Predictors of adherence to an on/off regimen for lenalidomide in patients with multiple myeloma.

Abstract

e19546 Background: The objective of this work was to evaluate the predictors of adherence to lenalidomide, taken in successive 21-day/7-day on/off cycles. Electronic monitoring was used to track day-by-day adherence to cycles. Methods: This study involved participants diagnosed with multiple myeloma and initiating lenalidomide at 2 hematology centers in Israel. Medication adherence was measured using the Medication Event Monitoring System (MEMS, www.aardexgroup.com). Lenalidomide was stored in bottles closed with an electronic cap, which timestamped each opening. This work focuses on how well a patient implements the dosing regimen, while they are on treatment. Implementation during on cycles was operationalized as a daily variable, equal to 1 if there was exactly 1 intake, to 0 otherwise. 13 predictors were considered and are detailed below. The influence of the predictors was studied longitudinally through one multivariate analysis. The significance level was taken at 0.05/13 = 0.0038, using the Bonferroni correction. Results: 93 subjects were included in the study, and 8 were excluded from the analysis, because they did not use the MEMS Cap as intended. Characteristics of the 85 participants are presented in Table 1. In some categories, percentages do not add up to 100 % because of missing data. Table 1 shows that a low number of pills prescribed per day, low ISS stage and high ECOG status positively influenced implementation. In addition, the quality of implementation did not decrease over time. Conclusions: Electronic monitoring of adherence allows fine-grained quantification of adherence to on/off drug regimens. This study investigated the demographic predictors of implementation adherence to an on/off regimen. A low number of pills per day positively influenced adherence, probably because of healthier patients and less complex regimens. A low ISS stage positively influenced adherence, probably also because of healthier patients. A high ECOG status positively influenced adherence, which was unexpected and could be related to the degree of help received by less functional patients. Clinical trial information: NCT02733224 . Predictor Median (Q1 – Q3) or % per category Coefficient p Time 64 (30 – 102) -0.0047 0.420 Gender Male (68 %)Female (32 %) -0.23 0.537 Age at lenalidomide initiation (years) 68 (60 – 74) -0.044 0.033 Educational level Less than high school (9 %)12 years (41 %)Academic (39 %) 0.30 0.187 Living with partner No (16 %)Yes (79 %) -0.30 0.531 Participation in a myeloma patient support group No (76 %)Yes (20 %) 0.21 0.644 Number of pills per day (for any indication) 5 (4 – 7) -0.53 0.000 ISS stage at diagnosis 1 (16 %)2 (19 %)3 (19 %) -0.99 0.000 Disease duration at initiation (months) 19 (5.75 – 45.7) -0.024 0.051 ECOG performance status scale 0 (45 %)1 (14 %)2 (6 %) 3.1 0.000 Line of myeloma therapy 1 (15 %)2 (66 %)3 (15 %) -0.22 0.749 Route for additional medications Oral (47 %)Subcutaneous (48 %) 0.64 0.233 Hospital Rabin (42 %)Sheba (58 %) 0.88 0.050