Adherence to Lenalidomide in Older Patients with Multiple Myeloma (MM): A SEER-Medicare Analysis

Introduction: Lenalidomide (Len) is an effective, convenient orally-administered therapeutic option for older adults with MM. In one recent survey of 64 patients (pts) with MM on oral therapy, 92% of respondents self-reported taking every dose of their medication. However, little objective data is available about real-world adherence to Len, or factors associated with better adherence.

Methods: Using the SEER-Medicare linked dataset, we identified pts with MM over the age of 65 who received more than 1 prescription for Len within 90 days following MM diagnosis. Adherence can be estimated from claims data using the Medication Possession Ratio (MPR) which is the number of days of drug supply dispensed in a given treatment period. We calculated a modified MPR as [Number of days supply/(last claim date - index date + last days' supply) x100], adjusted for 7 days off therapy per cycle. Because most Len-containing regimens incorporate a 7 day break, but not all claims reflect this (e.g. 21 pills were dispensed for a 21 day supply, but with a 7 day delay in the next dispensation), a "grace period" of 7 days was calculated in these instances in order to avoid underestimation of adherence. Discontinuation of Len treatment was assumed if there was a 70 day gap in dispensations (corresponding to a 6 week delay following next anticipated dispensation). We investigated factors associated with the MPR using linear regression; p-values <0.05 were considered significant. Candidate predictors included demographics, performance status indicators, Charlson comorbidity index (CCI), cognitive impairment, depression, number of concomitant meds, socioeconomic variables, and cancer center characteristics. A MPR <80% was considered poor adherence.

Results: 793 patients were eligible for analysis. The median age was 73 ± standard deviation (SD) 6.4. 52% were male; 81% white, 13% black & 6% other. The median CCI was 1 ± SD 1.6. 4% had a diagnosis of dementia. 5% had a diagnosis of depression. The median number of unique prescriptions in addition to Len was 10 ± SD 5.1.

Only 14.5% of patients were considered to have poor adherence, with an MPR <80%. The median MPR was 92 ± 9.3%. Analyzing MPR as a continuous variable, factors significantly associated with MPR on linear regression included: black race (Beta -2.4, p=0.023); age (Beta -0.15, p=0.042); Medicaid enrollment (Beta -1.74, p=0.041); number of concomitant meds (Beta -0.19, p= 0.0078). Factors that were not significantly associated with MPR included gender, performance status indicators, overall CCI score, depression, cognition, cancer center characteristics (i.e. academic vs community; National Cancer Institute designated comprehensive cancer centers vs other) or low-income subsidies.

Conclusions: We have demonstrated overall high rates of medication possession among older adults with MM. Only 14.5% of patients met the commonly applied definition for poor adherence of an MPR <80%. Factors associated with lower MPR included black race, increasing age, Medicaid coinsurance, and number of additional medications. Limitations of this study include that medication possession does not guarantee actual adherence, thus the MPR may overestimate of adherence. Conversely, we are unable to ascertain whether delays in dispensation actually reflect treatment breaks for toxicity. Future analyses will endeavor to distinguish delays in dispensations for toxicity from poorer adherence.