Prediabetes Treatment Failure: What Counts and What Comes Next

What Is Prediabetes and Why the Definition Matters

Prediabetes is diagnosed when fasting plasma glucose is 100-125 mg/dL, hemoglobin A1c is 5.7-6.4%, or a 2-hour oral glucose tolerance test (OGTT) result is 140-199 mg/dL. These thresholds come directly from the American Diabetes Association (ADA) Standards of Medical Care, which are updated annually and are the most widely used clinical reference in the United States. The 2024 ADA Standards of Medical Care in Diabetes place prediabetes management under Section 3 and recommend structured lifestyle intervention as first-line care for all adults who meet these criteria.

The definition matters because the cutoff for type 2 diabetes diagnosis is fasting glucose at or above 126 mg/dL or A1c at or above 6.5% confirmed on a second test. Treatment failure in prediabetes is, functionally, crossing that line.

The Three Diagnostic Criteria Compared

Each of the three prediabetes tests captures different physiology. Fasting glucose reflects hepatic glucose output. The 2-hour OGTT result reflects postprandial insulin secretion and peripheral glucose disposal. A1c reflects average glycemia over roughly 90 days.

A patient can be "normal" on fasting glucose and still have impaired glucose tolerance on OGTT. For this reason, the ADA acknowledges that A1c 5.7-6.4% has lower sensitivity for identifying prediabetes than OGTT. One large population-based study in JAMA Internal Medicine found that using A1c alone missed approximately 70% of individuals who had impaired glucose tolerance by OGTT criteria.

Why Some Patients Progress Faster

Risk of progression is not uniform. A person with A1c of 6.3%, BMI of 38, a parent with type 2 diabetes, and a prior gestational diabetes history has a materially different trajectory than someone with A1c of 5.8% and no other risk factors. The ADA identifies the following as high-risk features: A1c at or above 6.0%, BMI at or above 35, age <60, previous gestational diabetes, and a first-degree relative with type 2 diabetes. These patients should receive pharmacotherapy earlier, regardless of whether lifestyle intervention has technically "failed."

How to Define Treatment Failure in Clinical Practice

Treatment failure in prediabetes has no single universally adopted definition, but the major guidelines give clinicians enough information to make a clear determination. Failure is best understood as either glycemic progression or persistent high-risk glycemia despite adequate intervention.

Glycemic Progression to Diabetes

The clearest marker of treatment failure is confirmed progression to type 2 diabetes. Per ADA diagnostic criteria, this requires either:

• Fasting plasma glucose at or above 126 mg/dL on two separate tests
• A1c at or above 6.5% confirmed on a repeat sample
• 2-hour OGTT glucose at or above 200 mg/dL
• Random plasma glucose at or above 200 mg/dL with classic symptoms

If any of those thresholds appear on two tests, or once with classic symptoms, the diagnosis of type 2 diabetes is made. That is the definitive end-point of prediabetes treatment failure.

Persistent High-Risk Glycemia Without Full Progression

A patient who has completed 6 months of structured lifestyle intervention and still has A1c of 6.2% or fasting glucose consistently above 115 mg/dL may not yet have type 2 diabetes, but the trajectory is unfavorable. The 2023 AACE Comprehensive Diabetes Management Algorithm states that pharmacotherapy is appropriate for high-risk individuals who do not achieve target glycemic improvement with lifestyle measures alone.

"High-risk" by AACE standards includes A1c at or above 6.0%, obesity, or metabolic syndrome components. If a patient hits the 6-month mark without meaningful A1c reduction (typically defined as at least 0.2-0.3 percentage points) and still sits above 6.0%, that should prompt a pharmacotherapy conversation.

Rising A1c Trend as an Early Signal

An A1c that was 5.9% twelve months ago and is now 6.3% after lifestyle modification signals accelerating beta-cell dysfunction. The rate of change may be as clinically meaningful as the absolute value. A 2019 analysis published in Diabetes Care found that individuals with prediabetes who progressed to type 2 diabetes showed a steeper A1c slope in the 2 years before crossing the diagnostic threshold, underscoring the importance of tracking trajectory rather than single-point values.

The Landmark Evidence: What Lifestyle Can and Cannot Do

Before declaring failure, it is worth confirming that the lifestyle intervention was actually delivered at the standard the evidence supports. The Diabetes Prevention Program (DPP) is the benchmark.

The Diabetes Prevention Program Results

The DPP (N=3,234) randomized adults with prediabetes to intensive lifestyle modification, metformin 850 mg twice daily, or placebo. Published in the New England Journal of Medicine in 2002, the trial showed:

• Intensive lifestyle intervention reduced T2D incidence by 58% relative to placebo over a mean 2.8 years
• Metformin reduced T2D incidence by 31% relative to placebo
• The lifestyle group achieved a mean 5.6 kg weight loss at 6 months
• The A1c goal was not a specified primary target, but weight loss of 7% and 150 minutes per week of moderate activity were the behavioral targets

If a patient has genuinely achieved 7% weight loss and 150 minutes per week of exercise for at least 6 months and A1c is still climbing, that qualifies as lifestyle treatment failure by DPP standards. Most patients in real-world settings do not hit these targets. A clinician should assess adherence before labeling something a failure.

The DPP Outcomes Study: Long-Term Picture

The DPP Outcomes Study (DPPOS), which followed the original cohort for up to 10 years, found that the lifestyle group's advantage narrowed over time as weight regain occurred. By year 10, cumulative T2D incidence in the lifestyle group was 49% lower than placebo at the start but only modestly better than metformin. This tells us that lifestyle modification, even when it initially works, can fail over the long term without maintenance support.

What "Adequate Lifestyle Intervention" Looks Like

The CDC-recognized National Diabetes Prevention Program (DPP) requires at least 16 sessions over the first year with a trained lifestyle coach. CDC program data show that participants who complete at least 9 sessions achieve an average weight loss of 4-5%, and those who complete all 16 sessions average closer to 6-7%.

If a patient has only made informal dietary changes without a structured program, lifestyle modification has not truly been given a full trial. That is a different situation from a patient who completed all 16 sessions, hit the weight and exercise targets, and still progressed.

Pharmacotherapy After Lifestyle Failure

When treatment failure is established, the next question is which drug to use. The evidence base is clearest for metformin. It grows rapidly for GLP-1 receptor agonists.

Metformin: First-Line Pharmacotherapy

The ADA Standards of Medical Care explicitly recommend considering metformin for adults with prediabetes who are at very high risk, particularly those with BMI at or above 35, age <60, or prior gestational diabetes. Section 3 of the 2024 ADA Standards states: "Metformin therapy for prevention of type 2 diabetes should be considered in those with prediabetes, especially for those with BMI >35 kg/m2, those aged <60 years, and women with prior gestational diabetes mellitus."

The standard dosing used in DPP was 850 mg twice daily, titrated over several weeks to reduce GI side effects. Long-term safety data across more than 15 years of DPPOS follow-up confirm that metformin is well tolerated and does not meaningfully increase the risk of lactic acidosis in patients with normal renal function.

One practical note: metformin modestly reduces B12 absorption with long-term use. A Diabetes Care study found that B12 deficiency occurred in 7% of metformin users over 4 years. Annual B12 monitoring is reasonable for patients on long-term therapy.

GLP-1 Receptor Agonists: Emerging Evidence

GLP-1 receptor agonists are not yet FDA-approved specifically for prediabetes, but the weight-loss data are directly relevant. In STEP-1 (N=1,961), adults with BMI at or above 30 (or at or above 27 with a weight-related comorbidity) received semaglutide 2.4 mg subcutaneously once weekly or placebo. Published in NEJM in 2021, STEP-1 showed:

• Mean weight loss of 14.9% with semaglutide vs. 2.4% with placebo at 68 weeks (P<0.001)
• 84.1% of semaglutide participants achieved at least 5% weight loss
• Fasting glucose and A1c improved significantly in the semaglutide group

In STEP-2 (N=1,210), which enrolled adults with type 2 diabetes, semaglutide 2.4 mg produced 9.6% weight loss vs. 3.4% with placebo. The STEP trials as a group demonstrate that semaglutide at the 2.4 mg dose produces weight loss well above the 7% DPP target in most participants.

For a patient with prediabetes who has failed lifestyle modification and has BMI at or above 27, prescribing semaglutide 2.4 mg for weight management, with the expectation that glycemic improvement will follow, is clinically defensible. The FDA label for Wegovy does not restrict use to patients with established diabetes.

Tirzepatide: Early Prediabetes Data

Tirzepatide (Mounjaro/Zepbound), a dual GIP/GLP-1 receptor agonist, produced even larger weight reductions in SURMOUNT-1 (N=2,539). Published in NEJM in 2022, SURMOUNT-1 showed mean weight loss of 20.9% at the 15 mg dose vs. 3.1% placebo at 72 weeks. A post-hoc analysis of participants with prediabetes at baseline showed that over 95% had reverted to normoglycemia by week 72 in the highest-dose group.

The FDA approved tirzepatide (Zepbound) for chronic weight management in November 2023 for adults with BMI at or above 30, or at or above 27 with a weight-related condition.

Acarbose and Thiazolidinediones: Niche Roles

Two older agents have prediabetes trial data. Acarbose, an alpha-glucosidase inhibitor, reduced T2D progression by 25% relative to placebo in the STOP-NIDDM trial (N=1,429). Published in The Lancet in 2002, STOP-NIDDM showed a statistically significant reduction in incident diabetes, though GI side effects limited adherence. Pioglitazone reduced T2D progression by 72% in the ACT NOW trial (N=602), published in NEJM in 2011, but weight gain and fluid retention limit its routine use.

A Framework for Clinical Decision-Making

When a patient with prediabetes presents for reassessment, the following stepwise approach helps organize the decision about whether treatment has failed and what to do next.

Step 1: Confirm the Intervention Was Adequate

Before calling failure, verify that the patient completed at least 6 months of structured intervention, ideally a CDC-recognized DPP or equivalent. A patient who made casual dietary changes for 3 months has not had a full trial of lifestyle therapy.

Step 2: Assess Current Glycemia Against Defined Thresholds

Pull a current A1c and fasting glucose. If either has crossed the diagnostic threshold for type 2 diabetes, the clinical situation has changed and full diabetes management protocols apply. If still in the prediabetes range but A1c is at or above 6.0% and climbing, that qualifies as high-risk persistence.

Step 3: Stratify by BMI and Comorbidities

Patients with BMI at or above 35 should have pharmacotherapy initiated at first evidence of lifestyle inadequacy rather than waiting for full progression. Patients with prior gestational diabetes are at particularly high lifetime risk. A 2020 systematic review in Diabetes Care found that women with a history of gestational diabetes have a roughly 10-fold increased lifetime risk of developing type 2 diabetes compared to women without that history.

Step 4: Select the Pharmacotherapy

For most patients, metformin at 500 mg once daily (titrated to 850-1,000 mg twice daily over 4 weeks) is the starting point. For patients with BMI at or above 30 and either insurance coverage or affordable access, semaglutide 2.4 mg or tirzepatide deserves serious consideration given the magnitude of glycemic and weight benefits seen in STEP-1 and SURMOUNT-1.

Step 5: Set a Specific Reassessment Date

Recheck A1c and fasting glucose in 3-6 months after starting pharmacotherapy. The ADA recommends A1c monitoring every 3 months when adjusting therapy and every 6-12 months once stable. Document the target: A1c below 5.7% represents reversion to normoglycemia, which is achievable in a meaningful proportion of patients, particularly those who lose at least 10% of body weight.

Monitoring Intervals and When to Escalate

The ADA recommends testing prediabetes patients at least annually. Patients with A1c at the upper end of the prediabetes range (6.2-6.4%) or with multiple risk factors should be retested every 6 months. The 2024 ADA Standards recommend that patients with prediabetes be referred to an intensive behavioral lifestyle intervention program modeled on the DPP.

Escalation to an endocrinologist is appropriate when:

• A1c crosses 6.5% despite metformin and lifestyle optimization
• Fasting glucose is persistently above 120 mg/dL without explanation
• The patient has other endocrine comorbidities (PCOS, Cushing syndrome) driving insulin resistance
• GLP-1 or tirzepatide initiation requires prior authorization support or complex titration

One point worth making directly: prediabetes reversion is possible. A 2022 study in BMJ Open Diabetes Research and Care (pubmed.ncbi.nlm.nih.gov/35273077) found that among adults with prediabetes who lost at least 10% of body weight, approximately 50% reverted to normoglycemia at 2 years. Reversion is the goal, not merely slowing progression.