What hazard ratio (HR) and 95% confidence interval (CI) supports the conclusion that albiglutide is noninferior in efficacy to insulin lispro for type 2 diabetes mellitus treatment?

Hazard Ratio and Confidence Interval Supporting Albiglutide's Noninferiority to Insulin Lispro

The hazard ratio of 0.16 with 95% confidence interval of -0.2 to 0.41 best supports the conclusion that albiglutide is noninferior to insulin lispro for the treatment of type 2 diabetes mellitus.

Understanding Noninferiority in This Context

When evaluating noninferiority in a clinical trial comparing albiglutide with insulin lispro, we need to understand:

• The primary outcome was change from baseline in A1c at week 26
• The prespecified noninferiority margin was 0.4
• For noninferiority to be established, the upper bound of the 95% confidence interval must not exceed the noninferiority margin

Analysis of the Evidence

Based on the FDA label for albiglutide (Tanzeum), a clinical trial was conducted comparing albiglutide with insulin lispro as add-on therapy in patients inadequately controlled on insulin glargine 1. The results showed:

• The between-treatment difference was -0.2% with 95% confidence interval (-0.32%, 0.00%)
• This met the pre-specified non-inferiority margin of 0.4%

Additionally, in the Harmony 6 trial specifically examining this comparison:

• HbA1c decreased by -0.82 ± 0.06% with albiglutide and -0.66 ± 0.06% with lispro 2
• Treatment difference was -0.16% (95% CI -0.32 to 0.00)
• This met the noninferiority endpoint with a margin of 0.4%

Interpreting the Hazard Ratio and Confidence Interval

Looking at the options provided in the question:

1. HR -0.16 (95% CI, -0.2 to 0.41) - The upper bound (0.41) exceeds the noninferiority margin (0.4)
2. HR 0.16 (95% CI, 0 to 0.42) - The upper bound (0.42) exceeds the noninferiority margin (0.4)
3. HR 0.16 (95% CI, -0.1 to 0.5) - The upper bound (0.5) exceeds the noninferiority margin (0.4)
4. HR -0.16 (95% CI, -0.32 to 0) - The upper bound (0) is well below the noninferiority margin (0.4)

For noninferiority to be established, the upper bound of the confidence interval must not exceed the prespecified margin of 0.4. Therefore, option 4 (HR -0.16 with 95% CI of -0.32 to 0) is the only one that conclusively supports noninferiority.

Clinical Implications

This noninferiority finding has important clinical implications:

• Albiglutide provides comparable glycemic control to insulin lispro
• Albiglutide demonstrated weight loss (-0.73 kg) compared to weight gain with insulin lispro (+0.81 kg) 2
• Lower rates of documented symptomatic hypoglycemia with albiglutide (15.8% vs 29.9% with lispro) 2
• Once-weekly dosing of albiglutide offers a simpler therapeutic option than thrice-daily insulin lispro

Common Pitfalls in Interpreting Noninferiority Trials

• Confusing the direction of the effect - In this case, a negative value for the treatment difference (albiglutide minus lispro) indicates albiglutide performed better
• Misinterpreting the confidence interval - The entire confidence interval must be examined, not just the point estimate
• Forgetting to compare the upper bound to the noninferiority margin - This is the critical comparison for establishing noninferiority

In conclusion, the hazard ratio of -0.16 with 95% CI of -0.32 to 0 provides the strongest evidence supporting albiglutide's noninferiority to insulin lispro, as it's the only option where the upper bound of the confidence interval remains below the prespecified noninferiority margin of 0.4.