Early BCa: meta-analysis backs DFS endpoint for trastuzumab

  • Lancet Oncol

  • curated by Miriam Davis, PhD
  • Univadis Clinical Summaries
Access to the full content of this site is available only to registered healthcare professionals. Access to the full content of this site is available only to registered healthcare professionals.

Takeaway

  • A meta-analysis finds that DFS is strongly or moderately strongly correlated with OS in clinical trials of adjuvant trastuzumab given up to 1 year in patients with human epidermal growth factor receptor 2 (HER2)-positive early breast cancer (eBCa).

Why this matters

  • This is the first meta-analysis to validate DFS as a surrogate for OS in eBCa.
  • Using DFS as a surrogate should expedite drug development.

Study design

  • Meta-analysis of 8 clinical trials (n=21,480), after search of bibliographic databases (eg, Medline), clinical trial registries, and trial registries from pharmaceutical companies.
  • DFS refers to any type of recurrence (invasive or noninvasive) or death from any cause.
  • Patient-level (rs) and trial-level (R2) correlations were calculated between DFS and OS.
  • The surrogate threshold effect was the maximum HR for DFS that statistically predicts an HR for OS less than 1.00 in a future trial.
  • Funding: Roche Pharma AG.

Key results

  • Patient-level associations between DFS and OS were strong (rs=0.90; 95% CI, 0.89-0.90).
  • Trial-level associations between DFS and OS were moderate to strong (R2, 0.75; 95% CI, 0.50-1.00) in 1 model covering all 8 trials, or strong (R2, 0.84; 95% CI, 0.67-1.00) in a second model covering fewer trials.
  • Surrogate threshold effects ranged from 0.56 to 0.81, based on all 8 trials.

Limitations

  • Regression analyses are sensitive to outliers.

Please confirm your acceptance

To gain full access to GPnotebook please confirm:

By submitting here you confirm that you have accepted Terms of Use and Privacy Policy of GPnotebook.

Submit