The second-generation BCR-ABL inhibitors have demonstrated better efficacy than imatinib in the ENESTnd and DASISION trials for the first-line treatment of patients with Philadelphia chromosome-positive chronic myelogenous leukemia (CML). In these studies, the endpoints of major molecular response and complete cytogenetic response were significantly improved, resulting in FDA approval, states moderator Harry P. Erba, MD, PhD.

In terms of their frontline utilization, the second-generation treatments, nilotinib and dasatinib, have demonstrated better efficacy and tolerability than imatinib, making them a clear choice. However, imatinib remains a highly effective drug that costs less than the second-generation agents. If costs were not a concern, nilotinib or dasatinib would be used as the first-line standard, Erba believes.

To determine which agent to start for newly diagnosed patients with CML, Rami S. Komrokji, MD, recommends utilizing Sokal scores. Patients with intermediate to high scores should receive second-generation agents. For patients at lower risk, Komrokji feels comfortable offering imatinib as a frontline option, as long as the patient is monitored closely for progression.

In addition to the Sokal score, the toxicity profiles for each agent can be utilized to determine the most appropriate drug to administer. From a practical standpoint, imatinib is not as well tolerated as some of the newer agents, Mark J. Levis, MD, PhD, states. As the toxicity profile becomes more severe, patients are less likely to be compliant, which leads to progression, Levis notes. Going forward, as imatinib becomes available generically, payers may not provide physicians with a choice of first-line therapy, Levis believes.

If costs were not a concern, the second-generations TKIs would undeniably be used in the first-line setting, states Elias J. Jabbour, MD. However, since costs are a concern, better tools are needed to determine the optimal patients for each therapy, Jabbour states. These tools would need to look at features intrinsic to each patient, such as metabolites, polymorphisms, or immune response data, Rafael Bejar, MD, PhD, believes. This level of individualization could make research difficult.