Treatment Adherence in Patients with Chronic Myeloid Leukemia in a Real World Setting

Understanding adherence in oral CML therapy is an important part of selecting and managing long-term CML treatment. Literature indicates that imatinib adherence ranges widely (14–98%), and data for other BCR-ABL inhibitors (dasatinib and nilotinib) are currently sparse. Imatinib was approved as first-line therapy in the USA in 2001. Dasatinib and nilotinib were approved as first-line therapy at the end of 2010, leading to limited adherence data for this indication. Both dasatinib and nilotinib were approved for second-line therapy earlier in time (dasatinib approved in 2006, nilotinib approved in 2007). As survival rates improve for patients with CML and as multiple frontline therapeutic options exist, it is important to understand medication adherence patterns as part of long-term therapy.

Using the HealthCore Integrated Research Database (HIRD™), we identified patients with ≥1 International Classification of Diseases (9th edition) code for CML (205.1x) and ≥1 prescription for a BCR-ABL inhibitor dispensed 1/1/2001–6/30/2010. We used medication possession ratio (MPR; number of days supply of current prescription divided by total days between current and next prescription) to calculate adherence to treatment. Cox proportional hazard models were used to quantify rates of poor adherence (MPR <85%) comparing nilotinib to dasatinib. Models were adjusted for baseline characteristics, previous imatinib exposure, concomitant medications, and comorbidities.

We identified 2,145 patients with a CML diagnosis exposed to a BCR-ABL inhibitor from 2001 to 2010. Among these 2,145 patients, 2,064 received imatinib as first-line therapy, 65 received dasatinib first-line and 16 received nilotinib first-line during this time period. Among the 2,064 first-line imatinib users, 197 received dasatinib and 53 received nilotinib as second-line therapy. Sample size was too small to evaluate adherence in first-line dasatinib and nilotinib users in the current dataset. Among second-line users, mean exposure to dasatinib was 276 days (≤100 mg/day, 275 days; ≥140 mg/day, 276 days) and 170 days for nilotinib. Adjusted Cox proportional hazard ratios comparing poor adherence in nilotinib vs. dasatinib were 1.6 (95% confidence interval [CI] 1.0–2.4) overall, 1.9 (95% CI 1.2–3.0) for nilotinib vs. dasatinib ≤100 mg/day, and 1.2 (95% CI 0.7–2.0) for nilotinib vs. dasatinib ≥140 mg/day.

When comparing treatment adherence for second-line CML therapy, overall, patients treated with nilotinb were 60% more likely to have poor adherence than patients receiving dasatinib. Sample size was too small to adequately examine adherence among first-line users. However, the cohort is being extended beyond 2010 and analyses are underway to assess adherence in first-line therapies.

Off Label Use: Although dasatinib and nilotinib are now approved for first-line therapy, some patients in our study were prescribed dasatinib or nilotinib before they were approved as first line therapy. In our abstract, we report the number of users who may have been prescribed dasatinib or nilotinib as first-line therapy prior to approval of dasatinib or nilotinib as first-line therapy. Additional analyses may be reported. Hirji:BMS: Employment. Davis:BMS: Employment.