A testing procedure for survival data with few responders

Opportunities posed by novel patient selection biomarker approaches in oncology drug development: going beyond the cytotoxics

An area of unmet medical need in clinical oncology has been optimizing patient selection for a given therapeutic with the goal of getting the right drug to the right patient. Recent studies have developed preclinical approaches to identifying molecular ‘signatures of resistance’ for cytotoxic therapies and prospective validation of this strategy is ongoing in the clinic. New challenges in this setting include identifying approaches to patient selection for cytostatic compounds such as signaling pathway inhibitors and stem cell targets. Here, we discuss the biomarker methodologies developed using traditional cytotoxic drugs and how these approaches can be adapted to identify biomarkers of patient selection for novel signaling inhibitors and other novel targets. It has become increasingly clear that such biomarker discovery and validation needs to begin early and continue throughout the drug development process.

Adaptive signature design: an adaptive clinical trial design for generating and prospectively testing a gene expression signature for sensitive patients

PURPOSE

A new generation of molecularly targeted agents is entering the definitive stage of clinical evaluation. Many of these drugs benefit only a subset of treated patients and may be overlooked by the traditional, broad-eligibility approach to randomized clinical trials. Thus, there is a need for development of novel statistical methodology for rapid evaluation of these agents.

EXPERIMENTAL DESIGN

We propose a new adaptive design for randomized clinical trials of targeted agents in settings where an assay or signature that identifies sensitive patients is not available at the outset of the study. The design combines prospective development of a gene expression-based classifier to select sensitive patients with a properly powered test for overall effect.

RESULTS

Performance of the adaptive design, relative to the more traditional design, is evaluated in a simulation study. It is shown that when the proportion of patients sensitive to the new drug is low, the adaptive design substantially reduces the chance of false rejection of effective new treatments. When the new treatment is broadly effective, the adaptive design has power to detect the overall effect similar to the traditional design. Formulas are provided to determine the situations in which the new design is advantageous.

CONCLUSION

Development of a gene expression-based classifier to identify the subset of sensitive patients can be prospectively incorporated into a randomized phase III design without compromising the ability to detect an overall effect.