| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Letters to the Editor |
University of Washington Seattle, Washington
Noonday Asset Management Charlotte, NC
Massachusetts General Hospital, Boston, MA
In Response: We appreciate the opportunity to discuss the pertinent issues raised by Banerji et al. in their letter. The application of biomarkers in clinical cancer trials, and particularly in phase I trials, remains controversial. There is no doubt, however, that the role of biomarkers in early clinical trials requires refinement as the era of molecularly targeted drug development matures.
We agree that publication bias against phase I trials with negative pharmacodynamic biomarker studies may prevent an accurate measurement of how biomarkers affect the "go-no go" decision. Our review did not include unpublished trials, and therefore, we cannot estimate precisely what proportion of new drugs entering phase I studies will fail to make it to phase II trial based on a negative biomarker result. Nevertheless, our analysis did not show a single published phase I trial in which the studied compound had its development halted because of lack of target modulation evidenced by negative biomarker studies. The demonstration of target blockade by pharmacodynamic biomarkers that are validated as predictive or prognostic factors of clinical outcomes can potentially affect the decision-making process to continue or not with the development of a drug. However, current evidence shows that only very few biomarkers used in phase I trials of molecularly targeted agents were clearly validated as surrogates of important clinical end points such as survival or time to tumor progression (1–3). Information about target blockade generated by molecular biomarker studies is often difficult, if not impossible, to interpret when correlation between such biomarkers and clinical outcomes is lacking. Biomarker data about intracellular target modulation should also be interpreted with caution. Intracellular biochemical pathways that modulate cancer cell survival, differentiation, metastasis, angiogenesis, and drug resistance are highly complex, heterogeneous, and likely redundant. One might argue that it is unjustified, at this time, to assume that biomarker confirmation of target blockage of one step of a single intracellular pathway will reliably predict for effective reduction in tumor burden or improvement in clinical outcomes (2). This relationship between biomarker findings and efficacy needs to be established for each new target and pathway.
We are hopeful that broader application of biomarkers will ultimately expedite drug development and boast overall success rates. Based on our review of existing data, however, we must conclude that enthusiasm to use biomarker studies in early cancer trials should be tempered by the robustness of data supporting their validity, their incremental costs, and the risks and burden that such studies may impose on patients enrolled in clinical trials, especially the ones that require invasive studies such as tumor biopsies.
Disclosure of Potential Conflicts of Interest
B. Chabner is employed by Kosan Biosciences, Cell Genesys, Pharma Mar, Oncotech, Glead, Glaxo Smith Kline, Aegera, and Gloucester Pharma and has stock options with SAB.
References
Related Article
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
