A pilot prospective study of adherence to aromatase inhibitor adjuvant therapy in patients with stage 1-3 breast carcinoma.

Introduction: The most practical approach to assessment of medication adherence in clinical practice is via patient self-reporting, which can be measured using the 8-item Morisky Medication Adherence Scale (MMAS). The 8-item MMAS was developed from a previously validated 4-item scale by supplementing additional items measuring specific medication-taking behaviors to better capture barriers surrounding adherence behavior and has been determined to have better reliability than 4-item scale. Validated medication adherence surveys such as an 8-item MMAS, are untested in oncology, but have been validated in HTN, HIV, DM and other disorders in the setting of multiple chronic medications. Thus this study will serve as a pilot study of the MMAS in the oncology population. Methods: This is a prospective study conducted at the University of Maryland Greenebaum Cancer Center (UMGCC). The primary objective was to identify the adherence level (high, medium, low) to adjuvant (AI) in early stage (1-3) breast cancer (BC) patients in order to identify predictors of adherence. Subjects were stratified according to duration of therapy with AI (≤ 2 yrs or >2 yrs) at the time of the survey. The secondary objective was to identify baseline predictors for adherence. Patients received a maximum 8 points for maximum adherence to medication treatment or high adherence, 6-7 points for medium adherence and < 6 points for low adherence. Eligibility: Adult females, postmenopausal, signed consent, hormone receptor-positive BC (ER and/or PR >=1%), pathologically confirmed Stage 1-3 BC, completed surgery, adjuvant chemotherapy, and/or radiation therapy, and were recommended by their medical oncologist to take adjuvant AI for 5 years. Pts received a prescription for adjuvant AIs (anastrozole, letrozole or exemestane) after completion of adjuvant chemotherapy. Results: Patient characteristics: 73 BC pts were sequentially accrued so far at the UMGCC and filled out the self-reported questionnaire; Causcasian-62%, African American (AAF)- 37%, Asian-1%, median age at diagnosis-54 (range33-84); stage 1-36%, stage 2-47%, stage 3-17%; mastectomy-61%; lumpectomy-39%; local radiation therapy 62%; ER+ 95%; PR+ 78%, HER2+ 17 %; tumor type ductal-75%; lobular-20%; mixed ductal/lobular-5%; tumor grade-1-46%, 2-41%,3-12%; adjuvant chemotherapy- 63%. Educational level- N/A- 3%, <high school- 11%, high school grad- 12%, some college- 34%, bachelor's degree- 18%, graduate school- 22%; marital status-single- 19%, married- 52%, divorced/separated- 19%, widowed- 10%; median number of medications- 5 (range 1-19); median number of comorbidities-3 (range 0-9). Conclusions: Overall adherence to AI treatment measured by MMAS was High in 50.7%, Medium in 35.6% and Low in 13.7% pts; there was no difference in adherence by duration of treatment with AI ≤ 2 vs. > 2 years. Univariable regression models for continuous predictors and rxc contingency tables for categorical ones revealed that the adherence to AI therapy is not affected by age, marital status, education, prior cancer therapy, tumor stage, and concurrent medication. However, there was a trend for better adherence to AI therapy in Caucasian women vs. AAF (p = 0.013). The study is ongoing. [ABSTRACT FROM AUTHOR]