The Lancet October 26, 2019

Glycaemic durability of an early combination therapy with vildagliptin and metformin versus sequential metformin monotherapy in newly diagnosed type 2 diabetes (VERIFY): a 5-year, multicentre, randomised, double-blind trial

David R Matthews, Päivi M Paldánius, Pieter Proot, Yiyun Chiang, Michael Stumvoll, Stefano Del Prato, et al.

Bottom Line

In patients with newly diagnosed type 2 diabetes, early combination therapy with vildagliptin and metformin significantly reduced the risk of initial treatment failure and improved long-term glycemic durability compared to sequential metformin monotherapy.

Key Findings

1. Early combination therapy resulted in a 49% relative risk reduction for time to initial treatment failure (HbA1c ≥7.0% at two consecutive visits 13 weeks apart) compared to sequential metformin monotherapy (HR 0.51, 95% CI 0.45-0.58; p<0.0001) [2.2.4].
2. The incidence of initial treatment failure over 5 years was significantly lower in the early combination group (43.6%; 429 of 998 patients) versus the monotherapy group (62.1%; 614 of 1,003 patients).
3. The median time to initial treatment failure was extended to an estimated 61.9 months in the combination therapy arm, compared with 36.1 months in the standard stepwise monotherapy arm.
4. Among patients who experienced an initial treatment failure, those in the early combination group had a significantly lower risk of a secondary treatment failure compared to the stepwise group that sequentially added vildagliptin (HR 0.74, 95% CI 0.63-0.86; p<0.0001).
5. Adverse events and serious adverse events were comparable between the two treatment strategies, with no increased risk of hypoglycemia in the early combination group.

Study Design

Design
RCT
Double-Blind
Sample
2,001
Patients
Duration
5 yr
Median
Setting
34 countries
Population Adults aged 18-70 years with newly diagnosed type 2 diabetes (duration ≤2 years), treatment-naive or on minimal metformin, with an HbA1c of 6.5-7.5% and a BMI of 22-40 kg/m².
Intervention Early combination therapy of vildagliptin (50 mg twice daily) plus metformin (1,000-2,000 mg daily).
Comparator Standard-of-care sequential therapy starting with metformin monotherapy (1,000-2,000 mg daily) plus placebo, with vildagliptin (50 mg twice daily) added upon initial treatment failure.
Outcome Time to initial treatment failure, defined as an HbA1c ≥7.0% at two consecutive scheduled visits 13 weeks apart.

Study Limitations

The trial was strictly designed to evaluate glycemic durability and lacked the statistical power to assess differences in hard cardiovascular or microvascular outcomes.
The maximum allowed dose of metformin was capped at 2,000 mg/day, whereas standard clinical practice often permits titration up to 2,550 mg/day, which may have slightly disadvantaged the monotherapy arm.
Enrollment was restricted to patients with a relatively narrow baseline HbA1c range of 6.5% to 7.5%, limiting the generalizability of the findings to patients presenting with more severe initial hyperglycemia.
The trial protocol allowed for open-label insulin rescue therapy if clinically indicated, which could potentially confound the long-term observation of oral therapy efficacy.

Clinical Significance

The VERIFY trial established early combination therapy as a highly effective, safe, and proactive approach to managing newly diagnosed type 2 diabetes. By halving the risk of early treatment failure and significantly extending the time patients remain at their glycemic goal, the trial provided compelling evidence to abandon the traditional "treat-to-failure" monotherapy strategy in favor of early, dual-mechanism intervention.

Historical Context

For decades, major societal guidelines (such as ADA and EASD) recommended a stepwise approach for managing newly diagnosed type 2 diabetes: initiating therapy with lifestyle modifications and metformin monotherapy, and sequentially adding a second agent only after glycemic targets were lost. The VERIFY trial fundamentally challenged this clinical inertia. By demonstrating that early combination of a DPP-4 inhibitor (to enhance beta-cell function) and metformin (to improve insulin sensitivity) altered disease progression and preserved beta-cell function longer, VERIFY catalyzed a paradigm shift toward early combination therapy in modern clinical guidelines.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

What are the complementary mechanisms of action between metformin and the DPP-4 inhibitor vildagliptin that make them a physiologically logical early combination for newly diagnosed type 2 diabetes?

Key Response

Metformin primarily reduces hepatic glucose production and improves peripheral insulin sensitivity, but it does not halt progressive beta-cell decline. Vildagliptin prolongs the action of endogenous incretin hormones (GLP-1 and GIP), which stimulates glucose-dependent insulin release and suppresses inappropriate glucagon secretion. Together, they simultaneously address multiple core pathophysiologic defects of type 2 diabetes without increasing the risk of hypoglycemia.

Resident
Resident

Traditionally, clinical algorithms recommended initiating combination therapy only if a patient's baseline HbA1c was significantly elevated (e.g., >1.5% above target). Based on the VERIFY trial, how might you change your management approach for a newly diagnosed T2DM patient presenting with a mildly elevated HbA1c of 7.2%?

Key Response

The VERIFY trial specifically enrolled newly diagnosed patients with relatively mild hyperglycemia (HbA1c 6.5-7.5%). It demonstrated that even in this demographic, early combination therapy significantly delayed the time to treatment failure compared to sequential add-on therapy. For residents, this means considering early combination therapy not merely as a rescue strategy for severe hyperglycemia, but as a proactive, disease-modifying approach to preserve long-term glycemic durability.

Fellow
Fellow

The VERIFY trial observed a delay in both the first and second instances of treatment failure, suggesting a potential 'legacy effect.' What does this durability imply about underlying beta-cell function, and how does early intervention with a DPP-4 inhibitor compare theoretically to early combination with a GLP-1 RA or SGLT2i?

Key Response

The delay in the second failure (which occurred after both groups were eventually on combination therapy) suggests that early, intensive glycemic control may preserve beta-cell mass or function, fundamentally altering the disease's natural history rather than just acutely lowering glucose. Fellows should critically consider whether newer agents like GLP-1 RAs or SGLT2is, which offer robust cardiovascular, renal, and weight-loss benefits, would produce an even greater durability or legacy effect compared to the DPP-4 inhibitor utilized in this trial.

Attending
Attending

While the VERIFY trial provides strong evidence for the glycemic durability of early combination therapy, how should clinicians balance the upfront polypharmacy, increased costs, and potential for therapeutic inertia against the long-term clinical benefits in a real-world primary care setting?

Key Response

Attendings must weigh clinical trial efficacy against real-world effectiveness. While VERIFY supports early combination to bypass clinical inertia (the common real-world delay in adding medications when A1c rises on monotherapy), starting two drugs immediately increases initial cost and pill burden. The discussion should center on utilizing single-pill combinations to optimize medication adherence and framing this strategy to patients as a proactive measure for 'beta-cell preservation' rather than a sign of severe disease.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

The VERIFY trial utilized a primary endpoint of 'time to initial treatment failure' defined as an HbA1c >= 7.0% at two consecutive visits. How might the fixed interval of HbA1c testing (every 13 weeks) and the specific 7.0% threshold have introduced interval-censoring bias into the Kaplan-Meier survival curves?

Key Response

The rigid 13-week testing schedule and the strict 7.0% threshold directly influence the recorded timing of the failure event, creating interval censoring where the true failure time is only known to occur within a specific window. Furthermore, defining failure at 7.0% in patients starting near 7.0% (mean baseline 6.7%) might disproportionately capture early, minor glycemic fluctuations in the monotherapy group, potentially amplifying the early separation of the survival curves compared to utilizing a continuous longitudinal mixed-effects model of HbA1c trajectories over time.

Journal Editor
Journal Editor

As a peer reviewer assessing the external validity and trial design of VERIFY, what are the primary methodological critiques regarding the protocol-mandated choice of the sequential add-on agent (vildagliptin) rather than allowing standard-of-care escalation, and how might this impact the generalizability of the findings?

Key Response

A critical reviewer would flag that the trial forces the sequential monotherapy group to eventually receive vildagliptin as their second agent. In modern real-world practice, providers often escalate to agents with proven cardiorenal benefits (SGLT2 inhibitors or GLP-1 receptor agonists). The editor must evaluate whether the observed glycemic durability is a generalizable principle of 'early combination therapy' or specifically tied to vildagliptin, especially given that the lack of hard cardiovascular outcomes limits the trial's applicability to contemporary risk-stratified treatment algorithms.

Guideline Committee
Guideline Committee

Current ADA/EASD guidelines recommend early combination therapy primarily for patients with high baseline HbA1c or established high cardiovascular risk. Does the evidence from VERIFY warrant upgrading the guidelines to mandate early combination therapy for all newly diagnosed T2DM patients regardless of baseline HbA1c, and what level of evidence would support this change?

Key Response

The VERIFY trial provides Level A evidence that early combination therapy improves glycemic durability in newly diagnosed patients with baseline HbA1c between 6.5-7.5%. However, updating guidelines to mandate this approach requires synthesizing this glycemic benefit with costs, patient preferences, and the modern shift toward cardiorenal risk-driven prescribing. The committee would likely issue a strong recommendation (Level A) to *consider* early combination therapy to delay glycemic failure, but stop short of a universal mandate due to the trial's focus on intermediate glycemic endpoints rather than hard mortality or macrovascular outcomes.

Clinical Landscape

Noteworthy Related Trials

1998

UKPDS 34

n = 1,704 · Lancet

Tested

Metformin monotherapy

Population

Overweight patients with newly diagnosed T2DM

Comparator

Conventional therapy (diet alone)

Endpoint

Any diabetes-related endpoint or mortality

Key result: Metformin significantly reduced diabetes-related endpoints and all-cause mortality compared to conventional diet therapy.
2006

ADOPT Trial

n = 4,360 · NEJM

Tested

Monotherapy with rosiglitazone, metformin, or glyburide

Population

Recently diagnosed, drug-naive T2DM patients

Comparator

Active comparators (rosiglitazone vs metformin vs glyburide)

Endpoint

Time to monotherapy failure (FPG > 180 mg/dL)

Key result: All monotherapies showed a progressive loss of glycemic control over time, demonstrating that beta-cell function declines despite initial monotherapy success.
2022

GRADE Trial

n = 5,047 · NEJM

Tested

Addition of glargine, glimepiride, liraglutide, or sitagliptin

Population

T2DM patients with duration <10 years already on metformin

Comparator

Head-to-head comparison of the four add-on agents

Endpoint

Time to primary metabolic failure (HbA1c >= 7.0%)

Key result: Liraglutide and insulin glargine were more effective in maintaining target HbA1c levels than glimepiride and sitagliptin.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis