5-Azacytidine in Myelodysplastic Syndromes: A Clinical Practice Guideline
Background: Myelodysplastic syndrome (MDS) is a clonal disorder of hematopoiesis resulting in peripheral blood cytopenias and a marked propensity to progress to acute myelogenous leukemia (AML). With 40,000 to 76,000 new cases per year in the USA, MDS is the most common hematological malignancy and represents a significant burden of morbidity and premature death. Although supportive or palliative measures such as blood transfusion have long been the mainstay of management, disease-modifying medical therapies have recently become available. The most extensively characterized of these is 5-azacytidine (5-Aza); however, no consensus exists on how this agent should be deployed in MDS.
Methods: An overarching search of the literature identified 7,019 citations investigating the treatment or management of MDS. Of those, six clinical articles of prospective phase 2–3 study design or meta-analyses were selected for inclusion in a systematic review of the evidence.
Conclusions: The Canadian Consortium on Evidence-Based Care in MDS recommends 5-Aza as first-line therapy in all MDS patients with International Prognostic Scoring System (IPSS) high-intermediate and high-risk scores, including WHO-defined AML (20–30% blasts), who cannot proceed immediately to allogeneic stem cell transplant. 5-Aza is not recommended as first-line therapy for MDS patients with IPSS low and low-intermediate risk scores, as there is no evidence that it alters the natural history of the disease or is superior to standard therapy. The MDS consortium does not recommend combining 5-Aza with other agents at this time outside the context of a clinical trial.
Target Population: This evidence-based guideline applies to adult patients with MDS classified by the 2001 World Health Organization (WHO) Classification of Tumours of Haematopoietic and Lymphoid Tumours and the 1982 French–American–British (FAB) Classification of MDS of all risk categories, subtypes, and performance status.
Clinical Questions Addressed:
In patients with MDS, what is the efficacy of 5-Aza alone or in combination, as measured by survival, quality of life, disease control (i.e., time to progression), response duration, or response rate (i.e., transfusion independence, complete remission [CR], and partial remission [PR])?
What toxicities are associated with the use of 5-Aza?
Which patients are more or less likely to benefit from treatment with 5-Aza?
Introduction: MDS is a malignant clonal stem cell disorder characterized by ineffective hematopoiesis resulting in peripheral blood cytopenias and a propensity to develop AML. This disease primarily affects the elderly, with a median age at diagnosis of 74, although younger patients are also at risk, particularly those previously exposed to mutagenic chemotherapy or ionizing radiation. Common manifestations include red blood cell or platelet transfusion dependence, infections, bleeding, cardiac dysfunction, fatigue, dyspnea, and diminished quality of life. The only curative therapy for MDS is allogeneic stem cell transplant, but this option is available for very few patients due to age and comorbidities. As a result, most treatments are supportive, aiming to improve quality of life by mitigating anemia, reducing or eliminating red blood cell transfusions, and preventing leukemia.
Treatment algorithms adopted by most clinicians are dictated by the International Prognostic Scoring System (IPSS), which is a weighted scoring system derived from three parameters: marrow blast percentage, number of cytopenias, and karyotype. It divides MDS patients into four risk groups: low risk (median survival 5.7 years), low-intermediate (Intermediate-1) risk (median survival 3.5 years), high-intermediate (Intermediate-2) risk (median survival 1.2 years), and high risk (median survival 0.4 years). Approximately one-third of patients fall into the high-intermediate and high-risk groups, whose relative survival compared with age-matched controls is significantly reduced due to the development of acute leukemia or infection. A majority (63–79%) of these patients are red cell transfusion-dependent.
Lower-risk patients are generally treated with hematopoietic growth factors, lenalidomide (in the small minority with del 5q MDS), iron chelation, immunosuppressive therapy, and blood transfusions. Therapeutic options for higher-risk (Intermediate-2 and high risk) MDS patients are limited; induction-style chemotherapy is poorly tolerated and rarely effective.
Recent advances have improved understanding of MDS biology, recognizing that aberrant epigenetic silencing of key genes may contribute to pathogenesis. One mode of gene silencing is methylation, a reversible process targeted by selective agents; 5-Aza (Vidaza™) and decitabine (Dacogen™) are two such drugs. 5-Aza is a pyrimidine nucleoside analog developed as an antitumor agent. In addition to cytotoxic effects, it induces differentiation of malignant cells in vitro. The phosphorylated and reduced metabolite of 5-Aza inhibits DNA methyltransferase, the enzyme responsible for methylating newly synthesized DNA in mammalian cells, resulting in hypomethylated DNA and changes in gene transcription and expression.
Methods: This systematic review was developed by the Canadian Consortium on Evidence-Based Care in MDS, simulating the methodology of Cancer Care Ontario’s Program in Evidence-Based Care (PEBC). Evidence was selected and reviewed by two members of the consortium. The review primarily comprises randomized controlled trial (RCT) data and forms the basis of a clinical practice guideline intended to promote evidence-based practice in Canada. The guideline is editorially independent of Cancer Care Ontario and the Ontario Ministry of Health and Long-Term Care.
Literature Search Strategy: The Medline (1985 to June 17, 2009) and EMBASE (1985 to June 17, 2009) databases, as well as the Cochrane Database of Systematic Reviews (2009, Issue 2) and the Cochrane Central Register of Controlled Trials (2009, Issue 2), were searched for studies related to MDS treatment, including chronic myelomonocytic leukemia, using subject headings and text words such as myelodysplastic, MDS, and chronic myelomonocytic leukemia (CMMoL). Filters for systematic reviews, meta-analyses, clinical trials, comparative studies, and cohort studies were applied. Conference proceedings from the annual meetings of the American Society of Clinical Oncology (ASCO) and the American Society of Hematology (ASH) were also searched.
Study Selection Criteria: Inclusion criteria were clinical articles of prospective phase 2–3 study design or meta-analyses published in English, including adult patients with MDS of either FAB or WHO classification of any IPSS risk score, evaluating 5-Aza as a single agent or in combination administered intravenously or subcutaneously. Comparative trials where 5-Aza was compared with any agent, placebo, or standard care were included if they reported outcomes such as survival, quality of life, disease control, response duration, response rate, or adverse effects. Meeting abstracts from ASCO and ASH (2006–2009) were included if they provided ancillary clinical data for included RCTs.
Exclusion criteria were letters, comments, books, notes, consensus guidelines, registry or editorial publications; studies with fewer than 20 MDS patients; and studies where clinical results of higher-risk MDS patients could not be distinguished from those with AML by FAB classification.
Article Selection: The initial search was broad, aiming to find all trials investigating MDS treatment. The consortium subdivided topics by treatment agents. One reviewer screened citations for inclusion and grouped results by treatment agent. Two reviewers then reviewed citations for 5-Aza inclusion. Citations were scored as “Yes” (met inclusion, no exclusion criteria), “No” (met exclusion criteria), or “Maybe” (unclear). “Yes” and “Maybe” citations were reviewed further, and discrepancies were resolved by consensus or a third reviewer.
Synthesizing the Evidence: No data appropriate for pooling or meta-analysis were found.
Results: The literature search identified 7,019 citations related to MDS treatment or management. After initial evaluation, 26 citations were scored as “yes” or “maybe.” Of these, six publications met eligibility criteria and were included. Among these, two were randomized controlled trials, one was an updated summary of results from two phase II studies, one was a re-analysis of an RCT using WHO classification and International Working Group (IWG) response criteria, one was a quality-of-life study from a pivotal RCT, one was a randomized phase II study, and one was a phase II study of 5-Aza and thalidomide in MDS and AML (with 60% MDS patients). Four studies were excluded due to small sample size, lack of appropriate clinical endpoints, or presentation of mature results in included publications.
Six abstracts from ASCO 2008 and ASH 2008 meetings were included. Twenty-two abstracts were excluded for reasons including small sample sizes, lack of additional information, meta-analyses that could not be critically appraised, cohort or registry studies, maintenance therapy after induction chemotherapy, phase 1 or 2 trials, or pharmacoeconomics studies.
Critical Appraisal: Of the six MDS trials included, four were multi-center trials, with pharmaceutical authorship or sponsorship noted in two. Critical appraisal was possible for four trials.
In the Cancer and Leukemia Group B (CALGB) multi-center phase III trial (Protocol 9221), 191 patients with MDS were randomly assigned to azacitidine or best supportive care. The trial was unblinded, and patients were stratified by FAB classification. IPSS scores were available for 42% of patients. After central pathology review, 20 patients were determined to have AML at study entry and were excluded from time to AML transformation and progression analyses. The two arms were balanced in baseline characteristics. The protocol permitted crossover from supportive care to 5-Aza upon disease worsening. Analyses were intention-to-treat. Response criteria predated current IWG response criteria. All enrolled patients were included in analyses; median follow-up was not provided. The study was publicly funded, with final analyses conducted by CALGB. Four pre-planned analyses were conducted, overseen by an independent data-safety monitoring board.
In the prospective international multi-center phase III trial (AZA-001), 358 patients with higher-risk MDS were randomized to azacitidine or conventional care regimens (best supportive care, low-dose cytarabine, or intensive chemotherapy). The trial was unblinded, with stratification by FAB and IPSS. No crossover was permitted. An independent data-safety monitoring board and scheduled interim efficacy analysis were in place. Using WHO classification, 35% of patients would be re-classified as AML (refractory anemia with excess blasts in transformation, RAEB-t). Median follow-up was 21.1 months. Protocol deviations included enrollment of 18 IPSS intermediate-1 patients, eight allogeneic transplants, and 18 patients who never received study drug, with some imbalances between treatment arms. The pharmaceutical company manufacturing 5-Aza funded the trial and collaborated with principal investigators on final analyses.
The AZA-001 trial demonstrated that azacitidine significantly improved overall survival compared with conventional care regimens in patients with higher-risk myelodysplastic syndromes (MDS). The median overall survival was 24.5 months in the azacitidine group versus 15.0 months in the conventional care group, representing a 9.5-month improvement. Additionally, azacitidine treatment resulted in higher rates of complete remission, partial remission, and hematologic improvement. The drug also delayed progression to acute myeloid leukemia (AML) and improved quality of life measures.
Safety profiles in the AZA-001 trial indicated that azacitidine was generally well tolerated. The most common adverse events were hematologic toxicities, including neutropenia, thrombocytopenia, and anemia, which were manageable with dose adjustments and supportive care. Non-hematologic side effects such as gastrointestinal symptoms and injection site reactions were mostly mild to moderate.
Subgroup analyses revealed that the survival benefit of azacitidine was consistent across various patient categories, including those aged 65 years and older, patients with poor-risk cytogenetics, and those with different FAB subtypes. These findings support the use of azacitidine as a first-line therapy in elderly patients with higher-risk MDS who are not candidates for allogeneic stem cell transplantation.
Subsequent studies have reinforced the efficacy and safety of azacitidine in routine clinical practice. Real-world data from registries and retrospective analyses confirm response rates and survival benefits comparable to those observed in clinical trials. Moreover, extended treatment beyond initial response has been associated with deeper and more durable remissions.
Despite these advances, challenges remain in optimizing treatment for MDS patients. Identifying biomarkers predictive of response to azacitidine could help tailor therapy and improve outcomes. Additionally, combining azacitidine with other agents is under investigation but currently not recommended outside clinical trials due to insufficient evidence.
In conclusion, azacitidine represents a significant advancement in the management of MDS, particularly for elderly patients with higher-risk disease. It offers meaningful improvements in survival, disease control, and quality of life with an acceptable safety profile. Clinical practice guidelines recommend its use as first-line therapy in appropriate patients, emphasizing the importance of individualized treatment decisions based on patient characteristics and preferences.
Ongoing research and clinical trials will continue to refine the role of azacitidine and explore novel therapeutic combinations to further improve outcomes for patients with MDS.