Dr. Roboz has received honoraria from Cephalon, Inc.
Arsenic trioxide and low-dose cytarabine in older patients with untreated acute myeloid leukemia, excluding acute promyelocytic leukemia
Article first published online: 29 SEP 2008
Copyright © 2008 American Cancer Society
Volume 113, Issue 9, pages 2504–2511, 1 November 2008
How to Cite
Roboz, G. J., Ritchie, E. K., Curcio, T., Provenzano, J., Carlin, R., Samuel, M., Wittenberg, B., Mazumdar, M., Christos, P. J., Mathew, S., Allen-Bard, S. and Feldman, E. J. (2008), Arsenic trioxide and low-dose cytarabine in older patients with untreated acute myeloid leukemia, excluding acute promyelocytic leukemia. Cancer, 113: 2504–2511. doi: 10.1002/cncr.23855
- Issue published online: 17 OCT 2008
- Article first published online: 29 SEP 2008
- Manuscript Accepted: 9 JUN 2008
- Manuscript Revised: 3 JUN 2008
- Manuscript Received: 14 APR 2008
- Cell Therapeutics, Inc.
- Cephalon, Inc.
- acute myeloid leukemia;
- arsenic trioxide;
Acute myeloid leukemia (AML) carries a dismal prognosis in older patients. In this study, the authors evaluated the safety and efficacy of arsenic trioxide combined with low-dose cytarabine in untreated patients aged ≥60 years with AML.
In a phase 1/2 design, arsenic trioxide was administered intravenously at a dose of 0.25 mg/kg on Days 1 through 5 and on Days 8 through 12, and low-dose cytarabine was given subcutaneously twice daily on Days 1 through 14 in escalating doses to a target of 10 mg/m2 per dose. Of 64 patients who had pathologically confirmed AML, excluding patients with acute promyelocytic leukemia and using World Health Organization criteria, the median age was 71 years, 10 patients (16%) had treatment-related AML, 40 patients (63%) had an antecedent myelodysplastic syndrome or myeloproliferative disorder, and 35 patients (55%) had unfavorable cytogenetics. Thirty-four patients (53%) had an Eastern Cooperative Oncology Group performance status of 2 or 3.
Complete remission was achieved in 21 of 61 patients (34%), including 15 of 50 patients (30%) who had secondary or treatment-related AML, 10 of 33 patients (30%) who had unfavorable cytogenetics, and 6 of 34 patients (18%) who had a poor baseline performance status. The mortality rate within the first 4 weeks was 8%. Neutropenic fever was observed in >80% of patients, and 41% of patients had bacteremia. Nonhematologic toxicity generally was mild and reversible and included fatigue, nausea, diarrhea, rash, peripheral edema, and elevated transaminases. There were no clinically significant cardiac arrhythmias.
The addition of arsenic trioxide to low-dose cytarabine appeared to improve responses in elderly patients who had AML compared with either agent alone, and a randomized trial of the combination versus single-agent low-dose cytarabine is ongoing. Cancer 2008. © 2008 American Cancer society.
Acute myeloid leukemia (AML) has a poor prognosis in patients aged >60 years. Cytarabine and anthracycline combinations remain the cornerstone of therapy and produce remission in up to 60% of older patients with de novo disease, excellent performance status, and no major comorbidities.1, 2 However, even in this highly selected group, the majority of patients develop disease recurrence and die of their disease or associated complications, and the median survival is from 6 to 12 months. For most patients aged >60 years, conventional chemotherapy induction carries unacceptably high morbidity and mortality with little expectation of durable remission. The statistics are particularly grim in older patients who have adverse prognostic indicators, such as unfavorable cytogenetics, antecedent hematologic disorders, and/or poor performance status, all of which frequently are identified. In these patients, remission rates with standard induction fall and early deaths increase, causing many clinicians and investigators not to offer intensive chemotherapy at all.3 As the median age at diagnosis for AML is approximately 65 years, new treatment options are clearly needed.
Subcutaneously administered, low-dose cytarabine has been used as a lower intensity AML treatment for more than 25 years. Rates of complete remission (CR) are between 15% and 25%, and some patients may achieve improved hematologic parameters.4–6 The regimen generally is well-tolerated and often can be administered on an outpatient basis. The United Kingdom Medical Research Council recently completed a trial comparing low-dose cytarabine administered at 20 mg twice daily for 10 days every 4 to 6 weeks with hydroxyurea for older patients who were not ‘fit’ for chemotherapy.6 Among 217 patients, 18% achieved CR with a median duration of 80 weeks. Both the remission rate and overall survival improved significantly with low-dose cytarabine compared to hydroxyurea/supportive care.
In an effort to improve these results, we combined low-dose cytarabine with arsenic trioxide. Arsenic trioxide is highly effective in recurrent and newly diagnosed acute promyelocytic leukemia and produces modest responses in myelodysplastic syndrome.7, 8 Although no responses were observed in a small study of single-agent arsenic trioxide in patients with AML,9 in vitro and clinical data suggest synergy of arsenic trioxide with ascorbic acid, and there has been activity, including CR, in a few older patients with AML who were treated with arsenic trioxide and ascorbic acid.10 In addition, in vitro data demonstrate that arsenic trioxide has antiproliferative and apoptotic effects on a variety of AML cell lines.11 Arsenic trioxide has several mechanisms of action with potential importance in AML, including induction of cell death through cell cycle arrest, apoptosis, and autophagy; generation of reactive oxygen species and accumulation of intracellular hydrogen peroxide; release of cytochrome c and activation of caspases; inhibition of glutathione peroxidase; and promotion of differentiation and antiangiogenesis.12–14 We hypothesized that arsenic trioxide may augment the clinical activity of low-dose cytarabine in older patients with AML and conducted a phase 1/2 study to determine the safety and efficacy of this combination.
MATERIALS AND METHODS
Patients aged ≥60 years with previously untreated AML (defined according to World Health Organization guidelines) were eligible for study participation if they met the following criteria: 1) serum creatinine ≤2.5 times the upper limit of normal; 2) serum bilirubin ≤ upper limit of normal; 3) absence of t(8;21) and inv(16) cytogenetic abnormalities, unless the patient could not or would not tolerate conventional chemotherapy; 4) absence of t(15;17); 5) absence of uncontrolled cardiac or pulmonary disease; and 6) baseline electrocardiogram (ECG) with a heartrate-corrected QT interval (QTc) <500 msec. Prior therapy with hydroxyurea, 5'-azacytidine, and hematopoietic cytokines for myelodysplastic syndrome was permitted, but prior cytotoxic therapy for AML was not. Approval for the study was granted by the Institutional Review Board of Weill Medical College of Cornell University and the New York Presbyterian Hospital, and all study participants provided witnessed, written informed consent. The study was conducted in accordance with the principles of the Declaration of Helsinki.
Arsenic trioxide (Trisenox; Cephalon, Inc.) was administered as a 2-hour intravenous infusion at a dose of 0.25 mg/kg on Days 1 through 5 and Days 8 through 12. Serum potassium and magnesium levels were required to be ≥4 mEq/dL and ≥2 mg/dL, respectively, before each arsenic trioxide administration. ECGs were obtained at baseline and at least twice weekly during arsenic trioxide treatment. Early in the study, if the QTc was ≥500 msec, then arsenic trioxide was held, electrolytes were repleted, and concomitant medications were altered when possible to reduce the QTc. Arsenic trioxide was administered only when the QTc was <500 msec. If doses were missed, they were replaced at the end of the treatment cycle when possible. On the basis of published data, the protocol subsequently was amended such that the QT interval, rather than the QTc interval, was required to be <500 msec before arsenic trioxide administration.15
Low-dose cytarabine at a dose of 5 mg/m2, 7.5 mg/m2, or 10 mg/m2 was given subcutaneously twice daily on Days 1 through 14 (1 treatment cycle). Bone marrow aspiration was performed on Day 21. If the bone marrow was hypocellular and without evidence of residual disease, then the patient was given antimicrobials, transfusions of blood products, and other supportive interventions as needed until peripheral count recovery, when another bone marrow biopsy/aspiration were performed to assess response. Granulocyte–colony-stimulating factor was permitted at the discretion of the principal investigator. If the Day-21 bone marrow showed residual disease, then the patient was given a second treatment cycle with the addition of 1 g of ascorbic acid intravenously over 30 minutes within 30 minutes of arsenic trioxide. Patients could receive additional treatment cycles if clinical benefit was demonstrated.
For patients in CR, an additional, identical consolidation cycle was given within 4 weeks of count recovery. If the patient required 2 cycles of treatment to achieve CR, then the consolidation was identical to the second treatment cycle (ie, it included ascorbic acid). Responding patients who were unable or unwilling to undergo reduced-intensity allogeneic stem cell transplantation were offered monthly maintenance therapy with low-dose cytarabine at a dose of 10 mg/m2 subcutaneously twice daily on Days 1 through 5 and arsenic trioxide at a dose of 0.25 mg/kg intravenously on Days 1 and 4. Patients who received ascorbic acid during induction and consolidation also received it with the arsenic trioxide infusions as part of maintenance. Recovery of baseline hematopoiesis was required for each maintenance cycle. Supportive measures for optimal care were provided throughout all treatment cycles. Treatment was administered on an outpatient basis when clinically and logistically feasible.
CR was defined as trilineage maturation, absent Auer rods, ≤5% bone marrow blasts, resolution of extramedullary disease, and restoration of normal hematopoiesis with an absolute neutrophil count ≥1 × 109/L and a platelet count ≥100 × 109/L. No response was defined as the failure to achieve CR but without disease progression. Treatment failure/progression of disease was defined either as failure to achieve >50% reduction in blasts compared with baseline or as death from any cause. Induction mortality was death within 30 days of the initiation of treatment.
This was a phase 1/2 study of the combination of arsenic trioxide with low-dose cytarabine in patients aged ≥60 years with previously untreated AML and in untreated patients with intermediate-2 or high-risk myelodysplastic syndrome according to the International Prognostic Scoring System. To our knowledge, this was the first study of this regimen.
The AML and myelodysplastic syndrome study cohorts accrued separately and in parallel following a modified Fibonacci schema; the myelodysplastic syndrome data will be reported separately. Low-dose cytarabine was administered at the same schedule in 3 dosing cohorts, 5 mg/m2, 7.5 mg/m2, and 10 mg/m2, with the latter target dose based on published data. Arsenic trioxide was administered at a fixed dose of 0.25 mg/kg on the basis of previously published safety and efficacy data in patients with myelodysplastic syndrome.7, 8 Other endpoints were CR and induction mortality.
Phase 2 sample size for the AML cohort was determined according to the Simon 2-stage optimal design.16 On the basis of published data, the predicted response rate to single-agent, low-dose cytarabine was 18%.6 Thus, a response rate ≤18% was considered unacceptable for the current study, and an overall response proportion ≥30% was considered worthy of further exploration. The null hypothesis that the overall response proportion was ≤18% was tested against the alternative hypothesis that the response proportion was ≥30%. Sample size computations were performed assuming a 5% level of significance and 80% power. In the first stage, 40 patients would enter the study. If ≤7 patients responded, then the study would be terminated early, and the results would be categorized as negative. If ≥8 patients responded, then enrollment would be extended to 77 patients at most for stage 2, and the results would be declared effective and worthy of further testing if ≥20 patients responded. This 2-stage design would yield a ≥0.8 probability of a positive result if the true percentage of overall responders was ≥30%. It would yield a ≥0.95 probability of a negative result if the true percentage of overall responders was ≤18%.
Overall survival was analyzed using the Kaplan-Meier method, and 95% confidence intervals were constructed using the Greenwood formula. Median follow-up for the cohort was computed from the patients who remained alive. All P values are 2-sided, and statistical significance was evaluated at the .05 α level. All analyses were performed using SAS version 9.1 (SAS Institute, Inc., Cary, NC) and Stata version 8.0 (Stata Corporation, College Station, Tex) statistical software.
Sixty-four patients with untreated AML were enrolled. Patient characteristics are summarized in Table 1. The median age of the study group was 71 years (range, 54-85 years) and 44% of patients were men. One man aged 54 years who had medical comorbidities that precluded conventional induction was treated. The estimated Eastern Cooperative Oncology Group performance status was estimated at 0 in 4 patients (6%), 1 in 26 patients (41%), 2 in 25 patients (39%), and 3 in 9 patients (14%). Ten patients (16%) had treatment-related AML after chemotherapy with or without radiotherapy for another malignancy, and 40 patients (63%) had secondary AML after an antecedent hematologic disorder (eg, myelodysplastic syndrome or myeloproliferative disease). The baseline karyotype was obtained on all patients: Thirty-five patients (55%) had unfavorable cytogenetics (−5/5q−, −7/7q−, t(6;9), t(9;22), and abnormal 3q, 9q, 11q, 20q, 21q, 17p, complex), 26 patients (41%) had intermediate cytogenetics (diploid, +8, −Y, and all other unclassified), and 3 patients (5%) had secondary inv16.
|Characteristic||No. of Patients (%)|
|Median age [range], y||71 [54-85]|
|Treatment-related acute myeloid leukemia||10 (16)|
|Secondary acute myeloid leukemia*||40 (63)|
|Complex (≥3 abnormalities)||20|
|ECOG performance status|
Sixty-one patients (95%) were treated at the second-phase target dose of 10 mg/m2 low-dose cytarabine twice daily on Days 1 through 14. The median age was 72 years (range, 54-85 years), 26 patients (43%) were men, 11 patients (18%) had secondary AML, and 34 patients (56%) had an antecedent hematologic disorder. Thirty-three patients (54%) with unfavorable cytogenetics, 25 patients (41%) with intermediate cytogenetics, and all 3 patients with secondary inv16 were in this group.
Response and Outcome
There were no responses in the 3 patients who were treated below the target dose of low-dose cytarabine. Among the 61 patients who were treated at the target dose, 21 patients (34%) achieved CR, including 15 of 50 patients (30%) with secondary or treatment-related AML, 10 of 33 patients (30%) with unfavorable cytogenetics, and 6 of 34 patients (18%) patients with baseline poor performance status. One additional patient with long-standing, antecedent myelofibrosis met criteria for CR other than persistent splenomegaly, and a second patient had normalization of blood counts and disappearance of peripheral blasts but refused confirmatory bone marrow evaluation. The characteristics of responders are summarized in Tables 2 and 3.
|Patient||Age, y||ECOG PS||Sex||Karyotype||sAML/tAML||No. of Cycles to Response||Response Duration, mo*|
|Characteristic||No. of Responders/Total (%)|
|ECOG performance status|
|Baseline WBC, ×109/L|
|Treatment-related AML||5/10 (50)|
|Secondary AML*||12/40 (30)|
The median age of patients who responded to treatment was 74 years, and the median age of nonresponders was 71 years. Eight of 23 patients (35%) aged ≥75 years, 4 of 25 patients (16%) with a performance status of 2, and 2 of 9 patients (22%) with a performance status of 3 achieved CR. Among responders, 15 patients (65%) had secondary AML or an antecedent hematologic disorder, 10 of 23 patients (43%) had an unfavorable karyotype, 10 of 23 patients (43%) had an intermediate karyotype, and 3 patients (13%) had secondary inv16. Normalization of cytogenetics was observed in addition to morphologic CR in 7 of 14 patients who had baseline abnormal karyotype (50%), including 3 of 3 patients with inv16. The mean white blood cell count among responders at presentation was 2.8 × 109/L (range, 1-36 × 109/L). Of 27 patients who had baseline white blood cell counts >10 × 109/L, 2 patients (7%) achieved CR. The mean baseline bone marrow blast percentage was 35%, and 2 patients met morphologic criteria for erythroleukemia.
The induction mortality rate was 11% for the whole cohort and 8% for the 61 patients who were treated at the low-dose cytarabine target dose, including 4 patients (6%) with sepsis and 3 patients (5%) with disease progression. At a median follow-up 298 days (range, 175-719 days), the median survival was 157 days (95% confidence interval [95% CI], 110-303 days) (Fig. 1). For patients who were treated at the low-dose cytarabine target dose, the median survival was 187 days at a median follow-up of 298 days (range, 175-719 days). In a landmark analysis that excluded patients who died before 3 months, the median survival for the 23 responders was 562 days (95% CI, from 308 days to upper limit not estimated), and the median survival for nonresponders was 191 days (95% CI, 127-342 days) (P < .0001; log-rank test) (Fig. 2). The average CR duration was 7.8 months (range, 3-21+ months), with several ongoing responses. Fourteen patients (67%) required 2 induction cycles to achieve remission. The median time to remission was 46 days. Patients were on study for a mean of 133 days. The mean days of hospitalization was 39, including elective admissions for patients who were unable to receive outpatient treatment for logistic reasons.
Four patients underwent reduced-intensity conditioning allogeneic stem cell transplantation after arsenic trioxide and low-dose cytarabine. One woman aged 61 years with leukemic transformation of myelofibrosis achieved morphologic CR but had residual splenomegaly. She subsequently died from complications of transplantation. A woman aged 59 years with secondary AML that evolved from myelodysplastic syndrome achieved CR with arsenic trioxide and low-dose cytarabine but died of transplantation complications. A women aged 66 years with treatment-related AML after chemotherapy for breast cancer and a woman aged 69 years with treatment-related AML after radiotherapy for breast cancer both had unfavorable cytogenetics at baseline and attained a CR after arsenic trioxide and low-dose cytarabine. They remained well clinically and in remission >18 months after undergoing reduced-intensity allogeneic stem cell transplantation.
Common toxicities are summarized in Table 4, The toxicities observed in >50% of patients included fatigue, nausea, diarrhea, rash, peripheral edema, and elevated transaminases. Grade 3/4 nonhematologic toxicities were infrequent. Grade 4 myelosuppression was observed in all patients, and >80% had at least 1 episode of febrile neutropenia. There were 37 episodes (58%) of bacteremia and 9 (14%) clinically significant bleeding events, including hemoptysis (1 patient), gastrointestinal bleeding (3 patients), subdural hematoma (2 patients), alveolar bleeding (1 patient), groin hematoma (1 patient), and vaginal bleeding (1 patient). Fifteen patients (23%) had pleural effusions requiring diuretics; neither these effusions nor the bleeding events were clearly drug-related. Fourteen patients (22%) required treatment or observation in the medical intensive care unit.
|Toxicity||Grade ≤2, %||Grade ≥3, %|
One thousand six hundred fifty electrocardiograms revealed no clinically significant drug-related arrhythmias. There were no episodes of torsade de pointes. By using the Bazett correction, there were 298 electrocardiograms with a QTc >500 msec and 13 with a QT >500 msec.
Overall survival in elderly patients with AML is dismal. Intensive induction has high mortality, significant morbidity, and little probability of long-term success in most older patients. Many modifications to standard anthracycline/cytarabine induction have been attempted, including intensification of anthracyclines, substitution of an anthracenedione, intensification of cytarabine in induction or consolidation, the addition of other cytotoxic agents, and priming with growth factors.17–25 Although randomized trials have demonstrated differences in CR and disease-free survival, there have been no significant differences in overall survival. In addition, there are inadequate quality-of-life data in elderly AML patients undergoing active treatment, and there is no consensus on determining which older patients are good candidates for intensive therapy. Thus, there is no accepted standard of care, and older patients with AML should be treated on clinical trials whenever possible.
The results of the current study demonstrate that arsenic trioxide can be combined with low-dose cytarabine. The extramedullary toxicity profile was favorable, with significantly lower induction mortality and much less severe nausea, emesis, diarrhea, and mucositis than with intensive induction chemotherapy. Although both drugs have been described as differentiating agents, there was no clinical or morphologic evidence of this. Alopecia was not observed. Rash, edema, and headache were common but mild and reversible. Prolongation of the QTc interval was common, but there were no clinically significant drug-related arrhythmias.
The regimen caused more significant myelosuppression than expected with either agent alone, requiring aggressive supportive care and frequent treatment for infection and bleeding. It is noteworthy that all responding patients achieved bone marrow aplasia. The addition of arsenic trioxide approximately doubled the CR rate expected with low-dose cytarabine alone, and there was a clear survival benefit for patients who achieved CR. In addition, the addition of arsenic trioxide may have hastened the time to CR, which was 114 days (range, 50-313 days) with low-dose cytarabine alone in the Medical Research Council trial.6 Most importantly, although there were no remissions in patients who had unfavorable cytogenetics in the Medical Research Council trial, in the current study, 10 patients in this poor-prognosis group achieved CR with the addition of arsenic trioxide. All 3 patients with inv(16) also achieved CR; although this abnormality is generally considered to have a favorable prognosis, these patients had the unusual and less favorable circumstance of having inv(16) in the setting of secondary disease. The arsenic trioxide and low-dose cytarabine combination was able to induce remission even in patients with the worst prognosis, including those with advanced age, poor performance status, secondary disease, antecedent hematologic disorders, and unfavorable cytogenetics. Finally, there were no unanticipated toxicities in the 4 patients who underwent reduced-intensity transplantation after treatment.
Low-dose cytarabine frequently is offered as an alternative to intensive induction and is currently the ‘control arm’ on several ongoing randomized trials in elderly AML. Nevertheless, a treatment that has a <20% remission rate obviously leaves considerable room for improvement. There have been many single-institution, small, phase 2 trials with promising CR rates and tolerability in older AML patients, but none has emerged as clearly superior, and it is hazardous to compare data retrospectively. The frustration of having treated thousands of older patients with a variety of regimens without any significant improvement in outcome has led investigators in the Medical Research Council to adopt a novel ‘pick-a-winner,’ randomized, phase 2 trial design to more efficiently compare several low-dose cytarabine-based combination regimens, including arsenic trioxide, gemtuzumab ozogamicin, clofarabine, and tipifarnib, with their standard low-dose cytarabine arm.26 It is noteworthy that the agents selected for combination with low-dose cytarabine in this and other ongoing trials have limited single-agent efficacy, including some, such as farnesyl transferase inhibitors and flt-3 inhibitors, that have convincing preclinical rationales. However, it is reasonable to systematically add agents with favorable, nonoverlapping toxicities systematically to a low-dose cytarabine ‘backbone’ in the hope of creating a superior regimen. If arsenic trioxide adds significantly to low-dose cytarabine in an ongoing randomized trial, then its availability as an oral formulation, which can be accomplished readily, would make its incorporation into elderly AML regimens even more appealing.
- 6A comparison of low-dose cytarabine and hydroxyurea with or without all-trans retinoic acid for acute myeloid leukemia and high-risk myelodysplastic syndrome in patients not considered fit for intensive treatment. Cancer. 2007; 109: 1114–1124., , , et al.
- 10Treatment of non-APL acute myelogenous leukemia with intravenous arsenic trioxide plus ascorbic acid [ASH Annual Meeting Abstracts]. Blood. 2006; 108. Abstract 1959., , , et al.
- 19A randomized placebo-controlled phase III study of granulocyte-macrophage colony-stimulating factor in adult patients (>55 to 70 years of age) with acute myelogenous leukemia: a study of the Eastern Cooperative Oncology Group (E1490). Blood. 1995; 86: 457–462., , , et al.
- 21Mitoxantrone versus daunorubicin in induction-consolidation chemotherapy–the value of low-dose cytarabine for maintenance of remission, and an assessment of prognostic factors in acute myeloid leukemia in the elderly: final report. European Organization for the Research and Treatment of Cancer and the Dutch-Belgian Hemato-Oncology Cooperative Hovon Group. J Clin Oncol. 1998; 16: 872–881., , , et al.
- 246-Thioguanine, cytarabine, and daunorubicin (TAD) and high-dose cytarabine and mitoxantrone (HAM) for induction, TAD for consolidation, and either prolonged maintenance by reduced monthly TAD or TAD-HAM-TAD and 1 course of intensive consolidation by sequential HAM in adult patients at all ages with de novo acute myeloid leukemia (AML): a randomized trial of the German AML Cooperative Group. J Clin Oncol. 2003; 21: 4496–4504., , , et al.