19IN - Does molecular triage help to identify highly sensitive disease?

Date 30 September 2012
Event ESMO Congress 2012
Session How to integrate new drugs in the current therapeutic landscape of metastatic triple negative breast cancer
Topics Breast Cancer, Metastatic
Pathology/Molecular Biology
Presenter Lajos Pusztai
Authors L. Pusztai
  • Medical Oncology, Yale University, 05620 - New Haven/US


It is helpful to consider predictive biomarkers within three distinct categories that also represent different degrees of difficulty in terms of the predictor discovery process. (1) There are broad treatment response markers that are based on detecting a fundamental biological process that is relevant for the action of a class of drugs. Such markers have a large transcriptional (or other molecular) footprint and hence represent relatively low hanging fruits in the discovery process. For example, high tumor proliferation is such a marker for general chemotherapy sensitivity or for worse baseline prognosis. It may be possible in the future to discover similar broad markers for metabolism-targeted agents or for immunotherapies. (2) There are drug class-specific predictors such as estrogen receptor (ER) expression or human epidermal growth factor 2-receptor (HER2) amplification or activating mutations in critical signaling genes (e.g EGFR, bRAF, c-KIT, etc..) that represent singular driver events in tumors. These alterations reflect partial dependence on a given molecular pathway and can predict sensitivity to a variety of agents that target the involved pathway. Similar driver events for distinct small subsets of cancers are likely to be discovered in the next decade and could yield predictors with clinically useful negative predictive value. (3) Single drug-specific response markers that could be used to select one agent over another have remained elusive despite extensive research efforts. This may be due to the case-to-case heterogeneity of molecular mechanisms that can converge to cause sensitivity or resistance to a particular molecule. Molecular markers under points (2) and (3) and their combinations exist and are useful to select adjuvant therapies in breast cancer (e.g. Ki67, OncotypeDX, MammaPrint, ER, HER2, PAM50) or triage patients to clinical trials who are unlikely to do well with existing treatment modalities. Prospectively designed molecular triaging studies that directly test the positive predictive value of candidate response markers also remain an important research strategy to rapidly assess the clinical utility of predictive markers.


The author has declared no conflicts of interest.