STI-571 (Gleevac/Glivec, imatinib) versus interferon (IFN) + cytarabine as initial therapy for patients with CML: results of a randomized study

Reviewer: Mary Kara Bucci, MD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: May 20, 2002

Presenter: B.J. Druker
Presenter's Affiliation: IRIS Study Group, Portland, OR
Type of Session: Plenary


    The standard therapy for newly diagnosed chronic myelogenous leukemia (CML) is interferon (IFN). The addition of cytarabine to IFN has been shown to improve response rates and overall survival. For patients who have failed treatment with IFN, recent data shows extremely encouraging results with STI-571, a bcr-abl tyrosine kinase inhibitor. The bcr-abl translocation, also known as the Philadelphia chromosome (Ph), is the genetic translocation underlying the vast majority of cases of CML. The disappearance of this mutation during treatment is associated with increased survival. STI-571 has shown a 95% complete hematological response rate (CHR), and a 4% complete cytogenetic response rate (CCR), defined as complete disappearance of the Philadelphia chromosome, in patients who relapsed after or progressed though IFN treatment. STI-571 has been shown to be well-tolerated and effective in such advanced cases, but there has, until now, been no data to support its use in newly diagnosed CML. This large, multi-center randomized study compared STI-571 as single agent therapy to the standard therapy, IFN + cytarabine, in newly diagnosed CML.

Materials and Methods

  • STI-571 was given at 400 mg/day, IFN was given with a target dose of 5 MIU/m2/day, and cytarabine was given at 20 mg/m2/d for 10 days/month.
  • Dose escalations of STI-571 to 600 and 800 mg/day were permitted in cases of no CHR at 3 months, or no MCR at 12 months.
  • The primary endpoint is time to progression (TTP), defined as either death, progression to accelerated phase (AP), progression to blast phase (BP), rapidly increasing white blood cell (WBC) count, or the loss of either a complete hematologic response (CHR) or a major cytogenetic response (MCR). MCR is defined as <35 % Ph positive.
  • Crossover from one arm of this study to the other was allowed for intolerance to therapy, loss of CHR or MCR, or increasing WBC.


  • Between 6/00 and 1/01, 1106 patients were enrolled on this study, with 553 patients randomized to each arm. Patient groups were balanced for patient characteristics and known prognostic factors (WBC, Sokal score, time from diagnosis).
  • Data presented is through January 2002, for a median follow-up of 14 months.
  • Median dose received of STI-571 was identical to the intended dose.
  • Median dose received of IFN was significantly less than intended, but comparable to doses received in previous major trials.
  • 495 (90%) of patients on the STI-571 arm (arm 1) remained on first-line therapy, compared to 165 (30%) of patients on the IFN/cytarabine arm (arm 2). The decreased number of patients remaining of IFN/cytarabine is due to crossover to STI-571 therapy.
  • 9% of patients discontinued STI-571, the majority for withdrawal of consent.
  • 31% of patients discontinued IFN/cytarabine, the majority for an adverse event.
  • 7 pts (1%) on arm 1 crossed over to arm 2; 218 pts (39%) on arm 2 crossed over to arm 1. The majority of crossovers were for intolerance to therapy (23%) or Grade 3-4 non-hematologic toxicity, despite dose reductions and symptomatic management.
  • Results were analyzed by intent-to-treat, which may overestimate the responses in the IFN/cytarabine arm, as these patients were then treated with STI-571, which has proven efficacy as salvage therapy pts who have failed IFN.
  • Results were also analyzed by Kaplan-Meier estimates, which censored patients at crossover.
  • The rates of MCR were 83% (457 pts) on arm 1, 20% (112 pts) on arm 2, (p<0.001).
  • The rates of CCR (0% Ph+) 68% (375 pts) on arm 1 and 7% (41 pts) on arm 2; rates of partial cytogenetic response (PCR, defined as 1-35% Ph+) were 15 % (82 pts) on arm 1 and 13% (71 pts) on arm 2.
  • Kaplan-Meier estimates of response at 12 months are 84% (arm 1) v. 30% (arm 2), (p<0.001).
  • CHR at 12 months by Kaplan-Meier estimates are 96 % (arm 1) v. 67% (arm 2) (p<0.001).
  • Progression-free survival (PFS), the primary endpoint of this study was 97.2% (arm 1) v. 80.3% (arm 2) by strict intent-to-treat analysis. Again, this overestimates the rate of PFS in arm 2 as 45% of these patients crossed over into arm 1, known effective salvage.
  • Pts without accelerated phase (AP) or blast crisis (BC) at 12 months were 98.5% (arm 1) and 93.1 % (arm 2) (p<0.001).
  • Non-hematologic toxicities for all patients who received at least one dose of STI-571 included edema, nausea and vomiting, mucositis, and skin rash, with rare grade 3-4 toxicities.
  • There were increased grade 3-4 toxicities in the IFN/cytarabine arm, including fatigue, depression, myalgias, arthralgias.
  • Hematologic toxicities included: neutropenia, 11% grade 3 and 2% grade 4 in arm 1, and 20% grade 3, 4% grade 4 in arm 2; and decreased hemoglobin: arm 1, 3% grade 3 and <1% grade 4; arm 2, 5% grade 3 and <1% grade 4.

Author's Conclusions

    STI-571 is more effective and better tolerated than current standard therapy. The rates in this study of complete hematolgic response (CHR), major cytogenetic response (MCR), and time to progression were increased in the group receiving STI-571. These results were highly statistically significant. STI-571 was extremely well-tolerated, with few adverse events or patients discontinuing therapy. This trial demonstrates a role for STI-571 as standard first-line therapy in newly-diagnosed CML.

Clinical/Scientific Implications

    The introduction of STI-571 as a tyrosine-kinase inhibitor specifically directed toward a known translocation was of great scientific importance; this drug is the first of its kind to be introduced and has shown such great clinical benefit. This study now defines a role for STI-571 as first-line therapy in newly diagnosed CML, with markedly greater response and RFS rates than the control arm. This study will greatly impact clinical practice, and changes the standard of care in treatment of this disease.

Oncolink's ASCO Coverage made possible by an unrestricted Educational Grant from Bristol-Myers Squibb Oncology.