Treatment for advanced breast cancer is haunted by the twin specters of drug failure and ruinous toxicity. Both problems have a similar root: the inability of modern medicine to match the right patient to the right drug. Individual drugs for advanced breast cancer, particularly chemotherapy, routinely fail to benefit the majority of women treated. As a result, women with advanced breast cancer are faced with progressively less active, progressively more toxic therapy. The tragedy of modern therapy is not just its toxicity, it is that so many experience so much toxicity for so little benefit.
Many hope that toxic therapies will disappear as new "targeted" agents are developed. We consider this a false hope, at least for the near future. So-called "targeted" therapies have toxicities that were initially unexpected; many of these agents are insufficiently targeted due to a poor understanding of how they work; and virtually all have entered the treatment of advanced breast cancer in combination with older agents. What is more, older agents genuinely benefit some patients with breast cancer, and often magnify the benefits of newer drugs (e.g., the combination of trastuzumab and chemotherapy).
Therapeutic individualization -- matching the right patient to the right drug -- has been a long-term but elusive goal. In this proposal, we argue that the time is ripe for an attempt to improve therapeutic individualization in breast cancer. This hope emanates from rapid improvements in three separate technologies. Genomics, which allows us to interrogate the criminal DNA of the breast cancer cell, has now been shown capable of predicting response (at least in part) to some chemotherapy agents. Proteomics, which performs a similar task with cancer cell proteins, potentially offers even greater possibilities to examine the workings of the cancer cell and its response to therapeutic agents. Finally, pharmacogenetics, the study of inherited differences in drug effects, adds a dimension to treatment selection not obtained through genomics or proteomics. Because it is unlikely, given what we already know about the biology of breast cancer, that any one approach will identify all responding patients, we propose a combined analysis using these three new technologies in combination with existing technology.
Our proposal starts and ends in the clinic. We will treat patients with advanced breast cancer with both older and newer agents, asking patients to volunteer a piece of cancer tissue and blood for analysis. The tissue will be processed at a central facility, and then examined by an impressive array of experts in genomics, proteomics, and pharmacogenetics/pharmacogenomics. We will then glean information obtained from these studies, combining the data to create an entirely novel treatment algorithm -- in essence, developing a recipe for matching the right drug to the right woman. Because it is not enough to assume that we "got it right the first time," we will then apply this treatment algorithm to another group of women receiving therapy for advanced breast cancer, attempting to show in a rigorous fashion that our "treatment recipe" would indeed lead us to the correct treatment decisions.
This is a first pass at an important problem, and as such, it is important to remain modest about what actually will be achieved. What might be achieved, however, is sufficiently important that we consider this an exciting avenue of research. If successful, this approach could benefit any patient receiving a standard chemotherapy agent for advanced breast cancer. Because the same approach might work for new agents as well, it could rapidly accelerate the introduction of these new drugs to the right patients. And finally, because this proposal both begins and ends in the clinic, it is realistic to believe that real patient benefits could become available with the completion of this work. The technology that we are examining is evolving at a stunning pace; what was unimaginable 5 years ago could become routine in 5-10 years. But it will only happen if we take an approach that examines specific agents in their correct clinical context using the best of modern technology.