.nlm.nih.gov/22686416/
2. ORIGIN Trial Investigators. Cardiovascular and other outcomes postintervention with insulin glargine and omega-3 fatty acids (ORIGINALE). Diabetes Care. 2014;37(11):2970-2978. https://pubmed.ncbi.nlm.nih.gov/25150157/
3. Holman RR, Paul SK, Bethel MA, et al. 10-year follow-up of intensive glucose control in type 2 diabetes (UKPDS 80). N Engl J Med. 2008;359(15):1577-1589. https://pubmed.ncbi.nlm.nih.gov/18784090/
4. Diabetes Prevention Program Research Group. Long-term effects of lifestyle intervention or metformin on diabetes development and microvascular complications (DPPOS). Lancet Diabetes Endocrinol. 2015;3(11):866-875. https://pubmed.ncbi.nlm.nih.gov/25964225/
5. American Diabetes Association Professional Practice Committee. Standards of Care in Diabetes, 2024. Diabetes Care. 2024;47(Suppl 1). https://pubmed.ncbi.nlm.nih.gov/38078589/
6. FDA. Lantus (insulin glargine) prescribing information. https://www.accessdata.fda.gov/drugsatfda_docs/label/2019/021081s073lbl.pdf