Homoharringtonine (HHT) has antileukemic activity in patients with Philadelphia chromosome (Ph) positive chronic myelogenous leukemia (CML). Combinations of HHT, interferon-α (IFN-α), and cytarabine (ara-C) have been studied in various CML phases. The objectives of this study were to evaluate the efficacy and toxicity profiles of a combination regimen of simultaneous HHT and IFN-α therapy in patients with chronic-phase CML who were not exposed previously to either agent.
Forty-seven patients were treated: 37 patients with early chronic-phase CML (2 patients with clonal evolution) and 10 patients with late chronic-phase CML. Their median age was 62 years (range, 23–73 years). HHT was given at a dose of 2.5 mg/m2 by continuous intravenous infusion over 24 hours daily for 5 days every month, and IFN-α was given daily at a target dose of 5 × 106 units/m2 subcutaneously. Response, survival, and treatment toxicity were analyzed.
Overall, the complete hematologic response (CHR) rate was 85%; the cytogenetic response rate was 66%, with major cytogenetic responses (Ph positive in < 35% of metaphases) in 49% of patients and complete cytogenetic responses in 21% of patients. The CHR rate, cytogenetic response rate, and major cytogenetic response rate were 84%, 69%, and 52%, respectively, in patients with early chronic-phase CML. Among the 10 patients with late chronic-phase CML, the CHR rate, cytogenetic response rate, and major cytogenetic response rate were 80%, 50%, and 40%, respectively. Response rates in patients age > 60 years were 84%, 62%, and 49% for CHR, cytogenetic response, and major cytogenetic response. Myelosuppression was frequent but manageable: Anemia with hemoglobin < 8.0 g/dL occurred in 36% of patients, requiring dose adjustments and erythropoietin therapy. Nonhematologic toxicities were mainly fatigue, aches, and gastrointestinal disturbances. Dose reductions with multiple courses were significant and were due to myelosuppression: After 6–24 courses, the median daily IFN-α dose was 1 MU/m2, and the median number of days on HHT per month was 2 days. With a median follow-up of 26 months, the estimated 2-year survival rate was 90% (95% confidence interval, 79–100%).
The prognosis of patients with Philadelphia chromosome (Ph) positive chronic myelogenous leukemia (CML) has improved significantly with interferon-α (IFN-α)-containing regimens and with allogeneic stem cell transplantation (SCT) for younger patients who have a matched donor. With IFN-α-based therapy, the major cytogenetic response rate was 38% (cytogenetic complete response [CR] rate, 26%), and the rate increased with the addition of cytarabine,1–4 with a median survival of 7 years.1, 2 It has been demonstrated that the achievement of minimal tumor burden, such as a cytogenetic CR, is associated independently with the prolongation of survival.3–8 Thus, new modalities to improve cytogenetic response and, ultimately, survival remain an important research strategy.
Homoharringtonine (HHT), a cephalotaxine ester, was investigated first in China and was reported to be active in patients with leukemia.9, 10 Antileukemic activity has been confirmed in studies conducted in the United States in patients with acute myelogenous leukemia (AML), acute promyelocytic leukemia (APL), and myelodysplastic syndrome (MDS).11–14 In patients with Ph positive late chronic-phase CML, HHT induced a hematologic response in 72% of patients and a major cytogenetic response in 32% of patients.15 In vitro studies demonstrated synergistic effects between HHT and IFN-α, and cytarabine (ara-C) and IFN-α in inhibiting CML colony growth.16 Sequential HHT and IFN-α in the treatment of patients with early chronic-phase CML (< 1 year from diagnosis) produced a complete hematologic response (CHR) rate of 92%; cytogenetic responses were observed in 60% of patients, with a major cytogenetic response rate (i.e., Ph positive < 35%) of 27%.17 Both CHR and cytogenetic response rates were significantly higher compared with the rates seen in patients after 6 months of IFN-α therapy. A combination regimen of HHT and low-dose ara-C in patients with chronic-phase, Ph positive CML who had experienced treatment failure with IFN-α therapy produced a CHR rate of 72% and a cytogenetic response rate of 32%, with a major cytogenetic response rate of 15%. Response rates were identical with HHT plus ara-C compared with HHT alone in patients with late chronic-phase CML, although survival was significantly longer for patients who received the combination after accounting for other prognostic effects.18 Simultaneous administration of HHT and IFN-α in patients with chronic-phase CML without prior exposure to either IFN-α or HHT was investigated. The results are summarized in this study.
MATERIALS AND METHODS
Adults with Ph positive, chronic-phase CML were entered on the study after informed consent was obtained. Eligibility criteria were 1) age ≥ 15 years, 2) no prior IFN-α or HHT therapy, 3) Zubrod performance status ≤ 2, 4) adequate hepatic function (bilirubin ≤ 2.0 mg/dL, alanine aminotransferase < 300 units/mL) and renal function (creatinine < 2.0 mg/dL or creatinine clearance ≥ 60 mL/minute), and 5) no evidence of severe heart disease (cardiac Class III and IV) or psychiatric problems. Patients with accelerated or blastic-phase CML were not eligible, except for patients with cytogenetic clonal evolution as the only accelerated-phase feature. Clonal evolution was defined as the presence of additional chromosomal abnormality other than Ph. Late chronic-phase CML was defined by time from diagnosis to start of therapy > 12 months.
HHT was given at a dose of 2.5 mg/m2 by continuous intravenous infusion over 24 hours daily for 5 days every month, and IFN-α was given daily at a target dose of 5 × 106 units/m2 subcutaneously. Subsequent HHT courses were adjusted by ± 1 day to achieve a lowest white blood cell (WBC) count of about 109/L and a lowest platelet count > 40 × 109/L with each course. HHT courses were given at monthly intervals provided the WBC count had recovered to > 2 × 109/L and the platelet count had recovered to > 60 × 109/L. If the nadir WBC and platelet counts were < 109/L and < 40 × 109/L, respectively, or recovery took > 42 days, then HHT was reduced by 1 day in the subsequent course. IFN-α was reduced by 25% for Grade 2 toxicities; for Grade 3–4 toxicities, it was discontinued until the toxicity subsided, then resumed at 50% of the initial dose. If toxicity was attributed to one agent, then dose reduction was required for that particular agent. Therapy was to be continued for at least 5 years in the absence of unacceptable toxicity.
Therapy was discontinued for the following reasons: 1) evidence of resistance with the optimal acceptable dose schedule; 2) disease transformation; 3) unacceptable toxicity (Grade 3–4) after dose reductions were made; 4) availability of other better options (e.g., allogeneic SCT); or 5) patient or physician choice (e.g., an increase in Ph positive bone marrow cells by at least 30% confirmed by two samples at least 1 month apart, or an increase to ≥ 65%), which resulted in some patients and physicians choosing to change to signal transduction inhibitor 571 ([STI571; imatinib mesylate; Gleevec) therapy. Toxicity was graded according to the National Cancer Institute criteria.19
Pretreatment and Follow-up Studies
Pretreatment evaluation included complete blood counts (CBC), platelet counts and differential, serum chemistries (SMA12), bone marrow aspiration, and cytogenetic studies. Follow-up studies included CBC, platelet counts, and differential weekly for 1 month, then every 2–4 weeks as indicated; SMA12 every other week for 1 month, then every month as indicated; and bone marrow aspiration and cytogenetic studies every 3 months in the first year, then every 4–6 months thereafter as indicated.
The response criteria used have been described previously.20, 21 A CHR was defined as normalization of peripheral counts and differential (WBC count < 10 × 109/L, platelet count ≤ 450 × 109/L) and the disappearance of all signs and symptoms of the disease, including palpable splenomegaly. Patients who achieved a CHR were categorized further by their cytogenetic response: cytogenetic CR (no Ph positive metaphases), cytogenetic partial response (PR; 1–34% Ph positive metaphases), and minor cytogenetic response (35–90% Ph positive metaphases). Major cytogenetic responses included both cytogenetic CRs and cytogenetic PRs (i.e., < 35% Ph positive metaphases). A partial hematologic response (PHR) was defined similar to a CHR but included a persistent reduction ≥ 50% of palpable splenomegaly and either thrombocytosis (platelets > 450 × 109/L) or the presence of a few immature peripheral cells.
Survival analysis was based on the Kaplan–Meier method.22 Survival was calculated from the time treatment was initiated until death or last follow-up. Differences in survival rates between groups were compared with the log-rank test.23P values ≤ 0.05 were considered statistically significant. The prognostic risk group stratification used has been described previously.24
Between October 1994 and October 2000, 47 patients were entered on the study. Their median age was 62 years (range, 23–73 years), and 21 patients (45%) were male. Two of 47 patients had cytogenetic clonal evolution. Their characteristics are detailed in Table 1. None of the patients had previously received IFN-α or HHT. Patients who received at least two courses of chemotherapy were evaluated for response.
Table 1. Characteristics of the Study Group (N = 47 Patients)
No. of patients
WBC: white blood cell.
Performance status (Zubrod)
Splenomegaly (cm bcm)
WBC count (× 109/L)
Platelet count (× 109/L)
Basophils ≥ 7%
Blasts ≥ 3%
Basophils > 4%
Blasts ≥ 5%
Diagnosis to therapy (months)
Risk group in early chronic phase (37 patients)—clinical model24
Among the 47 patients entered on the study, 4 patients did not receive two full courses of therapy (1 patient was taken off the study for neck surgery, 1 patients was taken off the study due to toxicity, and 2 patients refused further therapy). Two patients had evidence of clonal evolution at the start of therapy: Both of them achieved a cytogenetic response with suppression of Ph positive cells to the lowest values of 15% and 70%, respectively.
Among 35 patients with early chronic-phase CML and no clonal evolution, 3 patients were treated in CHR: All 3 patients achieved a cytogenetic response (included below; 1 CR and 2 minor responses). Twenty-seven of 32 patients with active CML achieved a CHR (84%), 1 patient (3%) had a PHR, and 1 patient did not respond to therapy. Cytogenetic responses were noted in 24 patients (69%), with major responses in 18 patients (52%), and CRs in 8 patients (23%) (Table 2).
Table 2. Treatment Response by Disease Status
No. of patients (%)
Early chronic (n = 35 patients)
Late chronic (n = 10 patients)
Chronic + clonal evolution (n = 2 patients)
Overall (n = 47 patients)
The patients were without active disease prior to treatment.
Fifty-two percent of patients with early chronic phase chronic myelogenous leukemia (CML) achieved a complete or partial cytogenetic response, and 40% of patients with late chronic phase CML achieved a complete or partial cytogenetic response, for an overall cytogenetic response rate (excluding minor responses) of 49%.
All 10 patients with late chronic-phase CML had evidence of active disease. Eight patients achieved a CHR (80%), 1 patient (10%) failed to achieve a response, and 1 patient was not evaluable. A cytogenetic response was observed in five patients (50%), with two CRs, two PRs, and one minor response. Responses in patients age ≤ 60 years included CHRs in 88% of patients and cytogenetic responses in 76% of patients (CRs in 13% of patients and major responses in 51% of patients). Among patients age > 60 years, responses included CHRs in 84% of patients and cytogenetic responses in 62% of patients (CRs in 26% of patients and major responses in 49% of patients) (Table 3).
Table 3. Treatment Response by Age (N = 47 Patients)
No. of patients (%)
Age ≤ 60 yrs (n = 16 patients)
Age > 60 yrs (n = 31 patients)
Fifty-one percent of patients age ≤ 60 years achieved a complete or partial cytogenetic response, and 49% of patients age > 60 years achieved a complete or partial cytogenetic response.
The most frequent and severe nonhematologic events during therapy were fatigue and bone aches, which were moderate in 14 patients (30%) and severe in 20 patients (43%) and were associated mostly with IFN-α. Gastrointestinal side effects (e.g., nausea, emesis, and diarrhea) also were observed frequently and were moderate in 13 patients (28%) and severe in 5 patients (11%). These were side effects of both IFN-α and HHT. Only one event (2%) of cardiac disturbance (chest pain and arrhythmia) and two hypotensive episodes (4%) were noted, mostly attributable to HHT. Other side effects, such as depression, weight loss, neurotoxicity, and skin rashes, were associated more with IFN-α. Side effects of the HHT and IFN-α combination are shown in Table 4. Rare side effects (< 5% of patients) included moderate liver dysfunction (2%), oral ulcers (2%), drug fever (2%), and sexual dysfunction (4%).
Table 4. Nonhematologic and Hematologic Side Effects
No. (%) of patients (n = 47 patients)
Percent of courses (n = 667 courses)
Nausea, emesis, diarrhea
Injection site related
Fever of unknown origin
Moderate (Hgb < 10.0 g/dL)
Severe (Hgb < 8.0 g/dL)
Thrombocytopenia (platelets < 50 × 109/L)
Granulocytopenia (neutrophils < 109/L)
Both HHT and IFN-α caused severe myelosuppression. Anemia was a frequent and unexpected complication of the combination regimen. There were 17 patients (36%) who experienced severe anemia with hemoglobin (Hgb) levels < 8.0 g/dL, and 2 of these patients had Hgb levels < 6.5 g/dL. Twenty-four patients (51%) received erythropoietin (Procrit) and/or blood transfusions during the study. Twelve patients had an increase in Hgb by ≥ 2 g/dL posterythropoietin. Thirty-six patients (77%) had severe thrombocytopenia with platelet counts < 50 × 109/L, and 1 patient had a platelet count < 10 × 109/L. Three patients experienced bleeding due to low platelet counts (two patients with gastrointestinal bleeding and one patient with intraocular bleeding). Granulocytopenia (< 109/L) was seen in 38 patients (81%) during therapy, and 17 patients (36%) had granulocytes < 0.5 × 109/L. These complication rates were cumulative throughout the treatment courses. The incidences of such problems by course of therapy after dose adjustments were lower (Table 4). Eleven patients had documented infections, 6 of which were located at the injection sites.
The median number of days on HHT therapy per month was 2 days after 12 months and 1 day after 24 months (Table 5). The median IFN-α dose was 0.8 mU/m2 daily at 12 months and 1.0 mU/m2 daily at 24 months. The total number of courses received was 667, with a median of 13 courses (range, 2–47 courses).
Table 5. Treatment Schedule Delivery with Continued Therapy
Time on therapy (course no.)
No. of patients
Median IFN-α daily dose (MU/m2)
Median no. of HHT infusions (days per month)
IFN-α: interferon α; HTT: homoharringtonine.
The median follow-up was 26 months (range, 3–77 months). Four patients died at times ranging from 13 months to 21 months after the start of therapy from postallogeneic SCT infectious complications (one patient), blastic-phase complications (one patient), accelerated-phase postdecitabine therapy in myelosuppression and infectious complications (one patient), and chronic-phase CML on imatinib mesylate with infections (pneumonia, sepsis, and suspected bowel perforation; one patient). The estimated 2-year survival rate was 90% (95% confidence interval, 79–100%) (Fig. 1). Twenty-three patients were taken off therapy to be treated with imatinib mesylate, and 1 patient was taken off therapy to undergo allogeneic SCT. Because it is possible that imatinib mesylate therapy may alter the natural course of CML significantly, Figure 1 shows survival without and with censoring at the time patients changed to imatinib mesylate therapy or to SCT. Survival was not significantly different among patients with early and late chronic-phase CML at this time of follow-up (estimated 2-year survival rates of 92% vs. 78%, respectively; [Fig. 2] P = 0.14). The median time on therapy was 22 months (range, 2–70 months; Fig. 3). At the time of last follow-up, 13 patients (28%) were continuing on therapy, and 34 patients were taken off therapy for the reasons detailed in Table 6.
Table 6. Patients Status (N = 47 Patients)
No. of patients
Ph: Philadelphia chromosome.
Off therapy due to
Cytogenetic resistance (Ph 100% after 12 months)
Loss of cytogenetic response
Resistance and toxicity
Stem cell transplantation
The course of patients who achieved a major cytogenetic response is shown in Table 7. At the time of last follow-up, 10 of 23 patients (43%) who achieved a cytogenetic response continued to have a cytogenetic response (3 patients still were in CR, 2 patients still were in PR, and 5 patients had minor responses), and 1 patient was in CHR on therapy. Nine patients were taken off therapy due to cytogenetic failure, i.e., increases in Ph positive cells by > 30% or Ph positive cells ≥ 65% after 12 months on therapy) and subsequently were treated with imatinib mesylate; 7 of these patients still were in minor cytogenetic response. Three patients went off study due to toxicity. No patient died in this group.
Table 7. Follow-Up of Patients with Major Response
This study analyzed the efficacy and safety of simultaneous HHT and IFN-α combination chemotherapy in patients with chronic-phase CML. Among 47 patients who were treated, the CHR rate was 85%, and the cytogenetic response rate was 66% (major response, 49%; CR, 21%). With a median follow-up of 26 months, the 2-year survival rate was 90%. Myelosuppression, particularly anemia, was a significant side effect with chronic therapy. Thirty-six percent of patients developed severe anemia (Hgb levels < 8 g/dL) that required dose reductions and was responsive to erythropoietin therapy. Both HHT and IFN-α doses had to be interrupted frequently and were reduced. Other toxicities, such as hypotension and cardiac arrhythmias, were minimal.
The cumulative myelosuppression with both agent required dose schedule reductions. The median number of days on HHT therapy per month was 2 days after 12 months and 1 day after 24 months (reduced from the starting dose of 5 days per month). The median IFN-α dose was about 1 MU/m2 (2 MU) daily at 12 months and at 24 months (reduced from the target dose of 5 MU/m2 daily). Despite these significant dose reductions, CHR and cytogenetic response rates as well as survival rates compared favorably with the rates achieved by IFN-α with or without ara-C therapy. However, the lower daily dose of IFN-α delivered in this combination, compared with our previous studies with IFN-α with or without ara-C (1.0 MU/m2 vs. 3.2 –5.0 MU/m2) and with sequential HHT and IFN-α (2.4 MU/m2),17 is of concern in relation to the durability of major cytogenetic responses. Longer follow-up may determine whether the combination was more effective (as measured by durability of major cytogenetic response rates and survival) with lower doses of IFN-α or whether lowering the IFN-α dose in the combination may have compromised its long-term efficacy. This is an important question not only for this study but also for future combination trials, such as trials combining imatinib mesylate and IFN-α, in which the known effective doses of each agent may be compromised because of cumulative myelosuppression.
IFN-α generally is associated with severe side effects among elderly patients with CML. In this study, 31 patients entered were age > 60 years. Despite their age, response and survival rates in this group were similar to the rates in younger patients (Table 3). This indicates that it may be safe to combine imatinib mesylate with the lower dose schedules of combined HHT and IFN-α in elderly patients. The lower IFN-α and HHT doses in this study also may be due in part to the older age of this study group compared with other groups in studies of patients with chronic-phase CML (median age, 62 years vs. 42–50 years, respectively).
Recent studies have reported that imatinib mesylate induced a CHR in 98% of patients and a cytogenetic CR in about 15% of patients with chronic-phase CML who had disease that failed to respond to IFN-α.25 Because of the encouraging results of imatinib mesylate therapy and its availability, 23 patients were taken off study to start imatinib mesylate (20 patients were taken off study for cytogenetic resistance and/or refractoriness, and 3 patients were taken off study for toxicity). This is important, because imatinib mesylate therapy may alter the natural course of CML significantly. This highlights the dilemma in analyzing the individual benefit of each element in a sequence of therapies on survival, especially if the course of the disease is relatively indolent, like CML or chronic lymphocytic leukemia. Survival data for patients with CML on IFN-α therapy often is reported with the censoring of patients with chronic-phase CML at the time of allogeneic SCT.5, 21 However, this issue has not been addressed for changing to imatinib mesylate therapy in patients with chronic-phase CML or for sequential therapies in patients with other leukemias or malignancies. Therefore, in this study, we elected to report the survival of patients both without and with censoring at the time they changed to imatinib mesylate therapy.
In summary, the HHT and IFN-α combination regimen was safe and effective. Future studies will evaluate combined HHT and IFN-α therapy in patients whose disease progress on imatinib mesylate therapy and will investigate combinations of imatinib mesylate with HHT, IFN-α, or both. This may result in high event free survival rates and potential long-term cures for patients with CML outside the setting of allogeneic transplantation.