As we enter the era of molecularly targeted therapy, drug trials will take on a new face. Specifically, if a drug is designed to perform a specific biochemical action and that action should result in other specific effects, these effects can be objectively monitored in tumor tissue before and after such therapy as tumor response criteria. In this way, we will immediately know if the desired effects were achieved. If these biochemical effects are achieved and the tumor does not respond by standard criteria, then the trial is essentially over. If the biochemical effects are not achieved, then pharmacologic measures or other agents can be altered to achieve the desired effects. In the current issue of the Journal of Clinical Oncology, Chan et al have reported an example of such an approach. This study is the first to demonstrate that treatment with azacitidine can produce demethylation of specific DNA sites in tumors. The investigators examined the methylation status of Epstein-Barr viral (EBV) promoters preand post-treatment with azacitidine in five patients with EBV-associated tumors (nasopharyngeal carcinoma and lymphoma), and demonstrated substantial demethylation in the target tissue of all examined viral promoters. These viral promoters regulate the expression of viral proteins which engender strong cytotoxic T cell (CD8 ) responses in seropositive individuals. The viral promoter regions are known to be methylated in tumors, resulting in a lack of viral gene expression and in turn allowing the tumor to escape from immune recognition. However, despite the changes in tumor cell viral promoter DNA methylation, Chan et al found no significant changes in tumor expression of viral antigens, and no clinical responses were seen. In an analogous situation, tumor suppressor genes have undergone tumor-acquired promoter methylation in a variety of solid tumors leading to their loss of expression and function in tumors, resulting in the malignant phenotype. In fact, a large variety of tumors have multiple genes methylated, associated with a loss of gene expression. Thus, the Chan study is a model for future trials in solid tumors in which the methylation status and re-expression of tumor suppressor genes are monitored. In preparation for this, many preclinical studies in the past 3 years have shown that in in vitro tissue culture, human tumor cells treated with azacitidine or decitabine will show demethylation and subsequent re-expression of a variety of key tumor suppressor genes and other genes whose loss of expression is involved in cancer pathogenesis. Azacitidine (5-azacitidine) and decitabine (5-aza-2 deoxycytidine) were originally synthesized in 1964 by Piskala et al and Pliml et al, and are the two most extensively studied agents that prevent DNA methylation. Despite this long history, the optimal role for these agents in the treatment of malignancy remains to be defined. Numerous clinical trials have evaluated the efficacy of azacitidine in hematologic malignancies. As a single agent, it has demonstrated activity in both pediatric and adult relapsed or refractory acute myelogenous leukemia (AML).In combination chemotherapy, azacytidine has been effectively used as first-line therapy in AML. However, no adequately powered phase III studies have defined the utility of azacytidine in the treatment of AML. Several phase II trials have also demonstrated that singleagent azacitidine has activity in myelodysplastic syndrome (MDS). Recently, the Cancer and Leukemia Group B reported the results of a phase III comparison of azacytidine and best supportive care in MDS. Azacitidine resulted in significantly superior response rates (60% v 5%), quality of life, and overall median survival (18 months v 11 months). A number of small phase II trials of single-agent azacitidine utilizing a variety of doses and schedules have also been performed in solid tumors, including breast, gastroinJOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 22 NUMBER 8 APRIL 15 2004
No takes yet. Share an insight, caveat, or question.
Dowell et al. (2004) studied this question.