Evolution of Decitabine Development

Abstract

Decitabine (5-aza-2′-deoxycytidine) is a hypomethylating agent with a dual mechanism of action: reactivation of silenced genes and differentiation at low doses, and cytotoxicity at high doses. The original studies in the 1980s used decitabine as a classical anticancer drug, at its maximum clinically tolerated dose, 1500 to 2500 mg/m² per course. At these doses, decitabine was found to be active in leukemia, but was associated with delayed and prolonged myelosuppression. After a better understanding of epigenetics in cancer and the role of decitabine in epigenetic (hypomethylating) therapy was gained, it was reevaluated at approximately 1/20th of the previous doses (ie, at ‘optimal biologic’ doses that modulate hypomethylation). In these dose schedules of decitabine (100 to 150 mg/m² per course), the drug was found to be active with manageable side effects in patients with myelodysplastic syndromes (MDS) and other myeloid tumors. Optimizing dosing schedules of decitabine to maximize hypomethylation (low dose, high dose intensity, and multiple cycles) have further improved results, suggesting that decitabine is an active therapy that alters the natural course of MDS. Combination therapies that augment the epigenetic effect of decitabine will likely improve responses and extend its use for the treatment of other malignancies.

The development of new therapeutic strategies for myelodysplastic syndrome (MDS) has been the result of extensive understanding of the pathobiology of the disease. Therapeutics targeting chromatin structure, angiogenesis, and the bone marrow microenvironment that nurtures the MDS phenotype have demonstrated significant activity and offered an opportunity to alter the natural history of the disease.¹ Chromatin remodeling is a powerful mechanism of regulating gene expression and protein function.² In extreme states, chromatin remodeling can permanently repress expression of a gene, a situation termed ‘epigenetic silencing.’ Such silencing is exploited by cancers to fully express the malignant phenotype.³ Evidence supporting a role of epigenetic gene silencing in tumorigenesis stems from studies revealing a large number of genes that are silenced by aberrant DNA methylation in different types of cancers, many of which are involved in the control of cell cycle progression, apoptosis, tissue invasion, and genomic stability.

DNA methylation is remarkably altered in most malignancies, with concomitant global hypomethylation and localized hypermethylation.⁴ This increased methylation affects the regulatory region (gene promoter) located in CpG islands and suppresses gene expression permanently, thereby providing cancers with an alternative to mutations or deletions for inactivation of tumor suppressor genes and other critical genes. Leukemias and MDS are characterized by the hypermethylation and silencing of multiple genes.^5,6 Several genes, including the cyclin-dependent kinase inhibitor p15, have aberrant methylation and are associated with resistance to chemotherapy.^7–9

The 2 cytosine analogs azacitidine (5-azacytidine) and decitabine (5-aza-2′-deoxycytidine) are methyltransferase inhibitors that exhibited encouraging in vitro antileukemic activity.⁵ Decitabine is phosphorylated to decitabine triphosphate, which incorporates into DNA, depletes DNA methyltransferase, and induces replication-dependent DNA hypomethylation.^10,11 At high doses decitabine produces DNA adducts that results in DNA synthesis arrest and cytotoxicity.^10,11 At low doses it induces gene expression profile changes that favor differentiation, reduced proliferation, and/or increased apoptosis.^12–15 This dual antitumor activity has generated significant interest in investigating hypomethylating agents as antineoplastic drugs and biologic response modifiers. This review details the role of decitabine in the treatment of MDS and other hematologic malignancies.

History of Decitabine

Decitabine (Fig. 1) was first synthesized in 1964 and its potential antileukemic activity was reported in 1968.¹⁶ Interest in decitabine was enhanced by pre-clinical studies indicating that decitabine is a more potent antileukemic agent in mice than cytosine arabinoside,¹¹ and by the report of Jones and Taylor¹⁰ that it could induce terminal differentiation of a murine embryonic cell line. Momparler et al.¹⁷ and Rivard et al.¹⁸ initiated what to our knowledge were the first clinical trials of decitabine in acute leukemia. The original studies were classic phase 1 trials that identified the maximum tolerated dose (MTD) as 1500 to 2250 mg/m² per course.^17,18 The dose-limiting toxicity (DLT) was prolonged myelosuppression. Further clinical studies were disappointing in solid tumors but more promising in acute myelogenous leukemia (AML), MDS, and chronic myelogenous leukemia (CML).¹⁹ The regimens tested involved high doses of decitabine given for 1 to 7 days per course. In patients with higher-risk MDS, a low-dose schedule (15 mg/m² every 8 hours for 3 days = 135 mg/m² per course) demonstrated encouraging activity.²⁰ The University of Texas M. D. Anderson Cancer Center leukemia group introduced decitabine into leukemia trials in the U.S. in 1992. After a better understanding of its hypomethylating effects, and optimization of dose schedules, a pivotal phase 3 study was designed and initiated in the year 2000 to compare low-dose decitabine versus supportive care in patients with MDS.²¹ Based on these data, decitabine (Dacogen; SuperGen, Dublin, Calif) received approval from the U.S. Food and Drug Administration for the treatment of MDS and chronic myelomonocytic leukemia (CMML) in May 2006.

FIGURE 1.

Chemical structure of decitabine.

Preclinical Studies With Decitabine

DNA hypermethylation and gene silencing

Aberrant DNA methylation is present in many malignancies, with concomitant global hypomethylation and localized hypermethylation.⁴ Increased methylation affects regulatory regions located in CpG islands and suppresses gene expression. This provides cancers with an alternative to mutations or deletions for inactivation of tumor suppressor genes and other critical genes. Leukemias and MDS are characterized by hypermethylation and silencing of multiple genes.^5,6 This process can occur early in the disease course and is also associated with disease progression. The cyclin-dependent kinase inhibitor p15^INK4b was described as a frequent target of aberrant methylation in MDS, and its inactivation was associated with an increased risk of progression to AML.⁷ Aberrant methylation of other similar genes was also described and was associated with resistance to therapy and tumor progression.^8,9,22

Dual modes of action of decitabine

The antineoplastic action of decitabine results from its incorporation into newly synthesized DNA. It is an S-phase-specific agent¹¹ and has a dual, dose-dependent mechanism of action. At high doses its cytotoxic activity is due to covalent trapping the enzyme DNA methyltransferase into DNA.^10,11 At lower doses its antitumor effect is likely due to its ability to inhibit DNA hypermethylation and to reactivate tumor suppressor genes.^10,11 At low doses decitabine does not block cell cycle progression of G₁ phase cells into the S-phase. Low doses of decitabine incorporated into DNA lead to trapping and depletion of the enzyme DNA methyltransferase. In clonogenic assays, a 1-hour exposure to a 10 μM concentration of decitabine produced loss of clonogenicity in the same range of cells in S-phase (30–50%). A longer exposure time of 24 hours demonstrated a markedly greater antineoplastic activity, with >95% loss of clonogenicity with a decitabine dose of 1 μM.²³ The hypomethylation induced by decitabine resulted in reexpression of tumor suppressor genes, induction of cellular differentiation, and suppression of tumor growth.^10,24

Hypomethylation and histone acetylation

Decitabine is phosphorylated and incorporated into DNA. It then covalently binds to DNA methyltransferases and traps the enzyme to DNA, acting as an irreversible inhibitor of its enzymatic activity. Consequently, decitabine induces marked DNA hypomethylation in vitro and in vivo,¹⁵ and restores silenced gene expression. Its detailed molecular mechanism is still being deciphered. There appears to be a cascade of biochemical events triggered by promoter DNA methylation that involve initial DNA binding proteins, which attract histone deacetylases and histone methylases, and eventually modify histones into a silenced chromatin state.^25–27 A feedback loop is operational between DNA methylation and histone methylation, whereby each of these biochemical modifications at a given gene trigger the other, thus creating a self-reinforcing silencing loop.^25,27 This silencing loop is interrupted by decitabine. Decitabine induces hypomethylation and reverses the silenced histone code at tumor suppressor gene loci.^17,18,22 This dual effect (hypomethylation/histone changes) may explain the superior effect of decitabine on gene expression compared with histone deacetylase inhibitors.²⁸ Decitabine also has significant effects on the expression of genes not silenced by CpG island methylation.

Decitabine induces the expression of p21, a gene that has demonstrated no DNA hypermethylation in cancer.^20,29 The effects of decitabine on the histone code are not limited to genes showing silencing by promoter-associated methylation.²² In bladder cancer cells, decitabine induced rapid and substantial remodeling of the heterochromatic domains of the p14ARF/p16INK4a locus, reducing levels of dimethylated H3-K9 and increasing levels of dimethylated H3-K4. It also increased acetylation and H3-K4 methylation at the unmethylated p14 promoter, suggesting it can induce chromatin remodeling independently of its effects on cytosine methylation.²⁷ This silencing-independent activity of decitabine is not well understood. Other changes may be reactive, due to the stress of exposing cells to a cytotoxic agent. Thus, the ultimate antineoplastic mechanism of decitabine may be very pleiotropic.

In vivo molecular effects of decitabine

Global hypomethylation after decitabine therapy in vivo was observed in early trials,³⁰ and was studied in more detail in recent studies in leukemia.^15,31 Hypomethylation after decitabine was dose-dependent, peaked 10 to 15 days after the initiation of therapy, and recovered to baseline at 4 to 6 weeks. In the AML cell line OCI-AML2, decitabine induced the expression of 81 of 22,000 genes; 96 genes were down-regulated (×2-fold change in expression).³² Similar results were obtained with primary AML and MDS cells after treatment with decitabine ex vivo and in vivo, respectively. In contrast, significantly fewer changes in gene expression and cytotoxicity were detected in normal peripheral blood mononuclear and bone marrow cells or transformed epithelial cells treated with decitabine.³² Hypomethylation demonstrated an inconsistent association with response, with a positive correlation in AML, but an inverse correlation in CML. The latter inverse correlation may be due to the death of responsive-hypomethylated cells and a shift to resistant cells that withstand more hypomethylation.

Daskalakis et al.³³ documented p15 demethylation in marrow DNA samples in 9 of 12 patients with MDS treated with decitabine, and evidence of p15 gene reactivation in 4 responding patients with low baseline expression. They also noted gene reactivation in morphologically dysplastic cells in patients not in complete remission at the time of study, demonstrating the in vivo effect of the drug. In a recent study, p15 hypomethylation after decitabine was observed, but there was no correlation between baseline p15 methylation or after therapy and response.³¹

Pharmacokinetics

After intravenous decitabine administration, plasma protein binding is negligible (<1%) and has an excellent distribution in body fluids. Due to the nucleoside transport system there is rapid equilibration of decitabine between extracellular and intracellular compartments.^19,34 The plasma half-life in humans is approximately 35 minutes due to rapid deamination by high levels of cytidine deaminase.¹⁸ Decitabine crosses the blood-brain barrier, achieving 27% to 58% of the plasma concentrations after continuous infusion.³⁵ The repeated administration of decitabine in patients with advanced MDS as a 3-hour infusion of 15 mg/m² every 8 hours for 3 days does not result in systemic accumulation of the drug, and pharmacokinetic remains unchanged from cycle to cycle.^36,37 When decitabine was given as a 1-hour infusion daily for 10 days every 28 days, the plasma drug concentration-time on Days 1 and 10 achieved a mean Cmax of 93 ng/mL and following a 2-compartment infusion model. The mean short and long half-lives were 2.7 minutes and 36.9 minutes, respectively, with a trend of decreasing the longer half-life on Day 10.³⁸ The oral administration of cytidine analogs such as decitabine and azacitidine is not optimal due to the rapid decomposition in acidic conditions and high first-pass metabolism; the reported bioavailability of these agents ranges from 9% to 41%.^35,38 Decitabine is initially activated by deoxycytidine kinase from a monophosphate form to the active triphosphate form, which is then incorporated into DNA by DNA polymerase. Decitabine can be inactivated through its major elimination pathway involving deamination by cytidine deaminase found principally in the liver, but also in granulocytes, intestinal epithelium, and plasma.³⁸ Urinary clearance of intact drug accounted for 29% of plasma clearance in mice. In humans, total body clearance in the range of 124 ± 19 mL/min/kg exceeds hepatic blood flow and is explained by extensive extrahepatic deamination.³⁹ Decitabine is not a substrate for the cytochrome P450 enzymes.³⁹

Deoxycytidine analogs enter cells rapidly by a nucleoside-specific transport mechanism.¹⁹ Azacitidine and decitabine are activated to triphosphate forms (azacytidine by uridine-cytidine kinase, decitabine by deoxycytidine kinase) which are subject to degradation by cytidine deaminase. Azacitidine incorporates primarily into RNA (80%–90%), and to a much lesser extent into DNA.⁴⁰ Incorporation into RNA produces disassembly of polyribosomes, altered RNA methylation, a defective acceptor function of transfer RNA, and marked inhibition of protein synthesis. In contrast, decitabine incorporates primarily into DNA. Its incorporation into DNA results in covalent trapping of DNA methyltransferase causing inhibition of DNA methylation at concentrations that do not cause major suppression of DNA synthesis.³⁹ The more direct metabolic activation pathway to DNA incorporation of decitabine makes it a more potent compound compared with azacitidine.^41,42

Clinical Experience With Decitabine

Early clinical trials

A phase 1 pharmacokinetic study of decitabine was conducted in 21 patients with advanced solid tumors.¹⁹ The drug was given at dose ranges of 25 to 100 mg/m² infused over 1 hour every 8 hours × 3, and the treatment repeated every 3 to 6 weeks. For the 75 mg/m² and 100 mg/m² doses, mean peak plasma concentrations were 0.93 μM/mL and 2.01 μM/mL, respectively. There was rapid disappearance of drug from plasma with a T_1/2 alpha and T_1/2 beta of 7 minutes and 35 minutes, respectively. Total urinary excretion was <1% of the administered dose, suggesting that decitabine was eliminated rapidly and largely by metabolic processes.¹⁹

Initial phase 1 trials

The first clinical studies of decitabine in hematologic malignancies used 1500 to 2500 mg/m² per course.^17,18 Response rates with decitabine as a single agent or in combination with other therapies were 30% to 60%.^{17,18,20,24,29,43,44} In a phase 1 trial in 30 children with refractory or recurrent leukemia, with administered decitabine doses 0.75 to 80 mg/kg, significant reductions of circulating blasts were reported.¹⁸ Using a dose of 37 mg/kg in a 36-hour infusion, 1 complete remission was noted in a patient with acute lymphocytic leukemia (ALL).¹⁷ Momparler et al.¹⁷ reported similar dose-dependent activity when decitabine was given to 27 patients at doses ranging from 31 to 81 mg/kg over 36 to 60 hours. Nonhematologic toxicity was rare; no MTD was reported. There was detection of 70% inhibition of DNA methylation, but no evidence of blast differentiation.¹⁷

Despite promising activity, high-dose decitabine regimens were not pursued because of delayed and prolonged myelosuppression. At these doses the decitabine effect was likely a cytotoxic one. At lower dose schedules (15 mg/m² 3 times a day for 3 days), decitabine was found to have encouraging activity in MDS.²⁰ At an even lower dose (0.15 mg/kg daily over 1 hour daily for 10 days) decitabine was reported to have biologic efficacy in reactivating hemoglobin F in patients with sickle cell disease.⁴⁵ These observations, combined with the short half-life of the drug and its absolute requirement for DNA synthesis for activity, led to a novel phase 1 trial of decitabine in patients with recurrent or refractory leukemia.³¹ This study tested low-dose longer exposure schedules, with the intent of finding an ‘optimal biologic dose’ to modulate the molecular target achieving a response at levels lower than the classic MTD. Fifty patients (44 with AML/MDS, 5 with CML, and 1 with ALL) were treated with increasing doses of decitabine (5 mg/m², 10 mg/m², 15 mg/m², and 20 mg/m²) intravenously over 1 hour daily, 5 days a week for 2 consecutive weeks. The starting dose per course in this study was 30 times less than the previous established MTD. The exposure duration was then increased to 15 days and 20 days. The treatment was well tolerated and responses were noted at all dose levels. The objective response rate was 32% (16 of 50 patients). In 37 patients with AML, 5 (14%) achieved a complete response (CR) and 3 (8%) achieved a partial response (PR). In 7 patients with MDS, 2 (29%) achieved a CR and 2 (29%) achieved a PR. In 5 patients with CML, 2 (40%) achieved a CR and 2 (40%) achieved a PR. Responses were slow and gradual; the median time to response was 45 days (range, 16–70 days). The dose of 15 mg/m² daily × 10 was judged to be optimally effective (11 responses in 17 treated patients; 65%); lower response rates were noted when the dose was escalated or prolonged (2 of 19 or 11%). The low response rate at high doses was consistent with earlier studies in which decitabine at a dose of 500 to 1000 mg/m² per course induced CR in only 1 of 17 patients for AML recurrence (unpublished data). In this study, p15 gene promoter DNA methylation studies in peripheral blood mononuclear cells were performed in 29 patients (including 7 patients who achieved CRs or PRs), only 15 (52%) of whom (including 2 patients who achieved CRs or PRs) had promoter hypermethylation at baseline consistent with gene silencing (ie, >10% methylated). There was no correlation between p15 methylation at baseline or after therapy and response to decitabine.

Phase 2 studies in MDS

Two large phase 2 studies of decitabine in MDS were recently reported (Table 1). In the studies of Wijermans et al.^20,46,47 in Europe, 169 older patients (median age of 70 years) with intermediate-risk or high-risk MDS received low-dose decitabine (135 mg/m² total dose per course). The overall response rate was 49%, and the induction mortality rate was 7%. Response rates were 51% with high-risk disease and 46% with intermediate-1 disease. Improvement in thrombocytopenia was noted in 63% of patients after 2 cycles.⁴⁸ Complete remissions were associated with cytogenetic remissions.⁴⁹ Cytogenetic responses by the International Prognostic Scoring System (IPSS) were: low-risk, 3 of 5 (60%) patients; intermediate-risk, 6 of 30 patients (20%); and high-risk, 10 of 26 patients (38%). Survival was longer among patients achieving a cytogenetic response compared with those who did not (P =.02).

TABLE 1.

Clinical Results of Single-Agent Decitabine in Patients With MDS (Phase 2 Trials)

Reference	Patient characteristics			Dose	Response
	Median age (range), years	No. evaluable	IPSS INT-2/HI, %		OR rate, %	Median duration	Median survival
Wijermans 200247	70 (38–89)	169	72	45–50 mg/m2/d iv × 3 d every 6 wk	49	40 wk	15 mo
Kantarjian 200750	66 (39–90)	95	66	20 mg/m2/d iv × 5 d every 4 wk (n = 64) vs 10 mg/m2/d iv × 10 d every 4 wk (n = 14) vs 10 mg/m2 twice daily sc 3 × 5 d every 4 wk (n = 17)	73 (CR in 34; PR in 1; HI in 38)	NR	19 mo

MDS indicates myelodysplastic syndromes; IPSS, International Prognostic Scoring System; INT-2, intermediate-2 risk; HI, hematologic improvement; OR, overall response; iv, intravenous; sc, subcutaneously; CR, complete response; PR, partial response; NR, not reported.

In a phase 2 trial of decitabine in patients with MDS,⁵⁰ testing both dose intensity and subcutaneous route of administration, patients received a total dose of 100 mg/m² per course, and were randomized in a Bayesian design to 1 of 3 arms: 1) 10 mg/m² administered intravenously over 1 hour daily for 10 days; 2) 20 mg/m² administered intravenously over 1 hour daily for 5 days; and 3) 20 mg/m² administered subcutaneously daily for 5 days (administered as 2 subcutaneous doses). Cycles were administered every 4 weeks and response or lack of response was evaluated only after at least 3 cycles were given. Ninety-five patients (median age of 67 years) were treated, 77 of whom had MDS, and 18 of whom had CMML. Thirty-two percent had secondary MDS and 66% had intermediate-2 and high-risk disease. The median number of cycles was 71 (range, 1–18 cycles). Overall, 32 patients (34%) achieved a CR and 69 (73%) had an objective response (CR, PR, or hematologic improvement [HI]) by the new modified International Working Group (IWG) criteria.⁵¹ The 5-day intravenous schedule, which had the highest dose intensity, was selected as optimal. The CR rate in that arm was 39%, compared with 21% in the 5-day subcutaneous arm and 24% in the 10-day intravenous arm (P < .05). The high-dose-intensity arm was also better at inducing hypomethylation at Day 5 and at activating p15 expression at Days 12 or 28 after therapy. The side effect profile was favorable and included primarily myelosuppression.

These results appear to be favorable in comparison with the contemporary historic experience at the University of Texas M. D. Anderson Cancer Center that included 2 cohorts. Group A was comprised of 115 patients receiving intensive chemotherapy from 1995 through 2005 and matched for age, IPSS, and cytogenetics. Group B was comprised of all 376 patients treated with intensive chemotherapy from 1995 through 2005 with similar entry criteria as the decitabine study⁵² The CR rates were 43% with decitabine, 46% with intensive chemotherapy in Group A, and 52% with intensive chemotherapy in Group B. Mortality at 6 weeks was 3% with decitabine versus 12% with intensive chemotherapy (P = .002); the 3-month mortality was 7% with decitabine versus 22% with intensive chemotherapy. The survival was better with decitabine versus intensive chemotherapy in Group A (median of 22 months vs 11 months; estimated 2-year survival rate of 47% vs 25% [P < .001]). A multivariate analysis for survival in all 491 patients receiving decitabine or intensive chemotherapy (Group B) identified decitabine as an independent, favorable prognostic factor for survival (P = .006; hazards ratio of 0.74).⁵²

Phase 3 studies in MDS

The encouraging European phase 2 results led to a multi-institutional randomized phase 3 trial in the U.S.²¹ A total of 170 patients with MDS were randomized to receive decitabine at a dose of 15 mg/m² administered intravenously over 3 hours every 8 hours for 3 days (at a dose of 135 mg/m² per course) repeated every 6 weeks, or best supportive care. Decitabine resulted in a higher objective response rate (17% [CR in 9% and PR in 8%] compared with a supportive care response rate of 0% [P < .001]); 12 additional patients treated with decitabine (13%) achieved HI. Responses were durable (median, 41 weeks) and observed across all French-American-British classification system subtypes. The median time to first response (CR or PR) was 3.3 months, or after 2 cycles of decitabine therapy. Unfortunately, 47% of patients received ≤2 courses of decitabine; the median number of courses was 3. On an intent-to-treat analysis there was a trend toward a longer median time to AML progression or death compared with patients who received supportive care alone (median survival of 12.1 months vs 7.8 months; P =.16). Subgroup analysis indicated a greater benefit in IPSS intermediate-2/high-risk disease (median survival of 12.0 months vs 6.8 months; P =.03), and in those with de novo disease (median survival of 12.6 months vs 9.4 months; P =.04). All patients who responded to decitabine had higher quality of life scores and became red blood cell and platelet transfusion-independent in the absence of growth factors during the time of the response.²¹ Therapy was well tolerated with a toxicity profile expected for this class of agent. The most common adverse events were related to existing and aggravated myelosuppression. Grade 3 to 4 nonhematologic toxicities possibly related to decitabine included hyperbilirubinemia (12%), pneumonia (10%), and constipation (1%).²¹ A phase 3 multicenter trial comparing the same dose/schedule of decitabine with supportive care in elderly patients (aged >60 years) with MDS is currently being performed by the European Organization for Research and Treatment of Cancer (EORTC) with enrollment anticipated to be complete in early 2008.

Optimal dosing of decitabine

The issue of optimal dosing of decitabine is currently under evaluation, given that it has dual activity (hypomethylation at low doses, cytotoxicity at high doses). The classic MTD mode of drug development was evidently not appropriate and may have hindered earlier studies and full evaluation. Favorable responses with decitabine were reported at doses 10 to 30 times lower than the classic MTD in patients with MDS.

Correlative studies suggest that the in vitro observation of rapid saturation of the hypomethylation effect (and loss of the differentiation effect) at lower concentrations is also true in vivo.¹⁰ Moreover, it was recently shown that greater hypomethylation and clinical efficacy were the result of a better increased dose intensity of decitabine.⁵⁰ The sum total of dose-finding studies suggest that 1) short bolus infusions may be better than continuous infusion schedules, 2) lower doses are better than high doses, and 3) dose intensity results in higher response rates, and therefore the 5-day intravenous regimen, which has the highest dose intensity, appears to be the most effective schedule. Pharmacologically, these data could be explained by a correlation between peak levels of the drug and responses. Simply put, a high enough dose of the drug is required for intracellular incorporation, after which the intracellular half-life of the drug may be long enough to achieve a therapeutic effect. Although the regimen comprised of 20 mg/m² administered intravenously over 1 hour for 5 days every 4 weeks has demonstrated demethylation, produced pronounced activity in MDS patients, and represents an additional therapeutic option for MDS, the question remains whether greater hypomethylation can be achieved and if responses can be increased by alternative dosing regimens. The issue of optimal treatment duration and maintenance therapy should also be investigated.

Decitabine in patients with MDS and disease recurrence

Twenty-two patients with MDS were retreated with decitabine after a median of 11 months (range, 3–27 months) from the last course of initial decitabine treatment.^53,54 Decitabine was administered in a manner similar to the initial treatment: 15 mg/m² was administered over 4 hours given 3 times per day on 3 consecutive days with a total dose of 135 mg/m², repeated every 6 weeks. Ten patients (45%) responded; there was 1 CR, 2 PRs, and 7 HIs (at the time of initial treatment there were 2 CRs, 4 PRs, and 1 HI).⁵³ These data suggest that disease recurrence may have been due to treatment interruption in some cases rather than intrinsic resistance to decitabine. They also suggest that longer decitabine treatment duration in MDS may be beneficial.

Decitabine after azacitidine failure

Fourteen patients were treated with decitabine at a dose of 20 mg/m² given intravenously daily × 5 days after failure of azacitidine therapy.⁵⁵ Their median age was 74 years (range, 58–85 years). Patients had received prior azacitidine for a median of 4 courses (range, 1–9 courses). Overall, 5 patients (36%) achieved response by the IWG criteria⁵¹: a CR in 3 patients (21%), a PR in 1 patient (7%), and bone marrow CR+/−, other HI in 1 patient (7%). Improvement of thrombocytopenia was noted in 2 of 5 patients (40%) with pretreatment platelets <50 × 10⁹/L. The median duration of remission was 5.3 months and the median survival was 6 months.

Future Directions

Combination therapy

Decitabine may be combined to augment its epigenetic effect. Combinations of decitabine and histone deacetylase (HDAC) inhibitors are synergistic in reactivating gene expression.²⁸ Combinations with inhibitors of methylated DNA-binding proteins or histone lysine 9 methyltransferases are also attractive possibilities. Based on in vitro synergic activity, trials combining decitabine with valproic acid (VPA), vorinostat, depsipeptide, and other HDAC inhibitors are ongoing. In vivo, decitabine has been shown to sensitize cells to the effects of biologic therapy such as retinoic acid⁵⁶ and to increase the expression of proapoptotic molecules,⁵⁷ which may enhance the efficacy of classic chemotherapeutic agents, possibly through reversal drug resistance in selected cases.⁵⁸ In a phase 1/2 study, 54 patients (48 with AML and 6 with MDS) were treated with a fixed dose of decitabine (15 mg/m² over 1 hour as an intravenous infusion daily for 10 days) administered concomitantly with escalating oral doses of VPA (which was well tolerated at a dose of 50 mg/kg/day for 10 days). Twelve patients (22%) achieved objective responses, including 10 (19%) CRs, and 2 (3%) CRs with incomplete platelet recovery (CRp). Among 10 elderly patients with AML or MDS, 5 patients (50%) achieved a response (4 CRs and 1 CRp). Induction mortality was observed in 1 patient (2%). Cytogenetic response was documented in 6 of 8 responders. The median duration of remission was 7.2 months; the median overall survival was 15.3 months in responders. Transient DNA hypomethylation and global histone H3 and H4 acetylation were induced, and were associated with p15 reactivation.⁵⁹ Patients with lower pretreatment levels of p15 methylation had a significantly higher response rate. Further investigation is needed to assess the impact of HDAC inhibitors on the efficacy of decitabine and whether there is the potential to increase response rates with the combination. A randomized phase 2 study comparing decitabine alone with the combination with VPA is currently ongoing at our institution.

Uses in Other Diseases

Low-dose decitabine is an attractive therapeutic modality for several disease states, as hypermethylation of multiple gene promoters and their subsequent silencing has been previously reported.^9,60

Acute myeloid leukemia

In a study by Lubbert et al.,⁶¹ 51 patients (median age, 72 years) with AML received decitabine at a dose of 135 mg/m² administered intravenously over 72 hours, repeated every 6 weeks for up to 4 courses, with all-trans retinoic acid given at a dose of 45 mg/m²/day for 28 days given during course 2 in decitabine-sensitive patients.⁶¹ Maintenance with 20 mg/m² of decitabine administered intravenously over 1 hour on 3 days every 8 weeks was offered to patients completing all 4 courses. In the 29 evaluable patients, the best response was a CR in 4 patients (14%) and a PR in 5 patients (17%), with a median duration of 13 weeks to best response. The median survival was 7.5 months (range, 0.3–21 + months) and the 1-year survival rate was 24%.⁶¹ In a review of patients on the MDS phase 3 randomized trial, 12 patients treated on the decitabine arm (n = 9 patients) and on the supportive care arm (n = 3 patients) were found retrospectively to have AML at baseline by central review.²¹ Of the 9 patients treated with decitabine, 5 (56%) achieved an objective response versus no response noted on the supportive care arm.

In another study, 27 patients (median age, 69 years) with AML were treated with decitabine at a dose of 20 mg/m² administered intravenously daily 35 every 4 weeks; 17 patients (63%) received ≥3 courses. The objective response rate was 26% (3 CRs and 4 CRs with incomplete recovery of platelet counts).⁶² A phase 3 multicenter trial comparing the same dose/schedule of decitabine with supportive care or low-dose cytarabine in elderly patients (aged >65 years) with newly diagnosed AML is currently ongoing. Trials are also evaluating the role of decitabine as maintenance therapy in patients with AML. Patients with unfavorable-risk cytogenetics in first CR or CRp and patients in second CR are randomized to received decitabine at a dose of 20 mg/m² administered intravenously over 1 hour daily for 5 days or no further therapy after consolidation treatment. Decitabine therapy is repeated every 4 to 8 weeks for a total of up to 12 cycles.

Chronic myeloid leukemia

Thirty-five patients with imatinib resistant CML (12 in chronic phase, 17 in accelerated phase, and 6 in blastic phase) received decitabine at a dose of 15 mg/m² intravenously over 1 hour daily × 10 doses.¹⁵ The overall response rate was 54% (34% complete hematologic response [CHR] and 20% partial hematologic responses [PHR]); 6 patients achieved a complete cytogenetic response. The median duration of response was 3.5 months (range, 2–13+ months).¹⁵ Twenty-eight patients with CML (25 with imatinib resistance, 18 of whom were in accelerated phase and 7 of whom were in blastic phase) received a combination of decitabine at a dose of 15 mg/m² administered intravenously daily, 5 days a week for 2 weeks; and imatinib at a dose of 600 mg orally daily.⁶³ CHRs, PHRs, and HI were observed in 9 patients (32%), 1 patient (4%), and 2 patients (7%), respectively. Major and minor cytogenetic responses were observed in 5 patients (18%) and 3 patients (11%) patiets. The hematologic response rate was higher in patients without BCR-ABL kinase mutations (10 of 19 patients; 53%) compared in those with mutations (1 of 7 patients; 14%). The median duration of hematologic response was 18 weeks (range, 4–107+ weeks).

Acute lymphocytic leukemia

In vitro, decitabine causes loss of cell viability and induces apoptosis in ALL-derived cell lines with known DNA methylation alterations. Exposure of these cell lines to decitabine results in hypomethylation and reactivation of putative tumor suppressor genes.⁶⁴ Aberrant DNA methylation of multiple promoter CpG islands is frequently observed in patients with ALL at the time of initial presentation⁶⁵ and at recurrence.⁶⁶ Methylation of specific molecular pathways has been associated with poor prognosis in patients with ALL.⁸ Based on these data, investigators are assessing the safety and activity of decitabine in patients with recurrent/refractory ALL.

Myelofibrosis

Seven patients with myelofibrosis received decitabine administered subcutaneously at a dose of 0.3 mg/kg/day on Days 1 to 5 and Days 8 to 12; cycles were repeated every 6 weeks.⁶⁷ Five patients were evaluable for response. Of these, 2 patients achieved a response, including 1 patient with a CR (normalization of blood counts including transfusion independence). One patient in the blast phase of myelofibrosis had an HI in platelet counts (from 62 K/μL to 200 K/μL) with a decrease (from 2.58 K/μL to 0.03 K/μL) noted in peripheral circulating blasts. The other 3 patients had stable disease; 2 of these patients remained on treatment at the time of last follow-up and had received 4 cycles and 7 cycles of treatment, respectively. A decrease in spleen size was observed in 3 of 4 patients with palpable splenomegaly at baseline. Overall, there was a significant reduction in circulating CD34+ levels, with a mean decrease of approximately 70% at Day 12 of Cycles 1 and 2 (P = .01). The treatment was associated with myelosuppression and minimal extramedullary toxicities.⁶⁷

Solid tumors

Early clinical studies of high-dose decitabine monotherapy have demonstrated limited efficacy.^{34,38,68–70} This finding is currently being re-evaluated using low-dose schedules for potentiation of immune response or resensitization to chemotherapy. In patients with malignant melanoma and renal cell carcinoma, interleukin-2 (IL-2) activates cell apoptosis through lymphocyte stimulation. Malignant cells develop resistance to IL-2 either by down-regulating human leukocyte antigen or decreased expression of apoptotic proteins. In a phase 1 study, decitabine was administered before IL-2 with the objective of increasing protein expression. Major responses were observed in 3 of 13 patients (23%), a response rate comparable to that noted with single-agent IL-2.⁷¹

Sickle cell disease

Decitabine increases the transcription of gammaglobulin, leading to an increase in hemoglobin F.^72,73 This effect was evaluated in a phase 1/2 study of 8 patients with sickle cell disease. Decitabine was administered at a dose of 0.2 mg/kg subcutaneously 1 to 3 times/week × 2 cycles. Hemoglobin levels increased from a mean of 7.6 g/dL to 9.6 g/dL (P < .001), and hemoglobin F levels increased from 6.5% to 20.4% (P <.0001). Further studies are needed to evaluate the long-term effects of such therapy in patients with sickle cell disease.

Conclusions

Given that MDS is a disease of older individuals, aggressive therapies such as combination chemotherapy and stem cell transplantation are not realistic for many patients. Therefore, there is interest in exploring low-intensity or targeted therapies in patients with MDS, and learning how to integrate them in a multiagent therapeutic approach. Decitabine, at lower dose schedules, has been reported to have significant activity in MDS. This activity can be further improved with dose schedule modulations and in combination strategies with growth factors, HDAC inhibitors, and other combinations. Decitabine may have a therapeutic role in AML in several ways: 1) as a single agent in elderly patients not fit for intensive chemotherapy; 2) in combinations with other agents such as HDAC inhibitors, cytarabine, clofarabine, and gentuzumab ozogamycin; and 3) as a maintenance therapy in AML patients in CR after the completion of consolidation therapy, to reduce the potential of disease recurrence through the prevention of DNA hypermethylation/recurrence of residual leukemia. Decitabine should be further explored in CML patients alone after treatment failure of tyrosine kinase inhibitors or in combination with tyrosine kinase inhibitors, in ALL patients, and in patients with other hematologic disorders. Finally, there is a compelling scientific rationale to re-explore decitabine in patients with solid tumors. Given the emerging understanding of epigenetic mechanisms, the role of decitabine in multiple cancer types is just beginning to be understood.