Duloxetine and Escitalopram for Hot Flushes: Efficacy and Compliance in Breast Cancer Survivors

Original Article

Duloxetine and escitalopram for hot ﬂushes: efﬁcacy and compliance in breast cancer survivors

N. BIGLIA, MD, PHD, Gynaecology and Obstetrics Unit, Umberto I Hospital, Department of Surgical Sciences, University of Turin, Turin, V.E. BOUNOUS, MD, Gynaecology and Obstetrics Unit, Umberto I Hospital, Department of Surgical Sciences, University of Turin, Turin, T. SUSINI, MD, PHD, Breast Unit Department of Health Science, OB & GYN Section, AOU Careggi, School of Medicine, University of Florence, Florence, S. PECCHIO, MD, Gynaecology and Obstetrics Unit, Umberto I Hospital, Department of Surgical Sciences, University of Turin, Turin, L.G. SGRO, MD, PHD, Gynaecology and Obstetrics Unit, Umberto I Hospital, Department of Surgical Sciences, University of Turin, Turin, V. TUNINETTI, Gynaecology and Obstetrics Unit, Umberto I Hospital, Department of Surgical Sciences, University of Turin, Turin, & R. TORTA, MD, PHD, Psycho-Oncology Unit, Department of Neurosciences, University of Turin, Turin, Italy

BIGLIA N., BOUNOUS V.E., SUSINI T., PECCHIO S., SGRO L.G., TUNINETTI V. & TORTA R. (2016) Euro- pean Journal of Cancer Care Duloxetine and escitalopram for hot ﬂushes: efﬁcacy and compliance in breast cancer survivors

Selective serotonin reuptake inhibitors (SSRI) and serotonin–norepinephrine reuptake inhibitors (SNRI) might be an effective treatment for hot flushes (HFs) in breast cancer survivors (BCSs). This study aims to compare the efficacy and tolerability of duloxetine (SNRI) versus escitalopram (SSRI) in reducing frequency and severity of HFs in BCSs and to assess the effect on depression. Thirty-four symptomatic BCSs with emotional impairment received randomly duloxetine 60 mg daily or escitalopram 20 mg daily for 12 weeks. Patients were asked to record in a diary HF frequency and severity at baseline and after 4 and 12 weeks of treatment. Depression was evaluated through validated questionnaires (Beck Depression Inventory and Montgomery Asberg Depression Rating Scale) at baseline and after 4 and 12 weeks of treatment. Both drugs showed a significant reduction of HF frequency and severity after 12 weeks of treatment with no significant difference between the two groups. A significant improvement in depression symptoms was observed at the end of the study period within both the groups, without difference between the two drugs. In conclusion, escitalopram and duloxetine are both effective treatment for the relief of HFs in BCSs, with similar beneficial effect. A significant improvement of depression was obtained with no major side effects.

Keywords: breast cancer, hot ﬂushes, duloxetine, escitalopram, serotonin–norepinephrine reuptake inhibitors, selective serotonin reuptake inhibitors.

ment in treatment. Every year, an increasing number of INTRODUCTION new cancers are diagnosed among women in reproduc- The number of breast cancer survivors (BCSs) is increas- tive age. The young women may experience premature ing over time, due to early detection and to improve- menopause, due to adjuvant chemotherapy and/or to ovarian suppression and menopausal symptoms are typi- cally more severe in young patients, due to the abrupt Correspondence address: Nicoletta Biglia, Department of Surgical Sciences, Gynaecology and Obstetrics Unit, “Umberto I” Hospital, Largo depletion of oestrogens (Knobf 2006). Despite the well- Turati 62, 10128 Turin, Italy (e-mail: [email protected]). established efﬁcacy of adjuvant treatments, up to 20% Accepted 9 February 2016 of breast cancer patients consider stopping or actually DOI: 10.1111/ecc.12484 cease endocrine therapy because of menopausal symp- European Journal of Cancer Care, 2016 toms (Hickey et al. 2008).

Hot flushes (HFs) are the most frequently reported women. In healthy women, escitalopram has been shown menopausal symptom, affecting around 65% of BCSs to be superior to other SSRIs in terms of efficacy for HFs (Biglia et al. 2003; Antoine et al. 2008). Vasomotor symp- (Shams et al. 2014). toms (HFs and night sweats) are related to estradiol deple- Among SNRIs, previous studies on venlafaxine tion both in the serum and in the hypothalamic (Loprinzi et al. 2000; Biglia et al. 2005; Evans et al. 2005; temperature regulating centre. Furthermore, the alteration Carpenter et al. 2007; Boekhout et al. 2011) and its active of serotonergic and noradrenergic activity due to oestrogen metabolite O-desmethylvenlafaxine (Speroff et al. 2008; deprivation determines mood swings, depression or Archer et al. 2009; Pinkerton et al. 2013) demonstrated somatic complaints (Freedman 2001). HFs, as well as night their efficacy in reducing the number and intensity of sweats, may disrupt sleep, leading to loss of energy and HFs. Data from the few studies performed on duloxetine performance, contributing to mood depression and with a showed a benefit in reducing number and severity of HFs consequent negative impact on quality of life (Conde et al. (Joffe et al. 2007; Henry et al. 2011; Freeman et al. 2013). 2005; Gupta et al. 2006; Nappi & Kokot-Kierepa 2010). In a pilot study, duloxetine was effective in decreasing Furthermore, in BCSs, the awareness of having had cancer musculoskeletal symptoms in breast cancer patients trea- can determine emotional impairment. ted with aromatase inhibitors as well (Henry et al. 2011). Systemic oestrogen supplementation is the most effec- Although evidence supports SSRIs and SNRIs efficacy in tive strategy in reducing menopausal symptoms in reducing HFs (Carroll & Kelley 2009; Handley & Williams healthy women but unfortunately it is contraindicated in 2015), only paroxetine has been recently approved by the BCSs (International Menopause Society 2013; The North Food and Drug Administration (FDA) for the treatment of American Menopause Society (NAMS), 2012). The safety moderate-to-severe vasomotor symptoms associated with of systemic oestrogens in BCSs has been studied and dee- menopause (Orleans et al. 2014). ply debated. The Stockholm trial (Von Schoultz & Rutq- The majority of the studies evaluating the role of SSRIs vist 2005) was prematurely stopped in 2003 when the and SNRIs on vasomotor symptoms were performed on parallel HABITS trial (Holmberg & Anderson 2004) healthy women and only few of them were performed on a reported a higher recurrence rate in breast cancer patients population of BCSs (Loprinzi et al. 2000; Biglia et al. who received systemic oestrogens compared to women 2005; Carpenter et al. 2007; Boekhout et al. 2011; Henry treated with placebo. At 4 years of follow-up, the HABITS et al. 2011). study still found an increased risk of recurrence (Holmberg Furthermore, to the best of our knowledge, so far no et al. 2008), whereas a recent updated analysis of the data study was carried out comparing SSRIs and a SNRIs effi- from the Stockholm trial at 10.8 years of follow-up did not cacy on HF treatment, neither in the general population show any excess of recurrence risk (Fahlen et al. 2013). nor in women with breast cancer. In this study we com- An alternative compound to conventional oestrogen/ pared, as primary outcome, the efficacy of two antidepres- progestogen treatment, tibolone was tested versus placebo sant drugs, a SSRI (escitalopram) versus a SNRI in the LIBERATE trial, showing a significant superiority (duloxetine), in reducing the frequency and severity of to placebo in reducing vasomotor symptoms and improv- vasomotor symptoms in post-menopausal BCSs s with ing sleep quality, sexual behaviour, mood and attraction depressive mood and as secondary outcome we evaluated (Sismondi et al. 2011), but unfortunately also a significant the effect of these antidepressants on mood depression. increase in the recurrence rate (Kenemans et al. 2009). Open and controlled trials demonstrated that selective MATERIALS AND METHODS serotonin reuptake inhibitors (SSRIs), as well as selective serotonin–norepinephrine reuptake inhibitors (SNRIs) are Fifty-eight breast cancer patients with troublesome HFs effective non-hormonal alternatives for the treatment of (>20/week), emotional impairment and with no evidence vasomotor symptoms (Sassarini & Lumsden 2013; ACOG of disease attending the outpatient clinic for menopause 2014). symptoms entered the study. They were randomly allo- Among SSRIs, paroxetine (Stearns et al. 2003, 2005; cated to receive either duloxetine oral tablets (30 mg/day Soares et al. 2008; Simon et al. 2013), sertraline (Gordon in the morning during the first week and then 60 mg/day) et al. 2006), fluoxetine (Loprinzi et al. 2002), citalopram or escitalopram (10 mg/day in the morning for the first (Kalay et al. 2007; Barton et al. 2010) and escitalopram week and then 20 mg/day) for 12 weeks. (Soares et al. 2006; Defronzo Dobkin et al. 2009; Freeman The study design had been approved by the local ethic et al. 2011) have been studied, showing benefit in reduc- committee (approval number 0047015) and all patients ing the number and intensity of HFs in menopausal gave their written informed consent before participating.

Patients were provided with a diary and asked to record ried out in order to compare the emotional assessments HF frequency and severity, at baseline (T0) (the first week (T0–T1–T2) (evaluated through MADRS and BDI score). without treatment), after 4 (T1) and 12 weeks (T2) of treat- Since data from literature show that placebo can deter- ment. In the HF diary, the patient recorded daily HF num- mine a reduction in HF frequency and score up to 40%, we ber and severity (mild, moderate, severe and very severe considered as effective a further reduction given by any of HFs). The HF score was calculated by scoring each individ- the two drugs on study. ual HF (one point for a mild HF, two points for a moderate Statistical significance was determined by using an HF, three points for a severe HF, and four points for a very alpha level of 0.05 and two-sided tests. severe HF). This type of questionnaire was already used and validated in previous trials enquiring about HFs (Sloan RESULTS et al. 2001). Furthermore, in the HF diary the patient had to record treatment related side effects. Of the 58 enrolled patients, 28 were allocated to dulox- Mood depression was measured at three different times: etine and 30 to escitalopram. During the first week, 12 T0, T1 and T2 by a self-evaluating (Beck Depression patients in the duloxetine group (42.8%) and 12 patients in Inventory, BDI) (Beck et al. 1998) and by a hetero-evalu- the escitalopram group (40%) decided to withdraw from ated rating scale (Montgomery Asberg Depression Rating the trial and never began the therapy. Of the remaining 34 Scale, MADRS; Montgomery & Asberg 1979). Remission women who started the treatment, three patients in the of depression was defined as MADRS scores less than 10. duloxetine group (18.7%) and three in the escitalopram Duloxetine is a SNRI approved by the FDA for treat- group (16.7%) dropped out for side effects after 4 weeks of ment of major depressive disorder (Goldstein et al. 2004), treatment. fibromyalgia, diabetic peripheral neuropathic pain and Baseline characteristics of patients are shown in Table 1 chronic musculoskeletal pain (Arnold et al. 2007; Pritch- and were compared using t-test. No significant differences ett et al. 2007; Chappell et al. 2009; Skljarevski et al. were seen between the two groups and they are homoge- 2010). neous. All subjects enrolled into the study had trouble- Escitalopram is the most selective among SSRIs, with some HFs, early breast cancer with no documented proven efficacy in treating major depressive disorder and metastasis and concomitant emotional impairment, with onset of antidepressant activity evident within 1–2 weeks MADRS total scores greater than 12 at study entry. A of treatment initiation (Signorovitch et al. 2011). It seems that escitalopram is one of the best treatments for depression in terms of efficacy and acceptability, with low rates Table 1. Baseline characteristics of patients of discontinuation due to adverse (Bielski et al. 2004; Duloxetine Escitalopram Montgomery et al. 2004; Khan et al. 2007; Nierenberg (n = 28) (n = 30) P value et al. 2007; Wade et al. 2007; Cipriani et al. 2009; Shams Age et al. 2014). Median (range) 45 (40–55) 46 (39–56) NS <50 years 20 (71.4%) 18 (60%) NS >50 years 8 (28.6%) 12 (40%) NS Statistical analysis Menopausal status Spontaneous 7 (25%) 14 (46.7%) NS Pharmacological 17 (60.7%) 14 (46.7%) NS Statistical analysis was carried out using SPSS software Surgical 4 (14.3%) 2 (6.6%) NS version 17 (SPSS Inc., Chicago, IL, USA). Concurrent hormone therapy As primary outcome, we compared the efficacy of esci- Tamoxifen 7 (25%) 8 (26.7%) NS + talopram and duloxetine in reducing the frequency of HFs Tamoxifen GnRh 17 (60.7%) 14 (46.7%) NS analogues (the number of HFs reported per day) and of HF score from Aromatase inhibitors 4 (14.3%) 6 (20%) NS baseline (T0) to 4 (T1) and to 12 (T2) weeks of treatment. None – 2 (6.6%) NS As a secondary outcome we compared the effect of esci- HF >9 months 16 (57.1%) 17 (56.7%) NS Daily HF frequency 13 12 NS talopram and duloxetine on mood depression, in terms of (mean) MADRS score and BDI score. Weekly HF frequency 90 82 NS Dependent samples t-test was used to analyse data (HF (mean) Weekly HF score (mean) 244 197 NS frequency and score) and Shapiro–Wilk test to confirm Pretreatment MADRS 12.9 19.4 NS that the sample is from a normal population. When nor- score (mean) mality of data was not confirmed, the Wilcoxon signed NS, not significant; HF, hot flushes; MADRS, Montgomery rank test was used. Repeated ANOVA measures were car- Asberg Depression Rating Scale.

non-significant higher basal level of depression was Table 3. Mean MADRS and BDI score modification since observed in the escitalopram group while a non-significant baseline (T0) to 4 (T1) and 12 weeks (T2) of treatment higher baseline number of HFs per week was seen in the Duloxetine Escitalopram duloxetine group. Most of the patients in both the groups MADRS score (mean) were under tamoxifen and GnRh analogues concurrent T0 12.9 19.4 T1 7.6 12.6 treatment and complained of vasomotor symptoms since T2 5.6 (56.6%; P < 0.05) 11.1 (42.8%; P < 0.005) more than 9 months. BDI score (mean) Both drugs were effective in reducing the frequency and T0 4.9 8.3 T1 3.7 7.39 severity of HFs (Table 2) (t-test analysis). T2 3.6 (26.5%; P < 0.05) 6.6 (20.5%; P < 0.005) The beneficial effect was fast with both drugs, showing MADRS, Montgomery Asberg Depression Rating Scale; BDI, a significant benefit already after 1 month of treatment in Beck Depression Inventory. reducing the number of HFs per week and their severity, as shown by the significant reduction in the HF score. both drugs were nausea, mouth dryness, weakness and The decrease, after 12 weeks of treatment, in the total drowsiness. Side effects were limited, in most of the number of HFs per week was 49.8% in the duloxetine patients, to the first week of treatment disappearing after- group (P = 0.003) and 53% in the escitalopram group wards. However, the side effects led to the dropout of three (P = 0.001). patients in the duloxetine group (18.7%) and of three The decrease, at the end of the study period, in the patients in the escitalopram group (16.7%) during the first weekly HF score was 53.6% in the duloxetine group weeks of treatment. In the duloxetine group, nausea was (P = 0.003) and 60.4% in the escitalopram group reported by four patients (25%); weakness and drowsiness (P = 0.001). by 3 (18.7%); insomnia, xerostomia (dry mouth) by 2 A significant reduction of the total number of HFs per (12.5%); and one patient reported constipation (6.1%). week and in the weekly HF score was seen both in the Nausea and weakness were reported by four patients duloxetine group and in the escitalopram group, with no (22.2%) employing escitalopram, and xerostomia and difference between the two groups (P = NS). drowsiness by two patients (11.1%). One patient (5.5%) The same results were found for the effects on mood reported insomnia and anxiety. depression: both drugs significantly reduced MADRS and BDI scores during the study period (Table 3) (repeated DISCUSSION ANOVA analysis). After 12 weeks of treatment, in the escitalopram group, there was a 42.8% MADRS score The management of menopausal symptoms and the reduction (P < 0.005) and a BDI score reduction of 20.5% related decrease of quality of life in BCSs s is still an (P < 0.005). Duloxetine determined a 56.6% reduction in unsolved problem. Vasomotor symptoms, such as HFs and the MADRS score (P < 0.05) and a 26.5% reduction in the night sweats, are the most frequently reported symptoms BDI score (P < 0.05) at the end of the study period. (Grady 2006) and their pathophysiology is still under dis- No difference in the efficacy on depression was observed cussion. Oestrogens modulate serotonin and noradrenalin between the groups treated with duloxetine or escitalo- activity in the brain and these neurotransmitters partici- pram (P = NS). pate to the hypothalamic thermoregulation (Boulant 2000; Both drugs were well tolerated and no major adverse Romanovsky 2007; Rossmanith & Ruebberdt 2009; Vilar- event was recorded among those patients who received Gonzalez et al. 2011). Serum serotonin concentrations in the drug. The most frequently reported side effects for post-menopausal women are inversely related to the intensity of climacteric symptoms (Słopien et al. 2003) Table 2. Mean hot flushes (HF) activity baseline (T0) and after and related to vasodilatation (Watts & Davis 2011). 4 (T1) and 12 weeks (T2) of treatment Oestrogens increase the availability of serotonin in the Duloxetine Escitalopram synaptic cleft by increasing the synthesis of the transmit- Weekly HF frequency (mean) ter, slowing its degradation and delaying the reuptake of T0 89.6 82.5 T1 39.7 (56%; P = 0.001) 46.2 (48.3%; P < 0.001) the transmitter (Lu & Bethea 2002; Serva et al. 2002; Le T2 43 (49.8%; P = 0.003) 39 (53%; P = 0.001) Saux & Di 2005). Weekly HF score (mean) Hormone replacement therapy can alleviate these T0 244.7 197.6 symptoms, but it is contraindicated in BCSs, as it can T1 86.6 (62.6%; P = 0.001) 97 (53.9%; P < 0.001) increase the risk of recurrence (NAMS 2012). Further- T2 98.7 (53.6%; P = 0.003) 80 (60.4%; P = 0.001) more, it is known that HFs are more likely to occur and to

be experienced as severe in women with moderate or sev- vasomotor symptoms relapsed after 3 weeks from the dis- ere depressive symptoms, compared to women with no or continuation of the therapy and that the relapse occurred mild mood symptoms (Reed et al. 2009). more commonly in those with higher levels of pretreat- Antidepressants are an optimal alternative to hormone ment insomnia symptoms and in those who had a weaker treatment, since they have the ability to treat both these response to escitalopram. symptoms. Few studies have been published on the potential activ- Selective serotonin reuptake inhibitor and SNRI antide- ity of duloxetine in reducing HFs. Duloxetine (60– pressants have been successfully used in the treatment of 120 mg/day) for 8 weeks, in healthy menopausal women vasomotor symptoms, reducing HFs by up to 65% (Nelson not responding to placebo, significantly decreased et al. 2006; Carroll & Kelley 2009; ACOG 2014; Handley MADRS score and vasomotor symptoms (Joffe et al. and Williams 2015). 2007). A recent extension of this study evaluating also the A selective serotonergic mechanism or a dual one (con- impact of 8 weeks of duloxetine on both day and night comitant serotonergic and noradrenergic) can theoreti- HFs, showed a significant improvement of depression, cally lead to a different vasomotor response in central and anxiety and daytime, but not night-time HFs (Freeman peripheral thermoregulation. et al. 2013). The only study of duloxetine in BCSs was per- Most of the studies on SSRIs have been carried out using formed on 29 women treated with aromatase inhibitors paroxetine (Stearns et al. 2003, 2005; Soares et al. 2008; and it was primarily targeted on pain. Among the sec- Simon et al. 2013). Only few and inconsistent data have ondary outcomes, there was the HF interference with been published on escitalopram, a last generation SSRI, daily life that significantly improved after 8 weeks of with a more selective and potent action on serotonergic treatment (Henry et al. 2011). Our study confirms the effi- reuptake (Soares et al. 2006; Defronzo Dobkin et al. 2009; cacy of duloxetine in treating menopausal vasomotor Freedman et al. 2011; Freeman et al. 2011). symptoms in breast cancer patients, since it significantly At our knowledge, until now, no study evaluating the reduced weekly HF frequency by 56% after 4 weeks of role of escitalopram in reducing HFs was performed in treatment and by 49.8% after 12 weeks. Also the severity BCSs. Our results are consistent with the previous find- of HFs was improved, as shown by the mean reduction of ings on escitalopram efficacy in improving menopausal weekly HF score by 62.6% after 4 weeks and by 53.6% vasomotor symptoms on healthy women (Soares et al. after 12 weeks of use. 2006; Defronzo Dobkin et al. 2009; Freeman et al. 2011), Several studies have been performed in order to compare but applied to BCSs, many of them receiving tamoxifene, SSRIs and SNRIs in the treatment of major depressive dis- with severe HFs. In our study, after 4 weeks of treatment, orders (Bielski et al. 2004; Montgomery et al. 2004; Khan escitalopram significantly reduced weekly HF frequency et al. 2007; Nierenberg et al. 2007; Wade et al. 2007; and score respectively by 48.3% and 53.9%. After Soares et al. 2010; Bose et al. 2012). Escitalopram seems 12 weeks, a further reduction was observed both for to be at least as effective (Khan et al. 2007) or superior weekly HF frequency (53%) and score (60.4%). (Wade et al. 2007) to duloxetine in acute treatment of Also in the study of Freeman et al. (2011), escitalopram major depressive disorder (Wade et al. 2007; Bose et al. (10–20 mg/day) for 8 weeks significantly decreased HF fre- 2012), but better tolerated in long-term use (Khan et al. quency and severity compared to placebo; moreover, also 2007; Wade et al. 2007). No difference has been shown HF interference in daily life (Carpenter et al. 2012), pain when comparing escitalopram to desvenlafaxine in the and quality of life (LaCroix et al. 2012) are significantly treatment of major depressive disorder (Soares et al. 2010). improved by escitalopram. Similar response to escitalo- This is the first study directly comparing the efficacy of pram on HFs was found by Defronzo Dobkin et al. (2009) a SSRI versus a SNRI in the treatment of vasomotor meno- in menopausal women without mood depression. Con- pausal symptoms in breast cancer patients. On the basis of versely, another placebo-controlled study (Freedman et al. our preliminary data, both duloxetine and escitalopram 2011) found no significant decrease in HF frequency with are effective in reducing HF frequency and severity after 4 either 10 or 20 mg/day of escitalopram. and 12 weeks of treatment, with no significant differences It has been proven that vasomotor menopausal symp- between the two drugs. The improvement is fast with toms recur after discontinuation of hormone therapy both drugs, leading to a mean weekly HF frequency reduc- (Brunner et al. 2010), but whether they recur after antide- tion from baseline of around 50% in the first month of pressants discontinuation is still unknown. Joffe et al. treatment (56% in the duloxetine group and 48.3% in (2013) demonstrated that in approximately one-third of the escitalopram group at T1); a small further improve- healthy symptomatic women responding to escitalopram, ment was observed only with escitalopram (53% at T2).

Also the severity of HFs significantly improved after patients in our study (nausea, weakness, drowsiness, 4 weeks of treatment with both drugs (mean weekly HF insomnia, xerostomia and constipation) (Arnold et al. score 62.6% in the duloxetine group and 53.9% in the 2007; Pritchett et al. 2007; Wade et al. 2007; Chappell escitalopram group at T1); a small additional improve- et al. 2009; Skljarevski et al. 2010; Henry et al. 2011; ment was observed after 3 months in the group treated Freeman et al. 2013). Consistent with the findings of Free- with escitalopram (60.4% at T2). man et al. (2011), nausea, weakness and drowsiness were As regards mood depression, both drugs administered the more frequently reported side effects by patients under for 12 weeks, significantly reduced MADRS and BDI escitalopram treatment. In a recent systematic review scores, with no difference between the duloxetine and the analysing 18 trials, escitalopram seems to be the antide- escitalopram group. pressant with the lower rate of side effects (Handley and In the comparison study of Wade et al. (2007), on the Williams 2015). Furthermore, new users of escitalopram contrary, escitalopram was superior to duloxetine in are reported to have a the highest number of days of unin- reducing MADRS total score from baseline to weeks 1, 2, terrupted treatment, the better adherence and the lower 4, 8, 12 and 16. Also in the study of Bose et al. (2012), proportion of switching, than other antidepressants treatment with escitalopram compared with duloxetine (Esposito et al. 2009; Wu et al. 2009; Aguglia et al. 2012). resulted in a significant greater improvement in MADRS For women treated with tamoxifen, there may be a pref- total score at the end of week 8. In our study, after erence to avoid some SSRIs and SNRIs that are potent 12 weeks of treatment, MADRS score was significantly inhibitors of CYP2D6 enzyme with a consequent decrease reduced by 56.6% in the duloxetine group, with complete in the efficacy of tamoxifen. Among SSRIs, citalopram and remission of depressive symptoms (MADRS 5.6 at T2). escitalopram have a mild effect and can be given with These data are consistent with the findings of Joffe et al. tamoxifen, unlike paroxetine and fluoxetine, which both (2007). In line with the results from Khan et al. (2007), a have large effect (Handley and Williams 2015). As regards significant reduction of MADRS score was observed in the SNRIs venlafaxine and desvenlafaxine are the safest escitalopram group as well. choices for tamoxifen users, whereas duloxetine has a It is reported in literature that the dropout rates from moderate effect (Freeman et al. 2013). antidepressant treatment are 50% at 3 months and >70% The strength of this study is being the first study at 6 months and that 15–45% of new users of antidepres- directly comparing the efficacy of a SSRI versus a SNRI in sants give up with the treatment within the first 2– HF treatment in breast cancer patients. In addition, at our 3 months since initiation. Early discontinuation of phar- knowledge, until now no study was carried out comparing macological treatment is generally due to low clinical effi- SSRIs and a SNRIs efficacy on HF treatment, neither on cacy, long time to onset of action, fear of pharmacological the general population nor in women with breast cancer. interactions and unfavourable safety profile (Aguglia et al. Furthermore, in this study, a strict methodology has been 2012). used to evaluate HF activity and the 12-week length of the In our study, we observed a high proportion of women study respected FDA requirements for HF trials duration (24 out of 58; 41%) who decided not to start the treatment (Guttuso & Evans 2010; Handley and Williams 2015). on the week after randomisation. This observation Limitations of the study are the small sample size and the deserves special attention. Before randomisation, women lack of placebo control group. were fully informed on the drugs used and signed the con- Since comparison of outcomes between different clini- sensus form. The most frequently reported reasons to cal trials is not recommended given the heterogeneity of withdraw were fear of psychotropic drugs, perplexity of patient populations, assessment tools used and outcome general practitioner and criticism by friends. In the study data reported, we wanted to directly compare escitalopram of Joffe et al. (2007), of 30 eligible women, only 20 initi- and duloxetine. In literature a great amount of previous ated treatment with duloxetine and 14 completed the studies have shown that SSRIs/SNRIs reduce HFs, but study period of 8 weeks. although these agents appear to have similar effectiveness, In our study, both duloxetine and escitalopram were no direct comparison between the two groups in BCSs was well tolerated and only minor side effects were recorded. done until now. Most of the side effects disappeared after the first week of treatment and led to the dropout of 18.7% of patients in CONCLUSION the duloxetine group and of 16.7% in the escitalopram group. The more frequently reported side effects of dulox- In conclusion, our data show that a 12-week treatment etine through literature are the same experienced by the both with escitalopram and duloxetine is effective on

reducing the frequency and severity of HFs in BCSs. Fur- longer term treatment and the time to relapse after antide- thermore, a significant improvement of mood depression pressants discontinuation. was obtained with both drugs with acceptable side effects. Both drugs can be recommended for the treatment of HFs CONFLICT OF INTEREST in patients previously treated for breast cancer. Further studies are needed in order to understand the efficacy of The authors declare that they have no conflict of interest.

REFERENCES breast cancer: a survey on breast cancer Windschitl H. (2009) Duloxetine, a survivors. Maturitas 45,29–38. centrally acting analgesic, in the ACOG (2014) ACOG Practice Bulletin Biglia N., Torta R., Roagna R., treatment of patients with osteoarthritis No. 141. Management of menopausal Maggiorotto F., Cacciari F., Ponzone R., knee pain: a 13-week, randomized, symptoms. Obstetrics and Gynecology Kubatzki F. & Sismondi P. (2005) placebo controlled trial. Pain 146, 253– 123, 202–216 Evaluation of low-dose venlafaxine 260. Aguglia E., Ravasio R., Simonetti M., hydrochloride for the therapy of hot Cipriani A., Furukawa T.A., Salanti G., Pecchioli S. & Mazzoleni F. (2012) Use flushes in breast cancer survivors. Geddes J.R., Higgins J.P., Churchill R., and treatment modalities for SSRI and Maturitas 52,78–85. Watanabe N., Nakagawa A., Omori SNRI antidepressants in Italy during the Boekhout A.H., Vincent A.D., Dalesio I.M., McGuire H., Tansella M. & Barbui period 2003-2009. Current Medical O.B., van den Bosch J., Foekema-Tons€ C. (2009) Comparative efficacy and Research and Opinion 28, 1475–1484. J.H., Adriaansz S., Sprangers S., Nuijen acceptability of 12 new-generation Antoine C., Vandromme J., Fastrez M., B., Beijnen J.H. & Schellens J.H. (2011) antidepressants: a multiple-treatments Carly B., Liebens F. & Rozenberg S. Management of hot flashes in patients metaanalysis. The Lancet 373, 746–758. (2008) A survey among breast cancer who have breast cancer with Conde D.M., Pinto-Neto A.M., Cabello C., survivors: treatment of the climacteric venlafaxine and clonidine: a Sa0 D.S., Costa-Paiva L. & Martinez E.Z. after breast cancer. Climacteric 11, 322– randomized, double-blind, placebo- (2005) Menopause symptoms and 328. controlled trial. Journal of Clinical quality of life in women aged 45 to Archer D.F., Seidman L., Constantine Oncology 29, 3862–3868. 65 years with and without breast G.D., Pickar J.H. & Olivier S (2009) A Bose A., Tsai J. & Li D. (2012) Early non- cancer. Menopause 12, 436–443. double-blind, randomly assigned, response in patients with severe Defronzo Dobkin R., Menza M., Allen placebo controlled study of depression: escitalopram up-titration L.A., Marin H., Bienfait K.L., Tiu J. & desvenlafaxine efficacy and safety for versus switch to duloxetine. Clinical Howarth J. (2009) Escitalopram reduces the treatment of vasomotor symptoms Drug Investigation 32, 373–385. hot flashes in nondepressed menopausal associated with menopause. American Boulant J.A. (2000) Role of the preoptic- women: a pilot study. Annals of Journal of Obstetrics and Gynecology anterior hypothalamus in Clinical Psychiatry 21,70–76. 200, 172, e1-10. thermoregulation and fever. Clinical Esposito D., Wahl P., Daniel G., Stoto Arnold L.M., Pritchett Y.L., D’Souza D.N., Infectious Diseases 31, 157–161. M.A., Erder M.H. & Croghan T.W. Kajdasz D.K., Iyengar S. & Wernicke J.F. Brunner R.L., Aragaki A., Barnabei V., (2009) Results of a retrospective claims (2007) Duloxetine for the treatment of Cochrane B.B., Gass M., Hendrix S., database analysis of differences in fibromyalgia in women: pooled results Lane D., Ockene J., Woods N.F., antidepressant treatment persistence from 2 randomized, placebocontrolled Yasmeen S. & Stefanick M. (2010) associated with escitalopram and other clinical trials. Journal of Women’s Menopausal symptom experience before selective serotonin reuptake inhibitors Health 16, 1145–1156. and after stopping estrogen therapy in in the United States. Clinical Barton D.L., Lavasseur B.I., Sloan J.A., the Women’s Health Initiative Therapeutics 31, 644–656. Stawis A.N., Flynn K.A., Dyar M., randomized, placebo-controlled trial. Evans M.L., Pritts E., Vittinghoff E., Johnson D.B., Atherton P.J., Diekmann Menopause 17, 946–954. McClish K., Morgan K.S. & Jaffe R.B. B. & Loprinzi C.L. (2010) Phase III, Carpenter J.S., Storniolo A.M., Johns S., (2005) Management of postmenopausal placebo-controlled trial of three doses of Monahan P.O., Azzouz F., Elam J.L., hot flushes with venlafaxine citalopram for the treatment of hot Johnson C.S. & Shelton R.C. (2007) hydrochloride: a randomized, controlled flashes: NCCTG trial N05C9. Journal of Randomized, double-blind, placebo- trial. Obstetrics and Gynecology 105, Clinical Oncology 28, 3278–3283. controlled crossover trials of venlafaxine 161–166. Beck A.T., Steer R.A. & Garbin M.G. for hot flashes after breast cancer. The Fahlen M., Fornander T., Johansson H., (1998) Psychometric properties of the Oncologist 12, 124–135. Johansson U., Rutqvist L.E., Wilking N. Beck depression inventory: twenty-five Carpenter J.S., Guthrie K.A., Larson J.C., & von Schoultz E. (2013) Hormone years of evaluation. Clinical Psychology Freeman E.W., Joffe H., Reed S.D., replacement therapy after breast cancer: Review 8,77–100. Ensrud K.E. & LaCroix A.Z. (2012) Effect 10 year follow up of the Stockholm Bielski R.J., Ventura D. & Chang C.C. of escitalopram on hot flash interference: randomised trial. European Journal of (2004) A double-blind comparison of a randomized, controlled trial. Fertility Cancer 49,52–59. escitalopram and venlafaxine extended and Sterility 97, 1399–404. Freedman R.R. (2001) Physiology of hot release in the treatment of major Carroll D.G. & Kelley K.W. (2009) Use of flashes. American Journal of Human depressive disorder. The Journal of antidepressants for management of hot Biology 13, 453–464. Clinical Psychiatry 65, 1190–1196. flashes. Pharmacotherapy 29, 1357– Freedman R.R., Kruger M.L. & Tancer Biglia N., Cozzarella M., Cacciari F., 1374. M.E. (2011) Escitalopram treatment of Ponzone R., Roagna R., Maggiorotto F. Chappell A.S., Ossanna M.J., Liu-Seifert menopausal hot flashes. Menopause 18, & Sismondi P. (2003) Menopause after H., Barton D.L., Lavasseur B.I. & 893–896.

Freeman E.W., Guthrie K.A., Caan B., Holmberg L. & Anderson H. (2004) For breast cancer survivors. The Oncologist Sternfeld B., Cohen L.S., Joffe H., the HABITS steering and data 11,96–110. Carpenter J.S., Anderson G.L., Larson monitoring committees. HABITS LaCroix A.Z., Freeman E.W., Larson J., J.C., Ensrud K.E., Reed S.D., Newton (hormonal replacement therapy after Carpenter J.S., Joffe H., Reed S.D., K.M., Sherman S., Sammel M.D. & breast cancer – is it safe?): a randomised Newton K.M., Seguin R.A., Sternfeld B., LaCroix A.Z. (2011) Efficacy of comparison: trial stopped. The Lancet Cohen L. & Ensrud K.E. (2012) Effects escitalopram for hot flashes in healthy 363, 453–455. of escitalopram on menopause-specific menopausal women: a randomized Holmberg L., Iversen O.E., Rudenstam quality of life and pain in healthy controlled trial. Journal of the C.M., Hammar M., Kumpulainen E., menopausal women with hot flashes: a American Medical Association 305, Jaskiewicz J., Jassem J., Dobaczewska randomized controlled trial. Maturitas 267–274. D., Fjosne H.E., Peralta O., Arriagada R., 73, 361–368. Freeman M.P., Hirschberg A.M., Wang B., Holmqvist M. & Maenpaa J. (2008) Le Saux M. & Di P.T. (2005) Changes in Petrillo L.F., Connors S., Regan S., Joffe HABITS Study Group. Increased risk of 5-HT1A receptor binding and G-protein H. & Cohen L.S. (2013) Duloxetine for recurrence after hormone replacement activation in the rat brain after major depressive disorder and daytime therapy in breast cancer survivors. estrogen treatment: comparison with and nighttime hot flashes associated Journal of the National Cancer Institute tamoxifen and raloxifene. Journal of with the menopausal transition. 100, 475–482. Psychiatry and Neuroscience 30, 110– Maturitas 75, 170–174. International Menopause Society (2013) 117. Goldstein D.J., Lu Y., Detke M.J., Wiltse Updated 2013 International Menopause Loprinzi C.L., Kugler J.W., Sloan J.A., C., Mallinckrodt C. & Demitrack M.A. Society recommendations on Mailliard J.A., LaVasseur B.I., Barton (2004) Duloxetine in the treatment of menopausal hormone therapy and D.L., Novotny P.J., Dakhil S.R., Rodger depression: a double-blind placebo- preventive strategies for midlife health. K., Rummans T.A. & Christensen B.J. controlled comparison with paroxetine. Climacteric 16, 316–337. (2000) Venlafaxine in management of Journal of Clinical Psychopharmacology Joffe H., Soares C.N., Petrillo L.F., Viguera hot flashes in survivors of breast cancer: 24, 389–399. A.C., Somley B.L., Koch J.K. & Cohen a randomised controlled trial. Lancet Gordon P.R., Kerwin J.P., Boesen K.G. & L.S. (2007) Treatment of depression and 356, 2059–2063. Senf J. (2006) Sertraline to treat hot menopause-related symptoms with the Loprinzi C.L., Sloan J.A., Perez E.A., flashes: a randomized controlled, serotonin-norepinephrine reuptake Quella S.K., Stella P.J., Mailliard J.A., double-blind, crossover trial in a general inhibitor duloxetine. Journal of Clinical Halyard M.Y., Pruthi S., Novotny P.J. & population. Menopause 13, 568–575. Psychiatry 68, 943–950. Rummans T.A. (2002) Phase III Grady D. (2006) Clinical practice. Joffe H., Guthrie K.A., Larson J., Cohen evaluation of fluoxetine for treatment of Management of menopausal symptoms. L.S., Carpenter J.S., Lacroix A.Z. & hot flashes. Journal of Clinical New England Journal of Medicine 355, Freeman E.W. (2013) Relapse of Oncology 20, 1578–1583 2338–2347. vasomotor symptoms after Lu N.Z. & Bethea C.L. (2002) Ovarian Gupta P., Sturdee D.W., Palin S.L., discontinuation of the selective steroid regulation of 5-HT1A receptor Majumder K., Fear R., Marshall T. & serotonin reuptake inhibitor binding and G protein activation in Paterson I. (2006) Menopausal escitalopram: results from the female monkeys. symptoms in women treated for breast menopause strategies: finding lasting Neuropsychopharmacology 27,12–24. cancer: the prevalence and severity of answers for symptoms and health Montgomery S.A. & Asberg M. (1979) A symptoms and their perceived effects on research network. Menopause 20, 261– new depression scale designed to be quality of life. Climacteric 9,49–58. 268. sensitive to change. The British Journal Guttuso T. & Evans M. (2010) Minimum Kalay A.E., Demir B., Haberal A., Kalay of Psychiatry 134, 382–389. trial duration to reasonable assess long- M. & Kandemir O. (2007) Efficacy of Montgomery S.A., Huusom A.K. & term efficacy of non hormonal hot flash citalopram on climacteric symptoms. Bothmer J. (2004) A randomised study therapies. Journal of Women’s Health Menopause 14, 223–229. comparing escitalopram with 19, 699–702. Kenemans P., Bundred N.J., Foidart J.M., venlafaxine XR in primary care patients Handley A.P. & Williams M. (2015) The Kubista E., Von Schoultz B., Sismondi with major depressive disorder. efficacy and tolerability of SSRI/SNRIs P., Vassilopoulou-Sellin R., Yip C.H., Neuropsychobiology 50,57–64. in the treatment of vasomotor Egberts J., Mol-Arts M., Mulder R., Van Nappi R.E. & Kokot-Kierepa M. (2010) symptoms in menopausal women: a Os S. & Beckmann M.W. (2009) Women’s voices in the menopause: systematic review. Journal of the LIBERATE Study Group. Safety and results from an international survey on American Association of Nurse efficacy of tibolone in breast-cancer vaginal atrophy. Maturitas 67, 233–238. Practitioners 27,54–61. patients with vasomotor symptoms: a Nelson H.D., Vesco K.K., Haney E., Fu R., Henry N.L., Banerjee M., Wicha M., Van double blind, randomised, non- Nedrow A., Miller J., Nicolaidis C., Poznak C., Smerage J.B., Schott A.F., inferiority trial. The Lancet Oncology Walker M. & Humphrey L. (2006) Griggs J.J. & Hayes D.F. (2011) Pilot 10, 135–146. Nonhormonal therapies for menopausal study of duloxetine for treatment of Khan A., Bose A., Alexopoulos G.S., hot flashes: systematic review and aromatase inhibitor-associated Gommoll C., Li D. & Gandhi C. (2007) meta-analysis. Journal of the American musculoskeletal symptoms. Cancer 117, Double-blind comparison of Medical Association 295, 2057–2071. 5469–5475. escitalopram and duloxetine in the Nierenberg A.A., Greist J.H., Mallinkrodt Hickey M., Saunders C., Partridge A., acute treatment of major depressive C., Prakash A., Sambunaris A., Santoro N., Joffe H. & Sterns V. (2008) disorder. Clinical Drug Investigation 27, Tollefson G.D. & Wohlreich M.M. Practical guidelines for assessing and 481–492. (2007) Duloxetine versus escitalopram managing menopausal symptoms after Knobf M.T. (2006) The influence of and placebo in the treatment of patients breast cancer. Annals of Oncology 19, endocrine effects of adjuvant therapy on with major depressive disorder: onset of 1669–1680. quality of life outcomes in younger antidepressant action, a non-inferiority

study. Current Medical Research and adverse events: a comparison of depressive disorder. Menopause 17, 700– Opinion 23, 401–416. escitalopram versus serotonin 701. Orleans R.J., Li L., Kim M.J., Guo J., norepinephrine reuptake inhibitors. Speroff L., Gass M., Constantine G. & Sobhan M., Soule L. & Joffe H.V. (2014) Current Medical Research and Opinion Olivier S. (2008) Efficacy and tolerability FDA approval of paroxetine for 27, 1089–1096. of desvenlafaxine succinate treatment menopausal hot flushes. The New Simon J.A., Portman D.J., Kaunitz A.M., from menopausal vasomotor symptoms: England Journal of Medicine 370, 1777– Mekonnen H., Kazempour K., Bhaskar a randomized controlled trial. 1779. S. & Lippman J. (2013) Low-dose Obstetrics and Gynecology 111,77–87. Pinkerton J.V., Archer D.F., Guico-Pabia paroxetine 7.5 mg for menopausal Stearns V., Beebe K.L., Iyengar M. & Dube C.J., Hwang E. & Cheng R.F. (2013) vasomotor symptoms: two randomized E. (2003) Paroxetine controlled release Maintenance of the efficacy of controlled trials. Menopause 20, 1027– in the treatment of menopausal hot desvenlafaxine in menopausal 1035. flashes: a randomized controlled trial. vasomotor symptoms: a 1-year Sismondi P., Kimmig R., Kubista E., Biglia Journal of the American Medical randomized controlled trial. Menopause N., Egberts J., Mulder R., Planellas J., Association 289, 2827–2834. 20,38–46. Moggio G., Mol-Arts M. & Kenemans P. Stearns V., Slack R., Greep N., Henry- Pritchett Y.L., McCarberg B.H., Watkin (2011) Effects of tibolone on climacteric Tilman R., Osborne M., Bunnell C., J.G. & Robinson M.J. (2007) Duloxetine symptoms and quality of life in breast Ullmer L., Gallagher A., Cullen J., for the management of diabetic cancer patients – data from LIBERATE Gehan E., Hayes D.F. & Isaacs C. (2005) peripheral neuropathic pain: response trial. Maturitas 70, 365–372. Paroxetine is an effective treatment for profile. Pain Medicine 8, 397–409. Skljarevski V., Desaiah D., Liu-Seifert H., hot flashes: results from a prospective Reed S.D., Ludman E.J., Newton K.M., Zhang Q., Chappell A.S., Detke M.J., randomized clinical trial. Journal of Grothaus L.C., LaCroix A.Z., Iyengar S., Atkinson J.H. & Backonja M. Clinical Oncology 23, 6919–6930. Nekhlyudov L., Spangler L., Jordan L., (2010) Efficacy and safety of duloxetine The North American Menopause Society Ehrlich K. & Bush T. (2009) Depressive in patients with chronic low back pain. (NAMS) (2012) The 2012 hormone symptoms and menopausal burden in Spine 35, e578–e585. therapy position statement of the north the midlife. Maturitas 62, 306–310. Sloan J.A., Loprinzi C.L., Novotny P.J., American menopause society Romanovsky A.A. (2007) Barton D.L., Lavasseur B.I. & menopause. Menopause 19, 257–271. Thermoregulation: some concepts have Windschitl H. (2001) Methodological Vilar-Gonzalez S., Perez-Rozos A. & changed. Functional architecture of the lessons learned from hot flash study. Cabanillas-Farpon R. (2011) Mechanism thermoregulatory system. American Journal of Clinical Oncology 19, 4280– of hot flashes. Clinical and Journal of Physiology Regulatory 4290. Translational Oncology 13, 143–147. Integrative and Comparative Physiology Słopien R., Meczekalski B. & Warenik- Von Schoultz E. & Rutqvist L.E. (2005) 292,37–46. Szymankiewicz A. (2003) Relationship Stockholm Breast Cancer Study Group. Rossmanith W.G. & Ruebberdt W. (2009) between climacteric symptoms and Menopausal hormone replacement What causes hot flushes? The serum serotonin levels in therapy after breast cancer: the neuroendocrine origin of vasomotor postmenopausal women. Climacteric 6, Stockholm randomised trial. Journal of symptoms in the menopause. 53–57. the National Cancer Institute 97, 533– Gynecological Endocrinology 25, 303– Soares C.N., Arsenio H., Joffe H., Bankier 535. 314. B., Cassano P., Petrillo L.F. & Cohen Wade A., Gembert K. & Florea I. (2007) A Sassarini J. & Lumsden M.A. (2013) Non- L.S. (2006) Escitalopram versus ethinyl comparative study of the efficacy of hormonal management of vasomotor estradiol and norethindrone acetate for acute and continuation treatment with symptoms. Climacteric 16,31–36. symptomatic peri- and postmenopausal escitalopram versus duloxetine in Serva L., Rivkin M., Nakashima A. & women: impact on depression, patients with major depressive disorder. Sabban E.L. (2002) Estradiol stimulates vasomotor symptoms, sleep, and quality Current Medical Research and Opinion gene expression of norepinephrine of life. Menopause 13, 780–786. 23, 1605–1614. biosynthetic enzymes in rat locus Soares C.N., Joffe H., Viguera A.C., Watts S.W. & Davis R.P. (2011) 5- coeruleus. Neuroendocrinology 75, 193– Petrillo L., Rydzewski M., Yehezkel R., hydroxtryptamine receptors in systemic 200. Somley B. & Cohen L.S. (2008) hypertension: an arterial focus. Expert Shams T., Firwana B., Habib F., Paroxetine versus placebo for women in Review of Cardiovascular Therapy 29, Alshaharani A., Alnouh B., Murad M.H. midlife after hormone therapy 54–67. & Ferwana M. (2014) SSRIs for hot discontinuation. The American Journal Wu E.Q., Greenberg P.E., Yang E., Yu flushes: a systematic review and meta- of Medicine 121, 159–162. A.P., Ben-Hamadi R. & Erder M.H. analysis of randomized trials. Journal of Soares C.N., Thase M.E., Clayton A., (2009) Treatment persistence, healthcare General Internal Medicine 29, 204–213. Guico-Pabia C.J., Focht K., Jiang Q., utilisation and costs in adult patients Signorovitch J., Ramakrishnan K., Ben- Kornstein S.G., Ninan P., Kane C.P. & with major depressive disorder: a Hamadi R., Yu A.P., Wu E.Q., Dworak Cohen L.S. (2010) Desvenlafaxine and comparison between escitalopram and H. & Erder M.H. (2011) Remission of escitalopram for the treatment of other SSRI/SNRIs. Journal of Medical major depressive disorder without postmenopausal women with major Economy 12, 124–135.