Maurie Markman, MD
Editor-in-Chief of OncologyLive
Senior vice president for Clinical Affairs and National Director for Medical Oncology
Cancer Treatment Centers of America, Eastern Regional Medical Center
Several worthy candidates can surely be proposed as representing the single event that definitively shifted the oncology paradigm from rather simplistic morphology-based treatment to an intensive search for relevant molecular targets to which antineoplastic therapeutics could be directed more rationally. Few observers, however, would likely question that one of the major contenders for this singular honor should be the demonstrated impact of imatinib mesylate on outcomes for patients with chronic myeloid leukemia (CML).
Imatinib, a selective tyrosine kinase inhibitor of the bcr-abl
fusion oncogene (the molecular abnormality recognized to be essential in driving the malignant cell population), was shown in early-stage trials to produce rather profound biological and clinical remissions in the large majority of patients so treated. In fact, in a landmark phase I study, 53 of 54 patients achieved a complete hematologic response (Nat Rev Clin Oncol
As a result, it should really come as no surprise that while the definitive, evidence-based randomized phase III trial comparing imatinib with a standard-of-care regimen of interferon-alpha plus low-dose cytarabine revealed a substantially higher complete cytogenetic response rate and long-term relapse-free survival associated with the novel agent, overall survival was not improved (N Engl J Med
. 2006;355(23):2408-2417). This peculiar outcome was a direct result of the fact that, ethically, patients who progressed on the control arm had to be permitted to cross over to receive imatinib. In fact, 90% of patients randomized to the nonexperimental regimen received imatinib within 9 months from the initiation of antineoplastic treatment (N Eng J Med
An alternative approach to exploring the impact of this novel targeted strategy on overall survival in CML—even if perhaps it is a less definitive approach—would be to examine data from individual institutions with extensive experience managing this cancer, both before and following the introduction of the agent into routine clinical practice. At such centers, the general patterns of care (eg, patient populations, standard testing, follow-up strategies) and acknowledged clinical expertise have remained relatively constant over these time intervals, reasonably independent of the specific drug regimens employed. Although there is the potential for selection bias in any comparison between nonrandomized patient populations, the availability of comprehensive data from major centers helps to reduce (although certainly does not eliminate) the concern for such bias.
Investigators from the hematologic malignancy program at the University of Texas MD Anderson Cancer Center in Houston are widely acknowledged to rank among those with the most extensive experience worldwide in the management of CML. In a recent report on the survival of newly diagnosed individuals in the chronic phase of the disease process who were cared for at their institution over the preceding 45 years, investigators noted the truly striking improvement in 8-year overall survival based on the year treatment was initiated (Blood.
Table. 8-Year Survival For Newly Diagnosed Patients With Chronic-Phase CML, at MD Anderson Cancer Center
Adapted from Kantarjian H, O’Brien S, Jabbour E, et al. Improved survival in chronic myeloid leukemia since the introduction of imatinib therapy: a single-institution historical experience [published online ahead of print January 6, 2012]. Blood. 2012;119(9):1981-1987.
Prior to 1983, the majority of patients were treated with the combination chemotherapy regimen of hydroxyurea plus busulfan. The era of interferon-alpha-based treatment (with or without chemotherapy) began in approximately 1983 and ended in 2000, with patients receiving imatinib or another tyrosine kinase inhibitor since that time (2001). Of note, before 1975 the estimated 8-year survival in this population was only 6%.