The recent meetings of the American Society of Clinical Oncology confronted cancer specialists with the spectacle of the past and the future of their field. The contrast is stark. We saw a rare tumor – gastrointestinal stromal tumor (GIST) – transformed overnight by the power of modern biology. A relatively nontoxic oral medication, targeting a specific biologic process, turned a gruesomely untreatable cancer into a success story, seemingly in the blink of an eye. The same meetings saw literally hundreds of old-style abstracts, celebrating the mindless combination of relatively inactive drugs in an attempt to eke out incremental gains in incurable cancers. Brain-dead medical oncology, the order of business for a generation of clinical researchers, continued as if nothing new was happening. Why this disconnect in clinical research? It is hard to imagine that many of us still believe that some new combination of toothpaste A plus toothpaste B will conquer advanced (or even not-so-advanced) disease. Similarly, it is difficult to imagine that the intelligent, hard-working, imaginative clinical researchers who populate this field enjoy boredom and failure. It is, rather, that while we can sense the future and see its broad outlines, we cannot quite hold it in our grasp. Literally hundreds of novel therapeutic agents, the products of twin revolutions in combinatorial chemistry and tumor biology, are flooding onto the scene. Which of these agents is worthy of promoting into phase III, proof-of-concept trials? Which targeted therapies should we target? It is all well and good to celebrate STI-571, but it is unlikely that all cancers are as stupid as chronic myeloid leukemia or GIST. Breast cancer, with its infinite and perverse variety and its seemingly endless cornucopia of mutation and progression, is likely to defeat the best efforts of single agents. Barring some global solution (eg, a successful, broadspectrum antiangiogenic therapy or vaccine chemopreventative agent), I suspect that the way forward lies through genomics and proteomics. It is already possible to interrogate several thousand genes using cDNA microarrays, and it is likely that the same will be true for proteins in the near future. One of the more fascinating breast cancer abstracts presented at this year’s meetings suggested that sufficient tissue for a gene chip analysis is obtainable from a fine needle aspiration biopsy. Targeted therapies are likely to work only when we know the target. Tamoxifen does not work in estrogen receptor– negative tumors, nor does trastuzumab work in HER2negative tumors. Yet even the estrogen receptor and HER2 do a relatively poor job of predicting response to hormonal therapy and trastuzumab. Examining families of genes or proteins, rather than individual markers, should bring us closer to the elusive goal of individualizing therapy and should simplify tremendously the development of novel targeted agents, as well as the use of existing agents. Families of gene products, probed with cDNA microarrays, have already been used to separate BRCA-1 and BRCA-2 from sporadic breast cancers; there is every reason to believe that the gene chips should also help us determine therapeutic response. Such approaches may also resurrect older agents. The tragedy of chemotherapy is not that these drugs are toxic; many of the novel targeted therapies have very real side effects. Rather, the tragedy is that the drugs both fail patients and harm them. All therapies are targeted therapies. To believe otherwise is to believe that the universe is random and unexplainable. We just do not understand the targets very well. If we could take an old agent with a 10% response rate in a general population of cancer patients, and turn it into an agent with a 90% response rate in a selected subpopulation, would we be more willing to accept the agent’s toxicity? I suspect so. Dozens of minimally active agents were thrown away in past decades, their careers ended because they were not successful all-purpose agents. Might they not find new purpose if treated like relief pitchers or nickel-back linebackers, specialists brought in for a particular inning or down? The biologic revolution has been thought of as a means of eliminating older chemotherapeutic agents. It is equally likely that it will result in a renaissance for chemotherapeutic agents.