Fax: (011) 49 761 270 3697
Superiority of prolonged low-dose azanucleoside administration?†
Results of 5-aza-2′-deoxycytidine retreatment in high-risk myelodysplasia patients
Article first published online: 13 MAR 2006
Copyright © 2006 American Cancer Society
Volume 106, Issue 8, pages 1744–1750, 15 April 2006
How to Cite
Rüter, B., Wijermans, P. W. and Lübbert, M. (2006), Superiority of prolonged low-dose azanucleoside administration?. Cancer, 106: 1744–1750. doi: 10.1002/cncr.21796
See related editorial on pages 1650-2 and accompanying article on pages 1794-803, this issue.
- Issue published online: 4 APR 2006
- Article first published online: 13 MAR 2006
- Manuscript Accepted: 21 DEC 2005
- Manuscript Revised: 25 NOV 2005
- Manuscript Received: 29 SEP 2005
- DNA methylation;
The optimal treatment duration with decitabine (DAC) in patients with myelodysplastic syndromes (MDS) remains a matter of debate. Although at least 2 consolidating courses after best response usually are performed, the response to treatment after disease recurrence has not been systematically studied to date.
In the current study, the authors report on 22 of 108 patients with MDS (20%) treated with low-dose DAC in 3 Phase II trials who received DAC as retreatment at the time of disease recurrence.
According to the International Prognostic Scoring System (IPSS) at the time of initial treatment, 5 of 22 patients (23%) had a score of intermediate-1 (Int-1), 4 patients (18%) had a score of Int-2, and 13 patients (59%) were scored as high-risk. Patients initially received a median of 6 courses of DAC (range, 2 courses-6 courses), which resulted in a complete remission (complete response [CR]) in 12 of the 22 patients (55%). Retreatment with DAC at the time of disease recurrence was initiated at a median of 11 months (range, 3 mos-27 mos) after the last course of initial treatment. With regard to DAC retreatment, patients received a median of 3 courses (range, 1 courses-6 courses), with 10 of 22 patients (45%) responding (1 with a CR and 2 with partial remissions [partial response (PR)]; all 3 patients achieved a CR at the time of initial treatment) and 7 patients demonstrating a hematologic improvement (HI) (at the time of initial treatment there were 2 CRs, 4 PRs, and 1 HI). Twelve of the 22 patients (55%) did not demonstrate any objective responses to retreatment, including 4 patients with primary resistance to the first course of retreatment. The median survival of all patients from the initiation of the first DAC course was 27.5 months (range, 15 mos-50+ mos). The median survival of 43 patients who also had achieved a response to the initial treatment with DAC but who received best supportive care (n = 33 patients) or induction chemotherapy (n = 10 patients) was 18 months (range, 5 mos-72 mos). Second responders to DAC retreatment were found less frequently in the IPSS high-risk group compared with nonresponders (40% vs. 83%). Age, French–American–British classification subtype, serum lactate dehydrogenase level at retreatment, and previous response to DAC were not found to strongly differ between the groups; however, the subgroups were too small to perform a statistical analysis.
Retreatment with DAC was found to result in objective responses in 45% of previously DAC-responsive patients. However, the quality and duration of the second disease remissions were found to be inferior. Therefore, DAC-responsive patients might derive more clinical benefit from continuation of the initial treatment. Cancer 2006. © 2006 American Cancer Society.
With the exception of allogeneic blood stem cell transplantation, no current treatment of patients with high-risk myelodysplastic syndrome (MDS) results in a high rate of cure. However, because most patients with MDS are older than 60 years, and allogeneic transplantation is performed in only a subset of these patients, the majority of older MDS patients currently are offered best supportive care and, despite lack of a survival benefit, low-dose chemotherapy such as low—dose cytarabine.1 The International Working Group (IWG) on response criteria in MDS has acknowledged the necessity for determining treatment efficacy in this age group based not only on objective response rates but also on hematologic improvement, which can result in an enhanced quality of life (QOL).2 Therefore, therapy regimens that are effective in achieving hematologic improvement and improved QOL in these often elderly patients most likely should be designed as long-term treatment, with the intention to change high-risk MDS into a chronic disease that cannot be cured but can be controlled with extended treatment.
Decitabine (DAC) has been developed at nonintensive schedules that can be given repetitively because of its very low nonhematologic toxicity.3, 4 Similar to 5-azacytidine (a fraction of which also is incorporated into DNA and thereby mediates DNA demethylation), DAC is known to have maximum efficacy only after repeated courses; the median number of DAC courses necessary to achieve cytogenetic normalization is reported to be 3 (which is equal to approximately 4.5 mos), whereas the median time to disease remission with 5-azacytidine is reported to be 3.5 months.5, 6 However, to our knowledge, the optimal number of treatment courses for either drug is unknown.
Because overall response rates of approximately 50% were observed in previous European studies of low-dose DAC in patients with MDS, a Phase II protocol opened in 1997 examining the retreatment of patients who responded to their first treatment with repetitive DAC courses (maximum of 6 courses [duration of approximately 9 mos]). In the current study, we report on the response rate and outcome of both 22 responders to DAC who received retreatment with the same drug and dosing schedule and 43 initially responding patients who received other types of treatment (induction chemotherapy or best supportive care with or without low-dose chemotherapy) at the time of disease recurrence. Approximately 45% of patients receiving DAC retreatment demonstrated a second response, indicating continued sensitivity to the drug. However, in other patients, upfront resistance to second treatment was noted, suggesting that continued initial treatment beyond 6 to 8 courses might delay secondary resistance.
MATERIALS AND METHODS
Between March 1997 and December 2002, a total of 108 patients with MDS were treated with low-dose DAC in 3 Phase II trials. Patients with MDS of the refractory anemia (RA), refractory anemia with excess blasts (RAEB), RAEB in transformation (RAEB-t), or chronic myelomonocytic leukemia (CMMoL) subtypes were included in these trials. DAC at a dose of 15 mg/m2 over 4 hours, given 3 times per day on 3 consecutive days, was administered, with a total dose of 135 mg/m2 given per course. Courses were repeated every 6 weeks, up to a maximum of 6 courses (PCH 91–01 and PCH 95–11 protocols), which were extended to 8 courses in an amendment to the study protocol PCH 97–19. Of the 108 patients studied, 65 (60%) achieved a response (complete response [CR], partial response [PR], or hematologic improvement [HI]) to the initial treatment with DAC. Of these 65 initially responding patients, 43 (66%) received best supportive care (33 patients) or induction chemotherapy (10 patients) at the time of disease recurrence.
Of these initially responding patients, 22 patients (7 females and 15 males, with a median age of 71 yrs range, 51 yrs-81 yrs]) received DAC retreatment at the time of disease recurrence (Table 1). The patients had initially received a median of 6 courses of DAC (range, 2 courses-6 courses). As a result of the initial DAC treatment, 12 patients achieved a CR, 6 patients achieved a PR, and 4 patients achieved an HI. According to the French–American–British (FAB) classification, 3 patients had RA, 10 patients had RAEB, 8 patients had RAEB-t (ie, acute myelogenous leukemia [AML] according to the World Health Organization), and 1 patient had CMMoL (of proliferative type). The International Prognostic Scoring System (IPSS)7 scores before initial treatment for each of the patients was as follows: 5 of the 22 patients (23%) had a score of intermediate-1 (Int-1), 4 of the 22 patients (18%) had a score of Int-2, and 13 of the 22 patients (59%) were scored as high risk. Serum levels of lactate dehydrogenase (LDH) also were determined; the median value was 290 U/L (range, 174 U/L-774 U/L).
|Median no. of courses (range)||6 (2–6)||3 (1–6)|
|IPSS risk group|
DAC was administered in manner similar to the initial treatment: l5 mg/m2 over 4 hours given 3 times per day on 3 consecutive days with a total dose of 135 mg/m2, repeated every 6 weeks. The patients received a median of 3 courses (range 1 courses-6 courses). Retreatment with low-dose DAC was initiated at the time of disease recurrence, which was defined as patients in CR developing > 5% myeloblasts and/or deterioration of blood counts leading to transfusion or, in the case of a low blast count (patients with RA and RARS), a drop in the cell count toward pretreatment levels. The median time from the last course of initial treatment to disease recurrence was 11 months (range 3 mos-27 mos). At the time of retreatment, 1 patient had RA, 13 patients had RAEB, 7 patients had RAEB-t, and 1 patient had sAML. The distribution according to IPSS score at the time of retreatment was similar to those at the time of the initial treatment: 6 of 22 patients were scored as Int-1, 2 of 22 patients were scored as Int-2, and 14 of 22 patients were scored as high risk. The distribution of cytogenetic risk groups was similar to cytogenetics at the time of first treatment. The median serum LDH level was 289 U/L (range 154 U/L-1162 U/L). Table 1 lists the distribution of the patients according to the FAB subtype and IPSS risk score.
Evaluation of Therapy
A CR was defined as a normocellular or slightly hypocellular bone marrow with < 5% blasts and a normal hemogram (hemoglobin > 11g/dL; granulocytes > 1.0 × 109/L; and a platelet count > 100 × 109/L). A PR was defined as a decrease of > 50% in the number of bone marrow myeloblasts and a trilineage response of an increase in the hemoglobin level of at least >2g/dL, in the platelet count of > 50 × 109/L, and in the granulocyte count of > 1.0 × 109/L. A hematologic improvement was defined as a decrease of at least 50% in transfusion requirements, together with an improvement of at least 1 or 2 cell lineages of the peripheral cell counts, but not enough to qualify for a PR. Stable disease (SD) was defined by the absence of a CR, PR, or HI but without clear disease progression. The duration of response was measured from the date of the first response (HI, PR, or CR) until first disease progression. Disease progression (PD) was defined as deterioration of blood counts leading to increased transfusion requirements or an increase in the number of myeloblasts by > 10%.
Survival probabilities were estimated using the log-rank test.8 Statistical analyses were performed with the MedCalc statistical software package (MedCalc Software®;Mariakerke, Belgium) for Windows (Microsoft Corporation, Redmond, WA). This analysis contains all the available data as of January 1, 2005.
Of 108 elderly patients with high-risk MDS who were treated at 2 different centers with DAC in 3 consecutive Phase II trials, 65 (60%) responded. Of these, 22 patients (20%), all of whom had demonstrated a response to the initial treatment, received retreatment with DAC at time of disease recurrence (Table 1). Forty-three other patients who initially were responsive to DAC (40%) received best supportive care (33 patients) or induction chemotherapy (10 patients.) DAC retreatment (which was identical to the initial schedule with 15 mg/m2 of DAC given 3 times per day on 3 consecutive days with a total dose of 135 mg/m2 per course) was initiated at a median of 11 months (range, 3 mos-27 mos) after the last course of initial treatment. The patients received a median of 3 courses of DAC (range, 1 courses-6 courses), approximately half of the initial treatment (median of 6 courses; range, 3 courses -6 courses). Responses to retreatment were noted in 10 of the 22 patients (45%). Seven patients achieved a HI, 2 patients achieved a PR, and 1 patient achieved a second CR (Table 2). The median duration of the second response was 4 months (range, 1 mos-8 mos), which was approximately 2.5 times shorter than that of the first response (median of 10 mos; range, 1 mos-27 mos). Patients with HIs had the same median duration of second response as those with CRs and PRs. Twelve of the 22 patients (55%) did not demonstrate a response to DAC retreatment, including 4 patients with upfront resistance to the first retreatment course. Of 22 patients, 13 (59%) developed disease transformation to AML.
|Initial treatment (%)||Retreatment (%)|
|Complete remission||12 (55)||1 (4)*|
|Partial remission||6 (27)||2 (9)|
|Hematologic improvement||4 (18)||7 (32)|
|Stable disease||0 (0)||3 (14)|
|Progressive disease||0 (0)||9 (41)|
Two examples of second responses are shown in Figure 1. A patient with proliferative-type CMMoL (leukocyte count of > 50,000/μL before treatment) achieved long-lasting control of his leukocytosis (Fig. 1A) as well as HI (data not shown) for nearly 1 year. Several months after first treatment was withdrawn, leukocytosis reoccurred and retreatment was reinstituted, again resulting in rapid control of the leukocytosis, albeit for only 4 months when secondary resistance occurred. This patient spent > 85% of his survival time from his first course of DAC treatment at home, rather than in the hospital (Fig. 1A).
Another patient with RAEB who had achieved a CR after 6 courses of initial treatment (duration of 14 mos) and again after 6 courses of retreatment (duration of 7 mos) received 3 additional courses as second retreatment (for a total of 15 courses of DAC). However, he did not achieve another response to second retreatment. This patient spent > 80% of his survival time from the first course of DAC at home rather than in the hospital (Fig. 1B).
The median overall survival (OS) of all the patients from the first course of DAC given as the initial treatment was 27.5 months (range, 15 mos-50+ mos) (Fig. 2). Table 3 shows a comparison between responders (CR, PR, and HI) and nonresponders (SD, PD) to DAC retreatment. At the time of the initiation of retreatment, patients achieving a second response demonstrated a lower serum LDH level (197 U/L; range, 154 U/L-589 U/L) than the nonresponders (283 U/L; range, 219 U/L-1162 U/L). Of the 12 nonresponding patients, 10 (83%) were considered to be high risk by the IPSS classification at the initation of retreatment, compared with 4 of 10 patients (40%) who demonstrated a second response. No significant differences with regard to age, FAB subtype, serum LDH level, and IPSS score at the time of the first DAC treatment were found between responders and nonresponders to retreatment (Table 3).
|Response to retreatment (n = 10)||No response to retreatment (n = 12)|
|Median age, y (range)||71 (62–81)||71.5 (51–80)|
|Median duration of first response, mos (range)*||12.5 (4–27)||8.8 (3–16)|
|FAB subtype at initial DAC treatment (%)|
|RA||1 (10)||2 (17)|
|RAEB||5 (50)||5 (42)|
|RAEB-t||3 (30)||5 (42)|
|CMMoL||1 (10)||0 (0)|
|FAB subtype at DAC retreatment (%)|
|RA||0 (0)||1 (8)|
|RAEB||6 (60)||7 (58)|
|RAEB-t||3 (30)||4 (33)|
|sAML||1 (10)||0 (0)|
|IPSS score at initial DAC treatment (%)|
|Int-1||3 (30)||2 (17)|
|Int-2||2 (20)||2 (17)|
|High risk||5 (50)||8 (66)|
|IPSS score at DAC retreatment (%)|
|Int-1||4 (40)||0 (0)|
|Int-2||2 (20)||2 (17)|
|High risk||4 (40)||10 (83)|
|Serum LDH at initial DAC treatment, U/L||268 (202–461)||358 (188–774)|
|Serum LDH at DAC retreatment, U/L||197 (154–589)||283 (219–1162)|
The median OS of the other 43 patients who also had demonstrated a response to initial treatment with DAC was 18 months (range 5 mos-72 mos). The median survival from the time of disease recurrence was ten months.
Several Phase II clinical trials investigating low-dose DAC in the treatment of patients with MDS have been performed in Europe since the late1980s. Because the optimal treatment duration is not known, these studies were designed for a minimum of 4 treatment courses and a maximum of 6 to 8 treatment courses, with 2 consolidating courses administered after achievement of the best response. One of these trials (PCH 97–19) also allowed for the retreatment of patients who previously received DAC. For patients who demonstrated a first response to DAC, retreatment with this drug appeared to be a valid alternative to induction treatment which, however, can also be effective as a second-line treatment.9 Of 22 patients receiving DAC retreatment, 10 (45 %) again were found to be responsive to the drug, with 1 patient again achieving a CR. However, the responses in these patients were not of the quality and duration of the first response. Twelve of the 22 patients did not benefit from DAC retreatment, either because they were refractory to the first course of retreatment (4 patients) (with the rapid development of acute myeloid leukemia), or because drug treatment resulted in suppression of the abnormal hematopoiesis without outgrowth of normal hematopoietic cells despite repeated courses of treatment. Indeed, among these patients, 3 died of sequelae of aggravation of cytopenia.
There appear to be 2 major patterns of disease recurrence after effective DAC therapy. The abnormal clone first may be reduced below levels detectable by fluorescence in situ hybridization and then may expand again during a prolonged treatment pause, or the abnormal clone may persist even in the presence of hematologic improvement after the demethylating activity of the drug. Remethylation spreading from areas of incomplete demethylation may be 1 of the mechanisms contributing to clinical disease recurrence.10 In the current series, cytogenetic data in support of either mechanism are available. In 7 patients with an initially abnormal karyotype, response was associated with conversion to a normal karyotype, with subsequent disease recurrence with the same abnormal clone as before the initiation of treatment. In 3 patients, the abnormal clone persisted throughout the response, and was cytogenetically stable at time of disease recurrence.
The results of the current study point to the importance of extending therapy with low-dose azanucleosides beyond the point of first response, and strongly support the institution of a maintenance treatment. The former has been introduced into 2 Phase III studies with this treatment schedule, in which a maximum of 8 courses and 10 courses of treatment, respectively, were given.11, 12 In 1 of these studies, the median number of courses administered was 3, which may explain in part why the benefit on the survival of the entire group was not found to be significant.11 In a study of low-dose DAC for the treatment of AML in elderly patients, an outpatient maintenance treatment was incorporated to maintain the achieved disease remission.12 Although to our knowledge no published studies to date have formally addressed the question of the optimal treatment duration with 5-azacytidine, Silverman et al. reported on continued treatment with multiple courses of low-dose 5-azacytidine in patients with MDS.6 As prolonged treatment with these drugs is introduced, it becomes even more important to monitor patients for the possible late effects of the chronic exposure of normal hematopoietic cells to demethylation.
- 11Decitabine improves patient outcomes in myelodysplastic syndromes: results of a phase III randomized study. Cancer 2006; 106: xxx–xxx., , , et al.