Elderly patients with acute myeloid leukemia (AML) are generally unable to withstand the rigors of intensive induction chemotherapy and its attendant complications. Gemtuzumab ozogamicin (GO) is an immunoconjugate that had demonstrated activity in recurrent AML.
The objective of the current study was to determine the prognostic factors for achievement of complete remission (CR) in newly diagnosed elderly AML patients treated with GO as initial induction therapy. A retrospective study was performed of efficacy and toxicity associated with GO therapy, and factors potentially predictive of response were assessed in 49 previously untreated AML patients.
CR was achieved in 14% of all treated patients. Among the patients with an intermediate-risk karyotype, the CR rate was 30%, compared with none with an unfavorable karyotype. The median duration of overall survival was 3.7 months (95% confidence interval [95% CI], 1.4-6.9 months), and the median recurrence-free survival in patients who achieved CR was 11.8 months (95% CI, 5.0-ind months).
Acute myeloid leukemia (AML) is more prevalent in older adults, with a median age at diagnosis between 65 and 70 years.1 Differences in both host and disease biology distinguish AML in the elderly compared with younger patients. Elderly patients have poorer performance status at diagnosis, a higher incidence of poor-prognosis karyotypes, and are more resistant to chemotherapeutic agents. They also have a higher incidence of induction therapy deaths, a lower rate of complete response, and lower long-term survival.2, 3 Thus, new approaches are needed to improve treatment in elderly patients with AML.
Gemtuzumab ozogamicin (GO) is a targeted antineoplastic agent comprised of a recombinant anti-CD33 humanized antibody linked to N-acetyl-γ-calicheamicin.4 GO was approved in 2000 by the United States Food and Drug Administration for use in patients age ≥60 years with CD33+ AML in first recurrence who are not considered candidates for further cytotoxic chemotherapy.5 The most important associated hematologic toxicity is myelosuppression, and a frequently reported life-threatening nonhematologic adverse event is hepatic veno-occlusive or sinusoidal obstructive syndrome (VOD/SOS).6, 7 Several studies also evaluated the efficacy of GO in combination with cytotoxic chemotherapy for treating patients with refractory/recurrent AML8, 9 as well the use of GO as first-line therapy in newly diagnosed AML patients.10-14
To characterize outcomes of GO induction therapy in patients age ≥70 years with previously untreated AML treated at the University of Pittsburgh Cancer Institute (UPCI), we performed a retrospective study of efficacy and toxicity associated with GO therapy. We also assessed factors potentially predictive of response.
MATERIALS AND METHODS
One hundred twenty-eight patients with newly diagnosed, refractory, and recurrent AML (excluding acute promyelocytic leukemia) were treated with GO at the UPCI between February 2004 and October 2008. Among the 128 patients, only the newly diagnosed AML patients age ≥70 years (n = 49) treated with GO were included in the analysis (Table 1). Cases of AML were classified as secondary on the basis of a history of previous treatment with chemotherapy or radiotherapy or having antecedent hematologic conditions, including myelodysplasia and myeloproliferative disorders. The study was approved by the University of Pittsburgh Institutional Review Board according to institutional guidelines.
Table 1. Patient Characteristics
AML indicates acute myeloid leukemia; WBC, white blood cell.
Median age (range), y
AML at diagnosis
Median WBC count (range), ×109/L
% blasts in bone marrow
40.2 (range, 21-85)
Patients received 9 mg/m2 of GO as a single intravenous infusion on Days 1 and 15. The second dose was administered, irrespective of the peripheral counts, if patients had recovered from nonhematologic toxicities resulting from the first dose and had no evidence of uncontrolled infection or disease progression. The outcomes of patients were evaluated for the treatment period, which was defined as 28 days after the administration of GO and for the follow-up period, which was defined as the time subsequent to the treatment period.
Cytogenetic abnormalities were grouped by published criteria.15 The favorable risk category included patients with abnormalities (abn) of inv(16)/t(16;16)/del(16q) or t(8;21) without either a del(9q) or being part of a complex karyotype. The intermediate-risk category included patients characterized by +8, −Y, +6, del (12p), or normal karyotype. The unfavorable-risk category was defined by the presence of 1 or more of −5/del(5q), −7/del(7q), inv(3q), abn 11q, 20q, or 21q, del(9q), t(6;9), t(9;22), abn 17p, or by a complex karyotype, defined as 3 or more abnormalities.
Criteria for Response
Using standard criteria, complete response (CR) was defined by the presence of <5% blasts in the bone marrow, absence of extramedullary leukemia, peripheral blood count recovery with a neutrophil count of at least 1 × 109/L and platelet count of at least 100 × 109/L, and normal cytogenetics (in those with previously abnormal cytogenetics).15 Patients were considered to have no response (NR) if they did not meet all of these criteria.
Comparisons of demographic and baseline characteristics by response status were made by Fisher exact test for categorical variables and by Wilcoxon rank sum test for continuous variables. Overall survival and recurrence-free survival were estimated by the Kaplan-Meier method from our study database, locked on February 28, 2009. Overall survival was measured from the date of treatment start until the date of recorded death. Recurrence-free survival was measured from the date complete response was achieved until the date of documented recurrence or death. For patients presumably still alive or recurrence-free, follow-up was censored at the date of last examination or patient contact. In 2 instances, the upper 95% confidence limit on median survival time was indeterminate (labeled ind) due to an insufficient number of events. SAS statistical software (version 9.2; SAS Institute Inc, Cary, NC) was used for all statistical analyses.
A total of 49 patients with previously untreated AML were evaluated. Patient demographics and baseline characteristics are shown in Table 1. The median age was 77 years (range, 70-88 years) and 47% of patients had secondary AML. Twenty-six (53%) patients had an unfavorable-risk karyotype and 23 (47%) had an intermediate-risk karyotype.
Treatment and Response
All 49 patients completed the first dose of GO; 30 patients received a second dose, the remaining 19 did not continue treatment because of disease progression or infection. Seven of 49 patients (14%) achieved CR; all patients achieving CR received 2 doses of GO. The median time to CR was 23 days after the completion of GO administration. Seven of 23 (30%) intermediate-risk patients achieved CR, as compared with none of the 26 patients with an unfavorable karyotype (Table 2). Ten patients who did not respond to GO received additional treatment, including cytarabine, decitabine, hydroxyurea, etoposide, and idarubicin. Only 2 patients who received idarubicin and cytarabine (7 + 3) achieved a CR.
Table 2. Response Rates
CR Rates by Patient Characteristics
CR indicates complete response; AML, acute myeloid leukemia; NR, no response; WBC, white blood cell.
Risk category at diagnosis
Comparisons of Patient Characteristics Between NR and CR
NR Patients (n = 42)
CR Patients (n = 7)
Median age (range), y
Median WBC count at diagnosis (range), ×109/L
Median % blasts in bone marrow (range)
Median platelet count at diagnosis (range)
Survival of Patients Treated With GO
Forty-four of the 49 patients had died by the time of our analysis. The median overall survival for all patients was 3.7 months (95% confidence interval [95% CI], 1.4-6.9 months) (Fig. 1). The median survival for patients achieving CR was 25.7 months (95% CI, 10.0-ind), and was 1.6 months (95% CI, 1.1-52.0) for those not responsive to therapy (P = .001) (Fig. 2). The median survival for patients with an unfavorable karyotype was 1.2 months (95% CI, 0.6-4.4 months) and for patients with an intermediate-risk karyotype was 6.8 months (95% CI, 3.7-11.5 months) (P = .016) (Fig. 2). The median recurrence-free survival for the 7 patients who achieved CR was 11.8 months (5.0-ind) (Fig. 1).
All patients developed grade 3 to 4 neutropenia and thrombocytopenia, with complete responders requiring a median of 18 days (range, 9-40 days) from the completion of treatment to recover their neutrophils and a median of 23 days (range, 10-33 days) to recover their platelets. A total of 16 (32%) of the 49 patients died during the treatment period. The causes of death were disease progression (n = 7), sepsis (n = 5), multi-organ failure (n = 3), and hemorrhage (n = 1). Grade 3 to 4 transaminitis was observed in 4 patients (8%) and grade 3 elevation in total bilirubin in 1 patient (2%). Signs and symptoms of VOD were observed in 1 patient who died of multi-organ failure after receiving 2 doses of GO.
The therapeutic efficacy and toxicity of GO as first-line therapy in newly diagnosed patients with AML has been evaluated in recent studies, with reported rates of CR ranging from 8% to 27% and a median survival of only a few months (Table 3).10-14 Myelosuppression has been universally observed during GO treatment, and induction deaths ranged from 0% to 37% caused by infections, hemorrhage, and VOD of the liver.
Table 3. Efficacy of GO as First-Line Therapy in Patients With AML
In our study, 49 newly diagnosed patients with AML age ≥70 years were treated with GO. Although the overall CR rate was 14%, none of the 26 patients with high-risk cytogenetics achieved CR. The CR rate for patients with intermediate-risk cytogenetics was 30%, suggesting that this patient subset may benefit from GO treatment as opposed to those patients with high-risk cytogenetics. The beneficial effect of GO in patients with intermediate cytogenetic risk has been reported from the Medical Research Study AML15, which randomly assigned patients aged <60 years to receive induction chemotherapy ± GO/consolidation ± GO.16 Preliminary analysis on 1113 randomized patients demonstrated a significant reduction in recurrence risk and improvement in disease-free survival in patients with favorable or intermediate-risk cytogenetics, but not in patients with adverse cytogenetics. Although no patients with favorable cytogenetics were included in the current study, these patients would also be expected to benefit from GO.
The EORTC Leukemia Group in collaboration with the Italian Cooperative Group GIMEMA conducted a phase 2 study to assess the therapeutic activity and the toxicity profile of GO in older patients with newly diagnosed AML.10 Treatment-related mortality varied by age: 0% in patients ages 61 to 75 years versus 32% (7 of 22) in patients aged >75 years, suggesting that the GO dose and schedule is too toxic for patients aged >75 years. Similarly, in our study, the median age of the patients dying from treatment was 76 years.
An inherent limitation of this single institution retrospective study is that the basis for use of GO, as opposed to alternative options, is effectively unknown, and a comparable control group for comparison is unavailable. More specifically, comorbidity and performance status at the time of treatment might have played a role in the decision making, but regardless of this, we have no data on either 1 of these variables. In addition, because not all newly diagnosed AML patients at our institution received GO, one has to consider that the results of the study could differ. Whether our assessment of GO would be materially changed (ie, much poorer or better recurrence-free and overall survival would be found) cannot be assessed in the absence of data on the eligible newly diagnosed AML patients who did not receive GO. Countervailing strengths of our study are the inclusion of all patients (ie, an unselected patient series) with data regarding several key risk factors for response and survival, and our adjustment for these factors when needed. The role of GO as a single agent in previously untreated AML is still under investigation. In 1 study, CR and survival rates in newly diagnosed AML patients (median age, 71 years; range, 65-89 years) treated with GO were compared with a historical cohort of 31 AML patients treated at the same institution with idarubicin and cytarabine.12 A higher CR rate (48%) and a superior survival was associated with chemotherapy-treated patients compared with GO-treated patients. Based on this comparison, newly diagnosed elderly AML patients were believed to benefit more from induction chemotherapy than GO treatment. However, a prospective randomized clinical trial would be required to determine which approach is optimal.
Intensive chemotherapy may not be appropriate for many elderly AML patients, largely due to treatment-related morbidity and mortality. Low-dose cytarabine has been studied in elderly patients, producing superior CR rates compared with hydroxyurea (18% vs 1%) and better overall survival.17 However, patients with adverse cytogenetics did not benefit from treatment with cytarabine. Several new novel agents have entered clinical trials, and some have shown promising results. The therapeutic efficacy of the hypomethylating agents decitabine and azacitidine as first-line therapy in newly diagnosed patients with AML has been evaluated in recent prospective studies, with reported CR ranging from 24% to 50% and relatively lower toxicity and mortality rates.18-20 Furthermore, the combination of GO with decitabine and azacitidine has been evaluated in AML patients, and preliminary reports suggest that this combination appears to be safe and effective.21, 22
Our experience suggests that GO should be considered as a first-line treatment option in older patients with AML who cannot tolerate high-dose induction chemotherapy, but who have intermediate-risk cytogenetics.