Abstract
Myelodysplastic syndromes (MDS) represent a clonal hematopoietic stem cell disorder characterized by morphologic features of dyspoiesis, a hyperproliferative bone marrow, and one or more peripheral blood cytopenias. In patients classified according to the Revised International Prognostic Scoring System (R-IPSS) with intermediate or higher-risk disease, there is an increased risk of death due to progressive bone marrow failure or transformation to acute myeloid leukemia (AML). Azacitidine was the first DNA hypomethylating agent approved by the United States (US) Food and Drug Administration (FDA) for the treatment of MDS and the only therapy that has demonstrated a significant survival benefit over conventional care regimens (CCRs) in patients with intermediate or higher-risk disease. Prolonged survival is independent of achieving a complete remission. Azacitidine has been used in older patients with both clinical and hematological improvement as well as an acceptable side effect profile. The most common adverse effect is myelosuppression. These findings support the use of azacitidine as an effective treatment in older patients with higher-risk MDS.
Keywords
Introduction
Myelodysplastic syndromes (MDS), a group of clonal diseases of the bone marrow, are characterized by ineffective hematopoiesis leading to peripheral blood cytopenias and in approximately 35–40% of patients, progression to acute myeloid leukemia (AML) [Silverman, 2010]. MDS predominantly affect older individuals with a median age of 65–75 years. The disease is categorized morphologically according to the French-American-British (FAB) and World Health Organization (WHO) classification systems. The International Prognostic Scoring System (IPSS), which was revised in 2012 as the Revised IPSS (R-IPSS), was developed to predict individual prognosis and is based on the percentage of bone marrow myeloblasts, cytogenetics, and significant cytopenias [Greenberg et al. 1997, et al.2012]. Untreated patients with an R-IPSS score of intermediate or high/very high (collectively known as higher-risk disease) have a median survival of 3, 1.6 and 0.8 years, respectively. Higher-risk patients require treatment in order to modify the natural history of the disease and improve survival. Allogeneic stem cell transplant is the only curative treatment, but many patients are not candidates due to older age, comorbidities, or lack of a suitable donor [Fenaux et al. 2010a].
Aberrant DNA hypermethylation has been implicated in the pathogenesis of MDS. Azacitidine is a pyrimidine nucleotide analog that inhibits DNA methyltransferase, one of the enzymes responsible for DNA methylation. As a result, it alters gene expression and results in transcription of previously quiescent genes [Silverman, 2009]. Based on the results of two large phase III clinical trials, azacitidine has been approved in the United States (US) for all FAB subtypes and in Europe for the treatment of higher-risk MDS as well as AML with ⩽30% blasts and multilineage dysplasia. In this review, we examine the efficacy of azacitidine in elderly patients with MDS and discuss the side effects that can occur with treatment.
Efficacy
The clinical efficacy of azacitidine in patients with higher-risk MDS was demonstrated in two randomized phase III multicenter trials, the Cancer and Leukemia Group B (CALGB) 9221 and AZA-001 studies. The CALGB 9221 trial was a randomized phase III study of azacitidine 75 mg/m2/day subcutaneously for 7 days of each 28-day cycle, compared with best supportive care (BSC) in patients with FAB-defined MDS, of which 46% were defined as higher-risk disease [Silverman et al. 2002]. A total of 191 patients were enrolled with a median age of 68 years. After 4 months of BSC, patients with worsening disease were permitted to cross over to the azacitidine arm. A total of 60% of the azacitidine group demonstrated responses to treatment, including 7% of patients with complete response (CR), 16% with partial response (PR), and 37% with hematologic improvement (HI). This was in contrast with the BSC group, where 5% of patients experienced HI (p < 0.0001). Neither age nor sex influenced response rates. Median time to leukemic transformation or death was 21 months for azacitidine compared with 12 months for BSC (p = 0.007). The median overall survival was 20 months for azacitidine-treated patients compared with 14 months for patients treated with BSC (p = 0.10). This was not statistically significant, possibly due to the crossover design of the study. A landmark analysis was conducted to investigate the effect of crossover on the survival analysis. A significant survival advantage was reported for those patients who initially received azacitidine or crossed over before 6 months (p = 0.03). Among patients aged ⩾65 years with a FAB classification of refractory anemia with excess blasts (RAEB) or RAEB-T (31 patients treated with azacitidine and 37 BSC patients), median overall survival was 19.5 months in the azacitidine group compared with 14 months in the BSC group (p = 0.0388) [Silverman et al. 2005]. Patients who received azacitidine had delayed onset of red blood cell and platelet transfusions. Furthermore, they had improved quality of life compared with patients treated with BSC, as exhibited by greater improvements in physical functioning, dyspnea, fatigue, and overall quality of life (p ⩽ 0.01) [Kornblith et al. 2002].
A pooled analysis of three CALGB trials (8421, 8921, and 9221) demonstrated that the median number of cycles for azacitidine-treated patients to achieve any response was 3 (range, 1–17 cycles). Of those patients who responded, 75% of patients achieved their response by cycle 4 and 90% achieved their response by cycle 6 [Silverman et al. 2006]. This suggests that continued treatment is necessary to increase the probability of response.
The AZA-001 study was designed to confirm the survival benefit that was suggested by the CALGB 9221 study. This trial was an international, randomized phase III study of azacitidine 75 mg/m2/day subcutaneously for 7 days of each 28-day cycle for at least 6 cycles, compared with three conventional care regimens (CCRs) in higher-risk disease [Fenaux et al. 2009]. Prior to randomization, investigators determined which of three CCRs (induction chemotherapy, low-dose cytarabine, or BSC) was most appropriate for each patient, based on age, performance status, and patient preference. A total of 358 higher-risk patients with a median age of 69 years were randomized to the study. Patients had to have an Eastern Cooperative Oncology Group (ECOG) performance status of 0–2 and an estimated life expectancy of ⩾3 months. Azacitidine significantly improved median survival compared with CCRs (24.5 months versus 15 months; p = 0.0001) with a 2-year overall survival favoring the azacitidine-treated arm (50.8% versus 26.2%; p < 0.0001). The survival advantage was seen irrespective of age, including in those over 75, and in patients whose bone marrow blasts were between 20–30% [Seymour et al. 2010; Fenaux et al. 2010b]. In this study, 87 patients were over the age of 75, which is approximately one quarter of all randomized patients. In this subset, the 2-year overall survival rates were 55% in the azacitidine-treated group compared with 15% for the CCR group (p = 0.0003). Survival was superior regardless of karyotype, and responses occurred despite persistence of the abnormal clone. Of particular importance, patients with abnormalities in chromosome 7, who typically have poor outcomes, had a survival advantage when treated with azacitidine with a median overall survival of 13.1 months versus 4.6 months in the CCR group (p = 0.0017). Subgroup analysis revealed that there were statistically significant differences in overall survival between azacitidine and low-dose cytarabine as well as azacitidine and BSC. The survival difference between azacitidine and induction chemotherapy was not significant; however, this could be a reflection of the small numbers of patients preselected to be treated with this regimen.
Secondary end points in AZA-001 also favored treatment with azacitidine over CCRs. The median time to progression to AML was 17.8 months in the azacitidine group versus 11.5 months in the CCR group (p < 0.0001). Patients in the azacitidine-treated arm had higher rates of CR (17% versus 8%; p = 0.015), PR (12% versus 4%; p = 0.0094), and any HI (49% versus 29%; p < 0.0001). Although the CR rate with treatment is modest, it appears that any response to azacitidine translates into a survival benefit [List et al. 2008].
Assessing for HI using International Working Group (IWG) criteria is an important component of evaluating for response in MDS and can help in determining whether to continue or alter therapy. Major HI must last for at least 8 weeks; however, transient cytopenias may occur as part of the treatment as long as counts recover to the response levels prior to the next course of therapy [Cheson et al. 2006]. The proportions of major erythroid responses (40% versus 11%; p < 0.0001) and major platelet responses (33% versus 14%; p = 0.0003) were higher in the azacitidine group compared with the CCR group. No significant difference in major neutrophil response was observed.
In addition, red blood cell transfusion dependence and infections requiring intravenous antimicrobials were significantly reduced in the azacitidine-treated group. Of those patients who were transfusion-dependent at baseline, 45% of the azacitidine-treated patients became transfusion-independent compared with 11.4% (p < 0.0001) in the CCR group. The rate of infections treated with intravenous antimicrobials per patient-year was reduced by 33% with azacitidine compared with CCRs (p = 0.0032).
Results of the AZA-001 study confirmed that continued azacitidine treatment can further improve responses and outcomes in higher-risk patients. The median number of cycles required to demonstrate a response was 3 (range, 1–22). Among the 51% of patients who achieved CR, PR, or HI, a total of 81% achieved a response by cycle 6, and an additional 9% responded by cycle 9. Furthermore, 43% of responders showed an improvement beyond first response after a median of 4 additional treatment cycles (range, 1–11), suggesting that continued treatment is appropriate in the absence of disease progression or unacceptable toxicity.
The standard dosing schedule of azacitidine in the treatment of MDS is 75 mg/m2/day subcutaneously for 7 days in a 28-day cycle. Due to the difficulty in administering the drug for 7 consecutive days, a multicenter randomized phase II study was conducted to compare three modified azacitidine dosing schedules: (1) AZA 5-2-2 (75 mg/m2/day for 5 days, followed by 2 days with no treatment, then 75 mg/m2/day for 2 days), (2) AZA 5-2-5 (50 mg/m2/day for 5 days, followed by 2 days with no treatment, then 50 mg/m2/day for 5 days), or (3) AZA 5 (75 mg/m2/day for 5 days). A total of 151 patients, of which 63% had lower-risk disease, were enrolled. After six cycles of treatment, HI was reported in 44%, 45%, and 56% of the three treatment arms, respectively. Red blood cell transfusion independence was reported in 50%, 55%, and 64% of the groups, respectively [Lyons et al. 2009]. Based on these findings, it appeared that HI and transfusion independence were similar in the three treatment arms; however, these findings were predominantly in lower-risk patients, and overall survival was not compared using the different regimens. Therefore, in higher-risk patients, 75 mg/m2/day for 7 days remains the preferred dosing schedule.
Apart from these two large randomized studies, an analysis of the Surveillance, Epidemiology, and End Results (SEER)-Medicare database evaluating patients diagnosed with MDS during 2001–2007 revealed that patients who had received hypomethylating agents (HMAs) had an improved 24-month survival, especially among patients with refractory anemia or RAEB, compared with patients who did not receive HMA therapy. This large observational study demonstrated that adoption of azacitidine in older patients has resulted in a survival benefit not only in clinical trials but also in the general population [Wang et al. 2012].
Safety
Azacitidine is generally well tolerated in older patients. The most common grade 3 or 4 adverse event is myelosuppression [Fenaux et al. 2009; Silverman et al. 2002]. This toxicity is often transient and not cumulative. It is also more common during early treatment cycles and typically decreases over time in accordance with the achievement of a response [Lyons et al. 2009; Santini et al. 2010]. Although the majority of patients will have count recovery prior to their next cycle, some may require a delay in treatment by 1–2 weeks, and others may require a dose reduction. In the AZA-001 study, 23% of patients required a dose delay and 9% a dose reduction. Blood and platelet transfusions were required in 87% and 29% of patients, respectively. In CALGB 9221, 23% of patients required a dose delay and 11% a dose reduction due to adverse events [Santini et al. 2010]. In our practice, we typically continue with a 28-day cycle of standard dose azacitidine (75 mg/m2 subcutaneously for 7 days) during the first 4 cycles of therapy and support the patient with blood and platelet transfusions as well as prophylactic antibiotics if the patient has severe neutropenia (absolute neutrophil count ⩽ 250). We do a bone marrow after 3–4 cycles to assess for response and to determine the next steps of therapy.
The most common nonhematological adverse event is injection site reactions in patients receiving subcutaneous treatment, which is typically characterized by mild erythema. This can be treated with warm or cold compresses applied to the affected area. Rarely, patients will need additional treatment with corticosteroids or antihistamines. There are reports that topical administration of evening primrose oil can also alleviate this side-effect [Platzbecker et al. 2010]. Gastrointestinal disturbances are also observed. In AZA-001, constipation was the most common gastrointestinal disturbance, occurring in 50.3% of patients and lasting a median of 8 days. It has been postulated this could be secondary to anti-emetic therapy, such as ondansetron, that is administered prior to therapy. In CALGB 9221, the most frequently reported gastrointestinal event was nausea. In general, gastrointestinal side-effects can be managed with concomitant medications including anti-emetics, laxatives, stool softeners, or anti-diarrheals (Table 1). These also tend to improve over time with repeated cycles. Grade 3 or 4 gastrointestinal events occurred in <6% of patients in either study [Santini et al. 2010].
Management of common-side effects in patients treated with azacitidine.
A less common side-effect is pyrexia, which occurred in 5.7% of patients in AZA-001 and 12.7% of patients in CALGB 9221. This lasted for a median of 5–7 days. In patients who present with pyrexia, it is prudent to evaluate for infection by sending blood cultures, urine culture, and assessing for any localizing symptoms. In patients with cough or shortness of breath, additional imaging studies, such as a chest X-ray, is warranted. In addition, the new onset of fatigue or worsening of fatigue was also reported.
Infections and bleeding were not common adverse events although they are associated with MDS and can be exacerbated by treatment. However, in these two studies, there was no significant difference in the rate of these complications in patients receiving azacitidine compared with those receiving BSC. There are no established indications for prophylactic antibiotics or antifungals or for the use of granulocyte colony-stimulating factor (G-CSF) although it can be used in patients who develop neutropenic sepsis. In our practice, we prescribe levaquin prophylaxis to patients with ANC ⩽ 250. We also consider the use of azoles, such as posaconazole or voriconazole, in those who have prolonged neutropenia.
Azacitidine is predominantly excreted in the kidneys, and rare complications, including renal failure and renal tubular dysfunction, have been reported although this has occurred more frequently when azacitidine was used in combination with other chemotherapeutic agents and at higher doses [Peterson et al. 1981]. Elderly patients are more likely to have decreased renal function. For patients with renal insufficiency, there is no formal contraindication to the use of azacitidine; however, there is a subset that will require dose reductions due to worsened renal function and cytopenias during therapy [Batty et al. 2010]. A phase I study conducted in patients with both solid and hematologic malignancies revealed that azacitidine is dose proportional over the 25–100 mg/m2 dosing range, and renal impairment had no important effect on azacitidine pharmacokinetics. Therefore, initial dose adjustments at the start of therapy in patients with renal impairment are not required [Laille et al. 2014]. Patients with hepatic dysfunction should have frequent monitoring of their liver function tests and blood counts as they may also experience a greater incidence of hematologic side-effects. However, there is no recommended dosage adjustment or a threshold for liver function tests that prevents the use of azacitidine. In our practice, we use azacitidine with caution in those with liver function tests ⩾5-times the upper limit of normal.
Combination therapies
Azacitidine has been used in combination with other agents in an effort to improve response rates and ultimately survival in this patient population. A category of potentially synergistic drugs is the histone deacetylase inhibitors (HDACs). A number of clinical trials have tested the combination of HMAs with HDACs in both MDS and AML. HDACs include valproic acid, vorinostat, entinostat, and mocetinostat. The most compelling data came from a study combining azacitidine and vorinostat, concluding that the two agents can be safely administered in repetitive cycles and that the overall response rates and CR rates are superior with the combination than azacitidine alone [Silverman et al. 2013].
Azacitidine has also been combined with lenalidomide with positive results in early phase studies. The rationale stems from the mechanism of action of the two agents: the effects of azacitidine are dependent on cycling cells and lenalidomide, an immunomodulator, inhibits cell cycle progression. In a phase II study using azacitidine and lenalidomide in patients predominantly with higher-risk MDS, the combination was well-tolerated with an overall response rate of 72%. A total of 44% of patients achieved CR [Sekeres et al. 2012].
Subsequently, a US Intergroup Study, S-1117, was conducted that compared azacitidine monotherapy with azacitidine combined with either vorinostat or lenalidomide. Overall response rates and median overall survival were statistically similar for combination arms versus single agent azacitidine. Nonprotocol-defined dose modifications and protocol discontinuation due to toxicity occurred more frequently in combination arms [Sekeres et al. 2015].
Additional combination studies are underway in an effort to improve upon azacitidine monotherapy.
Outcome after azacitidine therapy
HMAs (i.e. azacitidine and decitabine) are the standard of care for older patients with higher-risk MDS. However, only azacitidine has been shown to extend survival in a randomized trial compared with BSC and is therefore the preferred agent in our practice. Patients who have relapsed or are refractory to HMAs have a poor prognosis with a median survival of 4–6 months [Prebet et al. 2011; Jabbour et al. 2010]. There are no approved second-line therapies for this patient population. The majority of patients are not treated with an optimal number of cycles of azacitidine therapy. The average physician treats only a limited number of MDS patients (2.8 patients), which may influence treatment decisions. There is a high rate of treatment discontinuation with 32% of patients receiving only one cycle of azacitidine. Following cessation of azacitidine, 80% of patients receive only BSC [Demakos et al. 2014]. This underscores the need for second line agents as well as avoidance of early cessation of treatment in patients who are tolerating therapy.
Conclusion
The management of MDS in older patients with higher-risk disease presents a clinical challenge. Without treatment, median survival is low due to the risk of infection, hemorrhage, and transformation to AML. Azacitidine is an effective and safe treatment in this patient population and can alter the natural history of the disease. The two randomized phase III studies, CALGB 9221 and AZA-001, have demonstrated that azacitidine is superior to BSC or conventional treatment with regards to improved survival, quality of life, and time to progression to AML. In the absence of disease progression or unacceptable toxicity, maintenance therapy is associated with improved responses and survival benefit. The treatment is generally well tolerated with the most common side effects being myelosuppression, gastrointestinal disturbances, and injection site reactions. Most adverse events are transient and resolve as therapy is continued. Clinical trials are underway to test combinations of azacitidine in order to further improve outcomes for older patients with MDS.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest statement
The authors declare that there is no conflict of interest.