Abstract
Abstract 2754
Background:
According to the NCCN guidelines, the recommended treatment option for patients diagnosed with Philadelphia-positive (Ph+) chronic myeloid leukemia (CML) in chronic or accelerated phases who are resistant or develop intolerance to first-line imatinib is to switch to one of the two second-generation BCR-ABL tyrosine kinase inhibitors, nilotinib or dasatinib. There is limited information on their comparative adherence. Analysis of adherence may reveal whether these two drugs are being used as intended and how adherence issues may potentially affect clinical outcomes. Nonadherence to imatinib has been associated with loss of cytogenetic response, imatinib failure, and higher rates of inpatient visits. A prior study covering the period up to 6/2009 has revealed a better adherence profile for nilotinib versus dasatinib. Since dasatinib was approved for once daily use in addition to twice daily use in 05/2009, this study replicated the analysis using more recent data to compare adherence between nilotinib and dasatinib as second-line therapies in CML.
Methods:
Medical and pharmacy records from two databases from January 1997 to April 2010 were combined to identify adult patients diagnosed with CML (ICD-9 code 205.1x) with ≥1 prescription for either nilotinib or dasatinib. Patients were required to have continuous enrollment ≥1 month prior to and after the index date, defined as the first prescription for nilotinib or dasatinib, and to continue their treatment ≥1 month before discontinuing or switching to another TKI. Adherence over an up-to 180-day period was evaluated based on the proportion of days covered (PDC; sum of days of medication on hand for nilotinib or dasatinib divided by the number of calendar days in the study period). Medication possession ratio (MPR) was also evaluated in a sensitivity analysis. PDC and MPR were compared between cohorts using linear regression models. Discontinuation was defined as a treatment gap ≥30 days, and rates were compared between cohorts using Kaplan-Meier survival analyses and Cox proportional hazard regression models. In a sensitivity analysis, discontinuation was defined as a treatment gap ≥90 days. All multivariate regression models were controlled for age, gender, CML complexity, adverse events at baseline, number of inpatient days, emergency room (ER) visits and outpatient visits during baseline, medical costs during baseline, CML year of diagnosis, Charlson Comorbidity Index, and bone marrow or stem cell transplant before the index date.
Results:
Adherence and discontinuation were analyzed for a total of 558 CML patients receiving a second-line TKI (473 dasatinib and 85 nilotinib). Patient characteristics were generally similar. Patients in the dasatinib cohort had a longer mean follow-up period compared with those in the nilotinib cohort (162 days vs. 146 days; P = 0.0007) and greater utilization and costs associated with inpatient and ER services. The mean PDC over the study period for nilotinib was higher compared with dasatinib (0.81 vs 0.68). After adjusting for confounding factors, the nilotinib group had a 0.115 higher PDC (approximately 17% higher) compared to the dasatinib group (P = 0.0004). Additionally, patients in the nilotinib group had a significantly higher MPR compared to patients in the dasatinib group (0.90 vs. 0.77 P = 0.0043). Discontinuation rates (gap ≥30 days) were significantly higher for dasatinib users than for nilotinib users, with an adjusted hazard ratio of 2.15 (P = 0.0013). Dasatinib-treated patients had higher discontinuation rates than nilotinib-treated patients after 3 months (40% vs 22%) and 6 months (50% vs 28%). These unadjusted differences were robust in the sensitivity analysis with discontinuation defined as a treatment gap ≥90 days, but the difference between the cohorts was no longer significant after adjusting for confounding factors (HR 1.76, P = 0.1328).
Conclusions:
This retrospective study showed that CML patients treated with nilotinib for second-line treatment were significantly more adherent to therapy, based on PDC and MPR, and had a lower discontinuation rate than did patients receiving dasatinib in this analysis. Further studies are needed to better understand treatment-specific factors affecting adherence and persistence (e.g., treatment cost, tolerability profile, dosing convenience).
Disclosures:
Guerin: Novartis Pharmaceutical Corporation: Annie Guerin is an employee of Analysis Group, Inc, which has received research funds from Novartis Pharmaceuticals. Bollu:Novartis Pharmaceutical Corporation: Employment, Vamsi K. Bollu worked for the Novartis Pharmaceuticals Corporation while performing this analysis but is currently employed by Sunovion Pharmaceuticals Inc. Sunovion Pharmaceuticals Inc. was not in any way associated with this study. Guo:Novartis Pharmaceutical Corporation: Employment. Stepner:Novartis Pharmaceutical Corporation: Michael Stepner is an employee of Analysis Group, Inc, which has received research funds from Novartis Pharmaceuticals. Griffin:Novartis Pharmaceutical Corporation: James D. Griffin is an employee of the Dana Farber Institute which has received consultancy fees from Analysis Group, Inc, which has received research funds from Novartis Pharmaceuticals Corporation. Wu:Novartis Pharmaceutical Corporation: Eric Q. Wu is an employee of Analysis Group, Inc, which has received consultancy funds from Novartis Pharmaceutical Corporation.
Portability of Predictive Academic Performance Models: An Empirical Sensitivity Analysis