**4. Discussion**

The majority of breast cancers are represented by hormone-receptor-positive tumors, and treatment with AET has shown grea<sup>t</sup> advantages in terms of disease recurrence and mortality. Despite these widely demonstrated benefits, AET is burdened by considerable side effects, especially in young women. Many clinical trials and epidemiological studies have shown that these side effects significantly impact quality of life and play a primary role in suboptimal adherence to AET [6–13].

Some recent studies have demonstrated that about 90% of patients on AET report side effects, which are more frequent in women taking tamoxifen than in those taking AIs [36–38]. According to these data, side effects were reported by 82% of our patients overall and, in particular, by 84% of those taking tamoxifen and 80% of those taking AIs. Moreover, the literature shows that arthralgia is reported by about 40% of patients taking AIs and 28% of those taking tamoxifen [39]. In our study, arthralgia was the most common side effect, reported by 53% of patients overall and, in particular, by 60% of those taking AIs and 32% of those taking tamoxifen. A possible explanation for this difference is the higher adherence to treatment of our patients, as arthralgia is a symptom that persists and worsens over time during AET. Hot flushes are reported by about 60–70% of patients taking AET in the literature, while in our study, they were reported by only 33% of patients. This difference may also be attributed to the higher adherence to treatment of our patients, as hot flushes tend to decrease in intensity and be more tolerated by women over time [40,41].

In our study, we evaluated the impact of GnRH agonists on the tolerance of tamoxifen and AI. It emerged that the addition of GnRH agonists to both tamoxifen and AI in high-risk premenopausal women significantly increased all analyzed side effects and worsened the tolerance to treatment overall. In particular, the most reported and less tolerated side effects among patients taking tamoxifen or AIs plus GnRH agonists, compared to those taking tamoxifen or AIs alone, were hot flushes, vaginal dryness, dyspareunia, mood changes and decreased libido. Our results are similar to those that emerged from the SOFT156 trial, which demonstrated that the addition of GnRH agonists to both tamoxifen and AIs was

related to a higher incidence of hot flushes, mood changes, vaginal dryness and decreased libido [42].

As there is consistent evidence of poorer AET tolerance in premenopausal women compared to postmenopausal ones, we evaluated side effects and tolerance in relation to the menopausal state [43]. The same results emerged from our study, as premenopausal women reported more side effects than those who were postmenopausal at the time of diagnosis (92% vs. 75%). Younger women more often reported hot flushes, vaginal dryness, decreased libido, dyspareunia and endometrial modifications.

In our study, the discontinuation rate was lower than that reported in the literature (12% vs. about 50%) [21,22,24,28]. It has to be considered that there are different methods of evaluation of adherence to AET in different studies. Self-assessment with a questionnaire or certified scales is the most used method, but in some cases, questionnaires include variables such as the percentage of tablets taken out of the total and consider as "non-adherents" those patients who take therapy irregularly or who simply report side effects [44], while we only considered non-adherent patients who definitively stopped the treatment. In fact, another study in which discontinuation is considered to be definitive suspension of treatment showed a discontinuation rate of 10%, although on a small sample of women [45].

Moreover, it emerged from our study that most of the patients who discontinued AET made this decision when extended therapy was proposed after the first five years of treatment or during extended therapy itself. The refusal of extended therapy may be explained not only by side effects, but also by the fact that patients could have perceived extended therapy as optional, without perceiving its real importance, maybe due to poor communication with the specialist. On the other hand, the discontinuation of AET during the extended therapy, after the initial acceptance, may also be explained by the worsening of the severity of side effects over time, as reported by 16% of our patients.

According to the literature, in our study, arthralgia was not only the most common side effect, but also the main reason for the discontinuation of therapy. A recent metanalysis, in fact, reported that the side effect most related to the discontinuation of AET was arthralgia, followed by weight gain and mood changes, while hot flushes, although very common, were considered physiological and generally did not lead patients to discontinue the treatment [36].

Moreover, from our analysis, it is evident that women who received adjuvant chemotherapy or radiotherapy had a lower discontinuation rate, despite a higher incidence of side effects in women who received chemotherapy. These data are in accordance with a recently published study and may be explained by a higher awareness of the severity of the disease, especially in patients who received chemotherapy, and of the importance of adjuvant treatments by these women [24].

It has been widely demonstrated that patient–physician communication plays a primary role in adherence to any medical treatment. Concerning AET, prior studies report that patients who have a referral specialist (gynecologist or oncologist) showed a higher compliance to AET than those followed up by a general practitioner, probably because the specialist can give the patients more detailed information about the importance of therapy and provide more specific medications to overcome side effects if needed, and this probably helps patients to continue the treatment [36,46,47]. In fact, the Necessity Concerns Framework (NCF) demonstrated that patients' adherence to treatment is related to their perception of the necessity and importance of treatment itself and the reduction of concerns about it more than to its side effects [48]. In fact, a recent study observed a meaningful difference in the necessity beliefs between women who accepted versus those who refused or discontinued AET, showing that women with ongoing AET intake had significantly higher trust in their treating physician and lower concerns regarding AET [35].

Concerning the evaluation of medical and psychological support, it emerged from our study that almost all of the patients taken in at our Breast Unit felt well supported. This may be attributed to a well-organized healthcare network, which allows us to take care of patients globally with regular follow-up visits and a direct communication channel

managed by a breast nurse. Patients have the possibility to refer to them with their problems and organize an appointment with the gynecologist to manage side effects or evaluate the possibility of changing their treatment. All these elements encouraged patients to talk with the gynecologist about their difficulties and side effects and try to manage them together, before discontinuing therapy by themselves. In our Breast Unit, patients are regularly monitored by a gynecologist, who is probably more aware of gynecological and sexual AET side effects and could manage them more easily than another specialist. Moreover, the gynecologist who follows up breast cancer patients in our Breast Unit is often the same person who performed their surgery, and this may result in a stronger patient–physician relationship and continuity of care, which may contribute to the improvement of patients' adherence to treatment.
