Demethylating Agents

DNA methylation is a normal mechanism of gene expression regulation, and abnormal DNA methyla-tion patterns that can interfere with the expression of genes used for growth, differentiation, and survival are seen in MDS. It has been known for some time that hypomethylating drugs can promote in vitro cellular differentiation,50 and that this occurs at doses much lower than those needed for maximal cytotoxic effect. The clinical trials using the DNA demethylating agents 5-azacytidine (azacitidine, Vidaza) and 5-aza-2'-deoxy-cytidine began decades ago, prior to a full understanding of their demethylating effects. As a result, they were initially studied as cytidine analogs, using a classic phase I design to determine their maximally tolerated dose in leukemia patients. Subsequent trials in MDS patients used lower, more tolerable doses, with the goal of inducing differentiation and minimizing toxicity, which manifested itself primarily as myelo-suppression in the earlier, high-dose trials.

Low-dose 5-azacytidine was first used to treat MDS by the Cancer and Leukemia Group B (CALGB) in the 1980s. Small studies evaluated 5-azacytidine at a dose of 75 mg/m2/day by either continuous intravenous infusion or subcutaneous injection for 7 days each month and found a 49% or 50% response rate, respectively; both showed a complete remission (CR) rate of 12%.5152 In a subsequent, multicenter, randomized, phase III trial, 191 patients with MDS of all subtypes were randomized to receive either 5-azacytidine given subcutaneously at a dose of 75 mg/m2/day for 7 days per month with supportive care, or supportive care alone.53 Patients on the supportive care arm were permitted to crossover to the 5-azacytidine arm after 4 months for worsening disease or persistent severe cytopenias. The overall response rate (CR + PR) on the 5-azacytidine arm was 23% (7% CR), with a median response duration of 15 months and a median time on treatment of 9.1 months. Hematologic improvement was seen in 37% of the patients. Responses were seen in all MDS subtypes, with a median time to best response of 93 days. Moreover, the median time to AML transformation or death was longer for patients randomized to receive 5-azacytidine (21 months) than those in the supportive care arm (12 months). A nonsignificant difference in median survival was detected (20 months vs. 14 months, p = 0.10), but a 6-month landmark analysis (to eliminate the confounding effect of the 49 crossover patients) detected an improvement in survival (18 months vs. 11 months, p = 0.03). Treatment with 5-azacytidine was also associated with significant improvements in QOL parameters, including fatigue, dyspnea, and psychological well-being.54 5-Azacytidine was approved by the FDA in May 2004 for all MDS subtypes based on this data, but treatment for the RA or RARS patients was approved only for those with moderate to severe neutropenia (ANC <1000/^L), thrombocytopenia (< 50,000/^L), or transfusion-dependent anemia.

The FDA approval of 5-azacytidine represents the first agent approved for treating patients with MDSs. The CALGB study of 5-azacytidine took many years to complete and report, and a confirmatory phase III study is underway comparing 5-azacytidine to several conventional therapies (supportive care, low-dose cytarabine, or intensive, AML-like therapies). The recent approval of 5-azacytidine in the United States suggests that this trial will primarily enroll patients in Europe. As the IPSS was not devised when the first phase III 5-azacytidine trial was started, this trial may better define which MDS patients have the highest likelihood of responding to 5-azacytidine.

Three large phase II studies of low-dose 5-aza-2'-deoxycitidine (decitabine) have been conducted in Europe, and the results recently summarized. Of 169

MDS patients treated with intravenous doses of 45-50 mg/m2/day for 3 days every 6 weeks (either 50 mg/m2/day by continuous i.v. infusion over 3 days, or 15 mg/m2 every 8 h for 3 days for a total of 9 doses), 20% had a CR, 10% had a partial remission (PR), and the overall response rate was 49% (including hematologic improvement).55 An early (after one cycle) and often dramatic rise in the platelet count was frequently seen, and was predictive of a subsequent tri-lineage response.56 On the basis of these positive results, a multicenter, phase III study was opened in the United States and Canada, which completed accrual in 2003. A total of 160 patients were randomized to receive decitabine 15 mg/m2 every 8 h over 3 days (total 45 mg/m2/day) plus supportive care, or supportive care alone, with no crossover allowed. Data from this trial was reported at the 2004 annual American Society of Hematology meeting with benefit seen in the IPSS high-risk and Int-2 risk patients, but not the Int-1 risk patients.57 Among the key questions are whether decitabine and 5-azacytidine work equally well in the same types of patients, or whether some MDS patients would benefit more from one versus the other drug. In the absence of FDA approval of decitabine, however, such questions are not yet clinically relevant. An oral formulation of decitabine is in preclinical trials, and new hypomethylating agents (e.g., zebularine)58 may also make their way into clinical trials.

These agents appear to induce hypomethylation by irreversibly binding and "trapping" DNA methyl-transferases so that the next round of DNA replication produces a demethylated base at a site that was previously methylated. Although it is clear that these agents have clinical activity in MDS, demethylation of specific genes (e.g., the CDK inhibitors p15, p16, p21) has not been definitively correlated with clinical responses. Therefore, the specific mechanism by which hypomethylation induces responses in MDS remains unclear. Identifying biological correlates of clinical responses will help define the mechanism of action of these agents, and will identify predictors of hematologic responses in this heterogeneous disease. A common side effect of low-dose decitabine and 5-azacytidine is some degree of myelosuppres-sion, which suggests that even at lower does, cytotox-icity may be important for the clinical responses seen in MDS.

Future studies with the demethylating agents 5-aza-cytidine and decitabine will hopefully improve our understanding of how these drugs improve the cytope-nias and bone marrow function in MDS. These agents will also be combined with other therapies in an attempt to improve the CR rate and duration, which may include maintenance therapy. Attractive combinations include the use of hypomethylating agents with other drugs that also have an impact on gene regulation, such as all-trans retinoic acid (ATRA), vitamin D analogs, or histone deacetylase inhibitors such as SAHA, depsipeptide, phenylbutyrate, or valproic acid. Mouse models are finally being developed for MDS (e.g., see Ref. 4) and they may prove useful for such preclinical studies.

0 0

Post a comment