Patient Nonadherence and Treatment Response to Imatinib in Patients with Chronic Myeloid Leukemia: Results from the ADAGIO Study

BACKGROUND. Imatinib therapy for chronic myeloid leukemia (CML) is a long-term treatment potentially compromised by patient nonadherence.

OBJECTIVE. To examine whether patients (pts) at different levels of treatment response differ in adherence to imatinib treatment.

DESIGN & PATIENTS. The ADAGIO study¹ is a prospective, 90-day (90d) observational, open-label, multicenter study of pts with chronic myeloid leukemia (CML) and treated with imatinib. 169 evaluable pts who had been on imatinib for a minimum 30 days at enrollment were studied. A sub-analysis included pts with optimal vs. suboptimal response (all patients) and complete vs. incomplete cytogenetic response (CgR; all patients and those treated with imatinib ≥12 months).

MEASUREMENTS. Adherence: imatinib pill count over 90d expressed as % of prescribed imatinib taken. Suboptimal response (SR): incomplete hematologic response at 3 months, and/or less than partial CgR at 6 months, and/or less than major molecular response and, in case of loss of major molecular response, other limitations or chromosomal abnormalities at 18 months (all else: optimal response [OR]). CgR: complete (0% Ph+ metaphases) or incomplete (≥1 Ph+ metaphases).

RESULTS. Pill count percentages ranged from 29%–202% of prescribed dose (M=90.9±20.1). Pts with SR (n=14) had significantly higher %s of imatinib not taken (23.2±23.8) than did those with OR (n=124; 7.3±19.3, P=0.005). Among pts treated with imatinib ≥12 months, those with complete CgR (n=98) had significantly lower mean percentages of imatinib not taken (9.0±18.6) than those with incomplete CgR (n=9; 26.0±24.4, P=0.012). Among all patients regardless of length of treatment, those with complete CgR (n=109) also had significantly lower mean percentages of drug not taken (9.1±18.1) than those with incomplete CgR (n=19; 23.9±19.2, P=0.004).

CONCLUSIONS. Proportions of CML patients with poor treatment response are low (10.1%, 8.4%, and 14.8% resp. for parameters above), underscoring the high efficacy of imatinib in CML. Pts with poor response tended to have higher % of imatinib not taken over 90d, an index of overall adherence behavior. Clinicians should be aware of the association between adherence and imatinib response and should query patients about their adherence behavior. Nonadherence should be ruled out prior to classifying a patient as imatinib-resistant. Enhanced adherence is likely to optimize the effectiveness of imatinib treatment in CML.