Breast Cancer Research and Treatment | 2021
RE: Predictors of increased risk for early treatment non-adherence to oral anti-estrogen therapies in early stage breast cancer patients
Abstract
We read with interest a study by Yusufov et al. evaluating the association of psychological and menopause symptoms and non-adherence to AET in breast cancer patients in the recent issue of the journal [1]. The authors found that the hypothesized set of psychological and menopause symptoms at baseline (pre-AET) together statistically distinguished between those who were non-adherent (n = 19; 26.0%) from adherent (n = 54; 74.0%) at 6 weeks, and discriminant function analysis was also statistically significant (Wilks’ = 0.782, chi(2) (6) = 15.50, p = 0.017) at the 6-week timepoint. The authors concluded psychological and menopause symptoms prior to AET initiation may help to identify early treatment nonadherence. We applaud the authors for finding this interesting conclusion in the AET in breast cancer patients. However, we think that psychological and menopause symptoms can only initially predicting risk for non-adherence to AET. Here are some concrete evidences. First, in the tables 1, we found the age was different between adherent and non-adherent groups (56.9 ± 10.2 years Vs. 50.0 ± 8.6 years, p = 0.0128) at 6 weeks of AET. Although psychological and menopause symptoms were statistically different between non-adherent and adherent groups at 6 weeks, we think it’s hard to tell which is the real non-adherence factor for AET in breast cancer patients, younger age or psychological and menopause symptoms? Younger age is an inherent factor for non-adherence in some diseases. Kaesler et al. found in chronic kidney disease patients, younger age was associated with lower adherence to dietary recommendations, as well as higher body mass index, male gender, lower educational attainment and various lifestyle variables [2]. Chow et al. found in HIV patients, younger age (odds ratio [OR] 2.36, p = 0.017) was also associated with lower adherence receiving singleor multiple-tablet darunavir, cobicistat, emtricitabine, and tenofovir alafenamide [3]. According these results, we suggest the authors do a stratified analysis based on younger age and older age. Second, the authors predicted the risk for early treatment non-adherence only by psychological and menopause symptoms. In fact, adherence is affected by many factors, and a single factor is difficult to explain and predict the adherence. Cavazza et al. found that prescribed tamoxifen-only (OR 0.69; 95% CI 0.57–0.83), treated for depression (OR 0.68; 95% CI 0.60–0.78), with surgery performed in high-volume hospitals (OR 0.85; 95% CI 0.75–0.97) related with lower adherence, and continued care in the surgical hospital (OR 1.73; 95% CI 1.51–2.00), undergoing chemotherapy before AET (OR 2.65; 95% CI 2.02–3.48), and earlier AET initiation related with higher adherence in adjuvant endocrine therapy breast cancer [4]. In Xu et al.’s study, they found the type of medication, duration of medication and side effects had an impact on adherence [5]. Predicting treatment adherence based on one factor will not be accurate. In this study, pre-AET psychological and menopause symptoms only correctly classified 77.9% 6-week treatment adherence. In order to estimate accurately, we suggest that the authors should establish prediction model according to multiple related factors. Third, only 73 participants were included in the study. The sample size was small. Small sample size is lack of representativeness, and the corresponding research results will lack robustness. In summary, the authors found that psychological and menopause symptoms can predict risk for non-adherence to AET in breast cancer patients. But we think that psychological and menopause symptoms can only be regarded as a preliminary assessment tool, accurate prediction needs further study. * Xiao-Min Li [email protected]