Surrogacy of progression free survival for overall survival in metastatic breast cancer studies: meta-analyses of published studies

Madan G. Kundu; Suddhasatta Acharyya

doi:10.1016/j.cct.2016.12.004

paper

Surrogacy of progression free survival for overall survival in metastatic breast cancer studies: meta-analyses of published studies

arXiv:1608.04300 · doi:10.1016/j.cct.2016.12.004

Abstract

Purpose: PFS is often used as a surrogate endpoint for OS in metastatic breast cancer studies. We have evaluated the association of treatment effect on PFS with significant HR $_{O S}$ (and how this association is affected by other factors) in published prospective metastatic breast cancer studies. Methods: A systematic literature search in PubMed identified prospective metastatic breast cancer studies. Treatments effects on PFS were determined using hazard ratio (HR $_{P F S}$ ), increase in median PFS ( $Δ$ MED $_{P F S}$ ) and % increase in median PFS (% $Δ$ MED $_{P F S}$ ). Diagnostic accuracy of PFS measures (HR $_{P F S}$ , $Δ$ MED $_{P F S}$ and % $Δ$ MED $_{P F S}$ ) in predicting significant HR $_{O S}$ was assessed using receiver operating characteristics (ROC) curves and classification trees approach. Results: Seventy-three cases (i.e., treatment to control comparisons) from 64 individual publications were identified for the analyses. Of these, 16 cases reported significant treatment effect on HR $_{O S}$ at 5% level of significance. Median number of deaths reported in these cases were 156. Area under the ROC curve (AUC) for diagnostic measures as HR $_{P F S}$ , $Δ$ MED $_{P F S}$ and % $Δ$ MED $_{P F S}$ were 0.69, 0.70 and 0.75, respectively. Classification tree results identified % $Δ$ MED $_{P F S}$ and number of deaths as diagnostic measure for significant HR $_{O S}$ . Only 7.9\% (3/39) cases with $Δ$ MED $_{P F S}$ shorter than 48.27\% reported significant HR $_{O S}$ . There were 7 cases with $Δ$ MED $_{P F S}$ of 48.27\% or more and number of deaths reported as 227 or more -- of these 5 cases reported significant HR $_{O S}$ . Conclusion: % $Δ$ MED $_{P F S}$ was found as better diagnostic measure for significant HR $_{O S}$ . Our analysis results also suggest that consideration of total number of deaths may further improve its diagnostic performance.