Cancer-drug discovery--let's get ready for the next period.

  • Dan L. Longo
  • Published 2014 in The New England journal of medicine


The use of drugs to treat cancer began about 70 years ago. After World War II, Louis Goodman and Alfred Gilman and their colleagues at Yale University noted that people who had been exposed to mustard gas often had bone marrow suppression. They identified the active chemical moiety in the gas, gave it to patients, and saw responses in patients with hematologic cancers.1 Thus, alkylating agents were developed and became key components of treatments for many forms of cancer. Not all useful drugs emerged as a product of accurate hypotheses. Sidney Farber mistakenly thought that leukemic cells looked megaloblastic, and so he gave folate to some children with acute leukemia. Their disease accelerated, and the children died. This led to the hypothesis that an antifolate agent might be effective, and aminopterin and amethopterin (methotrexate) were born.2 The most remarkable progress in the past two decades has come from a shift in focus from screening compounds for growth inhibition of tumor cell lines in culture in a mechanismagnostic fashion3 to identifying novel therapeutic targets through genetic and cell biology studies of cancers. The prototype of this class of agents is imatinib, an inhibitor of the BCR-ABL oncogene product that is formed by the t(9;22) translocation and that is central to the pathogenesis of chronic myeloid leukemia (CML).4 The efficacy of imatinib in patients with CML is high, but the agent is not curative (high-dose chemotherapy plus hematopoietic stem-cell transplantation is the only known curative therapy for CML). Patients take imatinib for years, for as long as it continues to work. CML is an unusual cancer in that, for the most part, a single factor, the BCRABL chimeric gene product, drives the tumor. The success of imatinib in patients with CML has given rise to a new strategy in cancer treatment: find the driver and disable it. In most solid tumors, multiple genes (10 to >100) are mutated,5 different cells in the tumor have distinct genetic lesions,6 and it is not always clear that a particular mutation is a critical one. Molecular analysis has led to the notion that many tumors that look alike under the microscope are not alike. The new nosology of tumors has identified multiple genetic varieties of tumors that used to carry the same diagnosis. Treating patients whose tumors have activating mutations in the epidermal growth factor (EGF) receptor with EGF-receptor blockers and tumors with activating mutations in BRAF with BRAF blockers, and so on, has produced some dramatic transient responses but no cures.7,8 After a few months, the tumor seems to adapt to the blockade and find a way around it by amplifying the key gene, activating an alternative pathway, mutating the target so that the drug no longer blocks its function, or pumping the drug out before it can do damage — or by other mechanisms that are being elucidated through a study of sequential biopsy specimens of tumors. In this issue of the Journal, Solomon and colleagues9 report the superiority of the ALK inhibitor crizotinib over chemotherapy as the primary treatment in the subgroup of patients with lung cancer whose tumors bear activating translocations involving the ALK gene. Responses to crizotinib were observed in 74% of the patients, and the median duration of response was nearly 11 months (as compared with a 45% response rate and a 7-month median duration of response with chemotherapy). Patients who were treated with crizotinib had improvements in quality-of-life measures as well. Unanswered questions include the following: Why did 26% of the patients who received crizotinib not have a response? Why did the drug stop working in those who had initially had a response? Are there combinations of agents that might convert good partial responses to durable complete responses? Would crizotinib be effective against tumors originating in other organs that also had ALK activation? Crizotinib joins the growing list of kinase inhibitors that are active in particular (sometimes very small) subgroups of patients. Many of these agents are producing tumor regressions in cancers for which no other therapy has shown even minimal activity. In the case of the ALK inhibitors, even more effective drugs are in the pipeline that produce even longer responses (median,

DOI: 10.1056/NEJMe1412624

Cite this paper

@article{Longo2014CancerdrugDG, title={Cancer-drug discovery--let's get ready for the next period.}, author={Dan L. Longo}, journal={The New England journal of medicine}, year={2014}, volume={371 23}, pages={2227-8} }