Impact of Exploratory Biomarkers on the Treatment Effects of Avastin (bevacizumab) in Metastatic Breast Cancer
Jubb AM, Miller KD, Rugo HS, et al
Clin Cancer Res. 2011;17:372-381
Summary
Results of phase 3 trials evaluating the vascular endothelial growth factor (VEGF) receptor inhibitor bevacizumab in pretreated metastatic breast cancer have indicated that certain patient populations are less likely to benefit from the drug. However, there are no known clinically relevant biomarkers that could identify breast cancers highly likely or highly unlikely to respond to bevacizumab.
Investigators conducted a retrospective analysis of patients treated on a phase 3 randomized trial that compared single-agent capecitabine vs capecitabine plus bevacizumab in women with previously treated metastatic breast cancer. Formalin-fixed, paraffin-embedded tumor tissue was available for analysis in 223 of the 462 patients who were treated in this study. Depending on the specific biomarker in question, investigators used either in situ hybridization or immunohistochemistry to analyze the relationships between the marker's expression and the subsequent progression-free survival (PFS) when treated with or without bevacizumab.
A significant difference in outcome was found in patients treated with capecitabine plus bevacizumab in the presence of low expression of delta-like ligand (P = .01), and a borderline difference was observed in patients with low expression of the VEGF-C component (P =.05) and neurophilin-1 (P =.07). Of note, these differences were no longer "significant" after performing a statistical correction for the fact that multiple hypotheses were tested in this retrospective analysis.
Viewpoint
There has been heighted interest in the discovery of a well-validated biomarker to predict for the clinical utility of bevacizumab in cancer treatment in general and in breast cancer management in particular. Although rather extensive evidence from multiple randomized trials confirmed the "statistically significant" benefits in either PFS or overall survival resulting from the administration of this agent across multiple tumor types, the overall degree of impact associated with this extremely expensive and not nontoxic agent is quite modest. The existence of a reliable biomarker that could assist clinicians in their decision to use this agent would be a major advance in standard cancer management.
The most prominent finding in the current study is that metastatic breast cancer patients with tumors demonstrating low expression of delta-like ligand who are treated with capecitabine plus bevacizumab experience a rather pronounced degree of clinical benefit, with a median PFS improvement of 4.5 months. This is particularly worthy of further exploration in a future prospective clinical trial. Retrospective analyses of previously conducted prospective therapeutic clinical trials, as was done here, are an essential initial step, but prospective trials are still needed to identify potentially promising tissue- or serum-based biomarkers for novel (and expensive) antineoplastic agents that can provide genuine clinical utility in routine cancer management.
http://clincancerres.aacrjournals.or.../17/2/372.long