INTRODUCTION
Depression is the psychiatric disorder that has received the most attention among cancer patients, particularly among breast cancer patients. In a recent literature review, Massie (Reference Massie2004) reported a prevalence rate of depression among women with breast cancer that varied between 1.5% and 46% across all stages of the disease. When the authors made a distinction between the prevalence of major and minor depression (e.g., clinically significant depressive symptoms, dysthymia, adjustment disorder with depressed mood or with mixed anxiety and depressed mood), it appeared that between 2% and 9.6% of women with breast cancer met the criteria for major depression, whereas between 24% and 57% suffered from minor depression (Sachs et al., Reference Sachs, Rasoul-Rockenschaub and Aschauer1995; Aragona et al., Reference Aragona, Muscatello and Losi1996; Pasacreta, Reference Pasacreta1997; Kissane et al., Reference Kissane, Clarke and Ikin1998).
Compared to breast cancer patients at other stages of the disease (Massie, Reference Massie2004), patients receiving palliative care appear to be more affected by depression. For example, Ciaramella and Poli (Reference Ciaramella and Poli2001) evaluated the prevalence of major depression in 100 cancer patients with various types of cancer and stages. Patients with distant metastases (30%) showed a significantly higher prevalence of major depression. In fact, a review of the literature indicates that among this group the prevalence of major depression as assessed by a diagnostic interview varies between 5% and 26% with a mean of 15% (Hotopf et al., Reference Hotopf, Chidgey and Addington-Hall2002).
Some studies have looked at the prevalence of depression specifically among women with metastatic breast cancer. However, most of these studies used the depression subscale from the Hospital Anxiety and Depression Scale (HADS-D; Zigmond & Snaith, Reference Zigmond and Snaith1983) rather than a diagnostic interview. By using a clinical score of 11 on the HADS-D, a prevalence of depression varying from 7% to 12.5% was observed across studies (Hopwood et al., Reference Hopwood, Howell and Maguire1991b; Pinder et al., Reference Pinder, Ramirez and Black1993; Fulton, Reference Fulton1997, Reference Fulton1998; Love et al., Reference Love, Grabsch and Clarke2004). To our knowledge, only one group of researchers (Kissane et al., Reference Kissane, Grabsch and Love2004; Love et al., Reference Love, Grabsch and Clarke2004) has evaluated the prevalence of depressive disorders in women with metastatic breast cancer using a structured interview based on the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV; American Psychiatric Association, 1994). Thirty-two percent of the 227 women interviewed met the diagnostic criteria for a depressive disorder: 7.0% with major depression, 25.5% with another depressive disorder (mostly an adjustment disorder with depressed mood or with mixed anxiety and depressed mood) and 1.3% with dysthymia. Another study by Okamura et al. (Reference Okamura, Yamawaki and Akechi2005) used the criteria of the DSM-III-R and the DSM-IV to evaluate the prevalence of psychiatric disorders in 50 women with a first recurrence of breast cancer (98% were metastastic). Only 2% of the participants met the criteria for major depression, and 18% had an adjustment disorder with depressed mood or with mixed anxiety and depressed mood.
In sum, few studies have until now looked at the prevalence of depression in women with metastatic breast cancer, and the results vary importantly from one study to the other. However, together these studies indicate that the prevalence of depression appears to be elevated among this population. Besides the diversity of measures, diagnostic criteria, and clinical thresholds which were used, the high variability in the results may be largely accounted for by participants' characteristics (e.g., age) as well as the time when the evaluation took place (e.g., soon following a diagnosis of metastasis, during palliative care, at the end of life). In fact, several factors associated with metastatic breast cancer may increase the likelihood of developing depressive symptoms.
RISK FACTORS AND CORRELATES
Medical
Severity of the Illness
As already mentioned, several studies suggest that cancer stage is a factor strongly associated with depression (Spiegel, Reference Spiegel1993; Massie et al., Reference Massie, Gagnon and Holland1994). The type of metastasis may also influence the presence of depressive symptoms given its strong association with prognosis. However, very few studies have investigated this hypothesis, and no evidence of an association was found (Pinder et al., Reference Pinder, Ramirez and Black1993; Okamura et al., Reference Okamura, Yamawaki and Akechi2005).
It has also been suggested that the progression of the disease toward a terminal stage may be associated with an increase in psychological distress. For example, a study conducted by Butler et al. (Reference Butler, Koopman and Cordova2003) on 59 women with metastatic breast cancer showed that psychological distress, as measured by the Profile of Mood States (POMS; McNair et al., Reference McNair, Lorr and Droppleman1971), remained generally stable or decreased during the course of the disease, but increased significantly in the few months preceding death. In the cross-sectional study by Pinder et al. (Reference Pinder, Ramirez and Black1993), patients with advanced breast cancer who died in the month after they had completed the HADS-D displayed significantly more depressive symptoms compared to patients who survived longer. Finally, a study by Hopwood et al. (Reference Hopwood, Howell and Maguire1991a), which also used the HADS-D on women with an advanced cancer, showed that the prevalence of depressive symptoms in patients who died during the course of their study was higher than in patients who survived longer, although this difference was not statistically significant.
Physical and Psychophysiological Symptoms
Only some studies have investigated the relationship between physical symptoms and depression in women with metastatic breast cancer, and the significant factors are noticeably the same as in the general cancer population (Ciaramella & Poli, Reference Ciaramella and Poli2001; Carpenter et al., Reference Carpenter, Elam and Ridner2004; Bender et al., Reference Bender, Ergyn and Rosenzweig2005; Mystakidou et al., Reference Mystakidou, Rosenfeld and Parpa2005a; Reuter et al., Reference Reuter, Classen and Roscoe2006). More precisely, pain (Hopwood et al., Reference Hopwood, Howell and Maguire1991a), fatigue (Kissane et al., Reference Kissane, Grabsch and Love2004), insomnia (Koopman et al., Reference Koopman, Nouriani and Erickson2002), an altered level of functioning (Hopwood et al., Reference Hopwood, Howell and Maguire1991a; Pinder et al., Reference Pinder, Ramirez and Black1993), as well as shortness of breath and gastro-intestinal symptoms (Hopwood et al., Reference Hopwood, Howell and Maguire1991a; Fulton, Reference Fulton1997) have all been associated with increased depression in women with metastatic breast cancer. However, none of these cross-sectional studies permits us to establish a causal relationship between these factors and the development of depression. Stommel et al. (Reference Stommel, Kurtz and Kurtz2004) assessed the longitudinal evolution of symptoms in 860 patients over the age of 65 suffering from a variety of cancers (i.e., breast, colon, lung, prostate). The analyses indicated that the severity of physical symptoms (e.g., nausea, pain, loss of appetite, fatigue, diarrhea) was a strong predictor of depressive symptoms in this population.
Cancer Treatments
The handful of studies that have looked at the impact of palliative treatments (i.e., hormone therapy, chemotherapy, Herceptin) on psychological distress suggest that chemotherapy may be associated with a higher prevalence of depression among women with advanced breast cancer (Miranda et al., Reference Miranda, de Resende and Melo2002; Okamura et al., Reference Okamura, Yamawaki and Akechi2005). Specifically, in the prospective study conducted by Miranda et al. (Reference Miranda, de Resende and Melo2002), the proportion of women with stage IIB and III breast cancer who were suffering from depression increased following their chemotherapy treatments, although this increase was not significant. The study by Okamura et al. (Reference Okamura, Yamawaki and Akechi2005) examined the presence of mood and anxiety disorders, as evaluated by the Structured Clinical Interview for DSM-III-R (SCID) (Spitzer et al., Reference Spitzer, Williams, Gibbon and First1990), among 50 women with breast cancer, mostly metastatic. The fact of receiving a certain type of chemotherapy (i.e., a combination of doxorubicin and cyclophosphamide) but no other type of treatment (docetaxel, paclitaxel, radiotherapy, hormone therapy) was significantly associated with the presence of a psychiatric disorder. A study by Byar et al. (Reference Byar, Berger and Bakken2006) indicated that depressive symptoms as evaluated by the HADS-D reached their peak at the time of the fourth chemotherapy treatment and tended to return to baseline level 2 months after the end of the treatments.
On the other hand, Stommel et al. (Reference Stommel, Kurtz and Kurtz2004) observed no effect of treatment (surgery, radiotherapy, chemotherapy) on psychological variables, and another study (Mystakidou et al., Reference Mystakidou, Rosenfeld and Parpa2005a) found that the fact of not receiving anti-neoplastic treatments (chemotherapy or radiotherapy) was a predictor of depressive symptoms among patients with advanced cancers of all types (16.7% breast cancer). Additional longitudinal studies are therefore needed to better document the relationship between cancer treatments and depression.
Sociodemographic Factors
Gender
Among the general population, women are recognized as being approximately twice as susceptible as men to developing depression (Kessler et al., Reference Kessler, McGonagle and Zhao1994). Hence, the prevalence of depression among breast cancer patients, almost all of whom are women, may be higher than in other types of cancer. In a review of the literature, Massie (Reference Massie2004) found that breast cancer was one of the four types of cancer that had the highest prevalence rates of depression (i.e., in order, oropharyngal 22%–57%, pancreas 33%–50%, breast 1.5%–46%, lung 11%–44%). However, the prevalence of depression among women with a gynecological cancer was lower (12%–23%), even if this is exclusively a feminine type of cancer. Conversely, Parker et al. (Reference Parker, Baile and de Moor2003) observed greater psychological distress (i.e., anxiety and depression) among women with exclusively feminine cancers like breast or gynecological cancer than among patients with urological and gastro-intestinal cancers, the former exclusively masculine and the latter mixed. It was, however, impossible to assess the effect of gender on patients with gastro-intestinal cancer given the small size of this subgroup. When statistically controlling for the type of cancer, certain authors observed a higher depression rate among women (Pettingale et al., Reference Pettingale, Burgess and Greer1988; Stommel et al., Reference Stommel, Kurtz and Kurtz2004; Mystakidou et al., Reference Mystakidou, Rosenfeld and Parpa2005a), whereas others did not arrive at this conclusion (Plumb & Holland, Reference Plumb and Holland1981; Kathol et al., Reference Kathol, Mutgi and Williams1990; Ciaramella & Poli, Reference Ciaramella and Poli2001; Hirai et al., Reference Hirai, Suzuki and Tsuneto2002; Lloyd-Williams et al., Reference Lloyd-Williams, Dennis and Taylor2004).
Age
Age also remains a controversial risk factor. In fact, most studies suggest that young adults with cancer are more at risk of depression than those who are older (Kathol et al., Reference Kathol, Mutgi and Williams1990; Potash & Breitbart, Reference Potash and Breitbart2002; Parker et al., Reference Parker, Baile and de Moor2003; Lloyd-Williams et al., Reference Lloyd-Williams, Dennis and Taylor2004; Wong-Kim & Bloom, Reference Wong-Kim and Bloom2005). For example, Kissane et al. (Reference Kissane, Grabsch and Love2004) found a significant association between lower age and depression among women with metastatic cancer, but not among women with a localized breast cancer. Contrary to this, other authors suggested that elderly cancer patients face greater losses (e.g., physical, financial, death of loved ones) and may therefore be more at risk of depression and suicidal ideation than younger patients (Massie et al., Reference Massie, Gagnon and Holland1994). Consistent with this hypothesis, Mystakidou et al. (Reference Mystakidou, Rosenfeld and Parpa2005a) observed a significant correlation between higher age and depressive symptoms among a group of patients with advanced cancers of all types.
Socioeconomic Status and Education Level
Among women with advanced breast cancer, those with a lower socioeconomic status would appear to be more at risk of developing depressive symptoms (Pinder et al., Reference Pinder, Ramirez and Black1993). Moreover, a higher level of education might prevent the onset of depression in cancer patients (Stommel et al., Reference Stommel, Kurtz and Kurtz2004), although this relation was not observed by some authors (Parker et al., Reference Parker, Baile and de Moor2003; Okamura et al., Reference Okamura, Yamawaki and Akechi2005).
Marital Status
Married persons or those living with a partner would appear to present fewer depressive symptoms than single, separated, and divorced people (Parker et al., Reference Parker, Baile and de Moor2003). Wong-Kim and Bloom (Reference Wong-Kim and Bloom2005) did not, however, observe this protecting role. Another study by Shapiro et al. (Reference Shapiro, Lopez and Schwartz2001) observed a significant relationship between the quality of the couple's relationship and the quality of life among women with stage II breast cancer. It may thus be hypothesized that the perception of received emotional support is a more important predictive factor than marital status.
Psychosocial Factors
Social Support
Different psychosocial variables may influence the risk of depression among patients with metastatic breast cancer. In particular, good social support may be effective in helping people with cancer adapt to their illness (Massie et al., Reference Massie, Gagnon and Holland1994; Shapiro et al., Reference Shapiro, Lopez and Schwartz2001; Potash & Breitbart, Reference Potash and Breitbart2002). For instance, the perception of adequate social support has been associated with better psychosocial indices (e.g., anxiety, depression, overall mental health) among a group of patients with cancers of various types and at various stages (Parker et al., Reference Parker, Baile and de Moor2003). However, results of a prospective study (Ranchor et al., Reference Ranchor, Sanderman and Steptoe2002) contradict these findings. What these researchers observed was an association between a high level of premorbid social support (i.e., before the cancer diagnosis) and an increase in long-term psychological distress (i.e., 1 year after the diagnosis). This unexpected relationship was explained by the potential negative influence of excessive support or overprotection on the ability of a person to adapt over the long term to a major life event like a cancer diagnosis. It appears, therefore, that the relation between social support and depression is more complex than was initially thought.
Coping Strategies
Coping strategies employed by women with breast cancer may also influence the development of depressive symptoms. Studies on the subject generally suggest that a higher perceived level of control and active coping styles are associated with better adjustment to cancer (Dunkel-Schetter et al., Reference Dunkel-Schetter, Feinstein and Taylor1992; Hirai et al., Reference Hirai, Suzuki and Tsuneto2002). Likewise, people with higher perceived self-efficacy would appear to be less depressed (Hirai et al., Reference Hirai, Suzuki and Tsuneto2002; Ranchor et al., Reference Ranchor, Sanderman and Steptoe2002). Additionally, it has been observed that older women with stage IV breast cancer employed coping strategies characterized as less fighting spirit, more hopelessness/helplessness, and more fatalism and were comparatively more anxious and depressed than younger women with nonmetastatic cancer (Schnoll et al., Reference Schnoll, Harlow and Stolbach1998). Thus, it would appear that coping strategies have a mediating role between some sociodemographics and depression.
Psychiatric History
Most authors agree that patients with a personal history of depression are more at risk of developing depressive symptoms following a cancer diagnosis (Plumb & Holland, Reference Plumb and Holland1981; Potash & Breitbart, Reference Potash and Breitbart2002; Meyer et al., Reference Meyer, Sinnott and Seed2003). In a longitudinal study conducted by Ranchor et al. (Reference Ranchor, Sanderman and Steptoe2002), 99 patients with mixed cancer sites and stages were evaluated before confirmation of their diagnosis (T0), as well as at 2 (T1), 6 (T2) and 12 (T3) months after the diagnosis. The results showed that premorbid distress (T0) was a significant predictor of short- (T1) and long- (T3) term distress. In the case of women with metastatic breast cancer, some researchers have observed that a past history of depression was significantly associated with the presence of current major depression (Kissane et al., Reference Kissane, Grabsch and Love2004; Okamura et al., Reference Okamura, Yamawaki and Akechi2005), whereas others were unable to show this relationship (Pinder et al., Reference Pinder, Ramirez and Black1993).
In summary, numerous demographic, medical, and psychosocial factors may increase the risk that women with metastatic breast cancer will develop depressive symptoms. The results have frequently been contradictory, notably with regard to cancer treatments, demographic factors, and social support. More large-scale longitudinal studies are needed to establish the contribution of each of these factors and their interaction with each other in the development of depression in women with metastatic cancer.
CONSEQUENCES
Psychological
In metastatic breast cancer patients, as for other types of cancer, the studies have frequently associated depression with a decreased quality of life (Weitzner et al., Reference Weitzner, Meyers and Stuebing1997; Skarstein et al., Reference Skarstein, Aass and Fossa2000; O'Brien, Reference O'Brien2003; Badger et al., Reference Badger, Braden and Mishel2004). Results of two studies that looked at the quality of life of people with cancer using the European Organization for Research and Treatment of Cancer Quality of life Questionnaire (Aaronson et al., Reference Aaronson, Ahmedzai and Bergman1993) suggest that depression is significantly associated with an alteration in physical, emotional, cognitive, and social dimensions of quality of life as well as with a decrease in overall perceived state of health and an increase in fatigue (Smith et al., Reference Smith, Gomm and Dickens2003; Mystakidou et al., Reference Mystakidou, Rosenfeld and Parpa2005a). Another study by Rustøen et al. (Reference Rustøen, Moum and Padilla2005) evaluated predictors for the quality of life among a sample of 157 patients with bone metastases (48% of whom were women with breast cancer). Results showed that depression was the most important predictor for quality of life, compared to pain and physical and social functioning. Likewise, a study conducted among breast cancer patients, for the most part with metastases, showed that participants with a psychiatric disorder had a significant decrease in the quality of life, with regard to the functional state (i.e., emotional functioning, self-image, perspective on the future), and to certain physical symptoms (i.e., loss of appetite, diarrhea, fatigue, nausea/vomiting), compared to women with no psychiatric diagnosis (Okamura et al., Reference Okamura, Yamawaki and Akechi2005). Finally, a recent longitudinal study showed that sleep disturbances over a period of 12 months were significantly predicted by depression at baseline (Palesh et al., Reference Palesh, Collie and Batiuchok2007).
Besides being associated with a noticeable reduction in the quality of life, depression may increase the risk of suicide in persons with cancer (Massie et al., Reference Massie, Gagnon and Holland1994). In one study, women with mixed cancer sites and stages were two to three times more likely to commit suicide than women in the general population (Björkenstam et al., Reference Björkenstam, Edberg and Ayoubi2005). Moreover, the suicide rate was higher among those with a poor prognosis.
Several authors have studied the desire for premature death among persons in the terminal phase of cancer. All the studies reviewed suggest that depression is associated with a desire for premature death or increased suicidal ideations (Chochinov et al., Reference Chochinov, Wilson and Enns1995; Breitbart et al., Reference Breitbart, Rosenfeld and Pessin2000; Filiberti et al., Reference Filiberti, Ripamonti and Totis2001; Suarez-Almazor et al., Reference Suarez-Almazor, Newman and Hanson2002; Kelly et al., Reference Kelly, Burnett and Pelusi2003; Mystakidou et al., Reference Mystakidou, Tsilika and Parpa2005b; O'Mahony et al., Reference O'Mahony, Goulet and Kornblith2005). A strong desire for premature death was reported in from 8.5% to 17% of patients in palliative care, and these rates may rise to between 40% and 47% in patients who are suffering from depression (Chochinov et al., Reference Chochinov, Wilson and Enns1995; Breitbart et al., Reference Breitbart, Rosenfeld and Pessin2000; Kelly et al., Reference Kelly, Burnett and Pelusi2003; Mystakidou et al., Reference Mystakidou, Tsilika and Parpa2005b).
Van der Lee et al. (Reference Van der Lee, van der Bom and Swarte2005) evaluated the relationship between depression and the rate of explicit requests for euthanasia in the Netherlands among 138 patients in the terminal phase of cancer. The results showed that the risk of requesting euthanasia among patients with depressive symptoms was 4.1 times higher than in nondepressed patients. Furthermore, there are some data suggesting that a decrease in depressive symptoms over time is associated with a decreased desire for premature death (O'Mahony et al., Reference O'Mahony, Goulet and Kornblith2005). These data suggest that it is essential that a systematic evaluation of depressive symptoms be made among patients who request euthanasia and highlight the importance of treating depression in terminally ill patients.
Medical
Although there is still no consensus due to the contradictory results obtained, certain studies have observed a link between depression and progression of cancer (Spiegel & Giese-Davis, Reference Spiegel and Giese-Davis2003). Although further research is needed to elucidate the mechanisms, this relationship may be explained by the impact of depression on the immune system, on the neuroendocrine system, and/or on health behaviors (e.g., adherence to treatments, tobacco use, alcohol use, sleep habits; Werth et al., Reference Werth, Gordon and Johnson2002; Reiche et al., Reference Reiche, Nunes and Morimoto2004; Watson et al., Reference Watson, Feldman-Stewart and Brundage2005). A group of researchers (Watson et al., Reference Watson, Haviland and Greer1999) looked at the type of coping strategies used by 578 women recently diagnosed with breast cancer. Higher levels of hopelessness and helplessness were found to significantly increased risk of recurrence or death at 5 years, but not depression. In a large-scale study by Goodwin et al. (Reference Goodwin, Zhang and Ostir2004), breast cancer participants who had been diagnosed with depression in the 2 years prior to their cancer diagnosis were significantly more likely to see their tumor and the stage of their cancer progress more rapidly. In addition, women with a previous history of depression were 42% more likely to die of their breast cancer within 3 years of their diagnosis. Finally, Meyer et al. (Reference Meyer, Sinnott and Seed2003) followed the evolution of depressive symptoms in 45 advanced cancer patients over a period of 6 months. They observed that an initial depression (i.e., at the first time of measurement) was not a predictor for survival, but an increase in depressive symptoms over time was significantly associated with an earlier decease.
In spite of the elevated prevalence of depression among cancer patients, particularly at the advanced stages, as well as its potentially harmful psychological and medical consequences, depression currently remains undertreated in this population (Block, Reference Block2000; Bowers & Boyle, Reference Bowers and Boyle2003; Greenberg, Reference Greenberg2004).
TREATMENT OF DEPRESSION IN METASTATIC BREAST CANCER
Pharmacological
Antidepressants are the most widely used treatment for depression, both among the general population and in cancer patients (Fisch, Reference Fisch2004; Williams & Dale, Reference Williams and Dale2006). Pharmacotherapy may also be used in patients with advanced cancer or in the terminal stage in order to reduce their depressive symptoms and suicidal ideation (Massie et al., Reference Massie, Gagnon and Holland1994; Potash & Breitbart, Reference Potash and Breitbart2002). It has been suggested that selective serotonin reuptake inhibitors should be used as the drug of choice because of their lesser side effects when compared to tricyclics (Potash & Breitbart, Reference Potash and Breitbart2002). Incidentally, certain antidepressants may also be effective for other symptoms frequently associated with breast cancer, such as fatigue, sleep difficulties, and hot flashes (Weitzner et al., Reference Weitzner, Moncello and Jacobsen2002; Ladd et al., Reference Ladd, Newport and Ragan2005).
Very few studies have, nonetheless, looked at the efficacy of pharmacotherapy for the treatment of depression in a context of cancer. A recent literature review identified six randomized and placebo-controlled studies on the subject, most of which were double blind (Williams & Dale, Reference Williams and Dale2006). In nonmetastatic patients, three studies evaluated the efficacy of paroxetine for the treatment of depression in patients with various types of cancer (Morrow et al., Reference Morrow, Hickok and Roscoe2003), with a melanoma (Musselman et al., Reference Musselman, Lawson and Gumnick2001), or with breast cancer (Roscoe et al., Reference Roscoe, Morrow and Hickok2005). The results indicated that a dose varying from 10 to 40 mg of paroxetine (average = 20 mg) was effective, both for the reduction of depressive symptoms and in reducing the number of patients who met the criteria of major depression in the short term (i.e., from 8 to 12 weeks of treatment). Only one of the three studies reported on the tolerability of the medication (Musselman et al., Reference Musselman, Lawson and Gumnick2001), and findings indicated no significant difference in the types of side effects reported after 12 weeks of treatment between the group treated with paroxetine and the placebo control group. Three people out of 20 from the paroxetine group suffered retinal hemorrhages, but they also had other risk factors. One study examined the efficacy of fluoxetine among a group of patients with mixed cancer sites and stages suffering from a major depressive or an adjustment disorder (n = 91; Razavi et al., Reference Razavi, Allilaire and Smith1996). This study revealed no significant difference between the fluoxetine group and the placebo group after 5 weeks of treatment as far as the percentage of positive response was concerned, defined by a score of less than eight on the HADS-D. With regard to tolerability, no significant difference in side effects was observed between the fluoxetine and control groups after 5 weeks of treatment. Finally, a research team looked at the efficacy of mianserine, a tetracyclic antidepressant, for the treatment of depression in 55 women with nonmetastatic breast cancer (Van Heeringen & Zivkov, Reference Van Heeringen and Zivkov1996). A greater reduction of Hamilton Rating Scale for Depression (HDRS; Hamilton, Reference Hamilton1960) scores was observed in the group treated with mianserine compared to the placebo control group after 4 and 6 weeks of treatment. The number of patients responding favorably to the treatment, as defined by a reduction of at least 50% of the HDRS score, was also significantly higher in the mianserine group than in the placebo group at the same time points. With regard to tolerability, no significant difference was observed between the groups.
Even fewer studies have looked at the efficacy of antidepressants in patients with advanced cancer, and none has been conducted specifically in metastatic breast cancer patients. Fisch et al. (Reference Fisch, Loehrer and Kristeller2003) assessed the efficacy of fluoxetine compared to a placebo to treat depressive symptoms and improve the quality of life in 163 patients with advanced cancer of various types. After 12 weeks of treatment, patients who had received fluoxetine showed a lower level of depression and a significantly higher level of quality of life than the placebo group. In this study the female gender was associated with better results, both with regard to depressive symptoms and quality of life. Another group (Holland et al., Reference Holland, Romano and Heiligenstein1998) assessed the efficacy and tolerability of fluoxetine and desipramine in the treatment of depressive symptoms in 40 women with advanced cancer (75% with breast cancer). Their results showed that these two medications were significantly more effective than a placebo in reducing depressive and anxiety symptoms and in improving the quality of life in this population after 6 weeks of treatment. Fluoxetine (20 mg/day) and desipramine (25–150 mg/day) would appear to be relatively well tolerated, though several side effects were felt, the most frequent being dryness of mouth, nausea, and pain. Six patients who were treated with fluoxetine along with four who were treated with desipramine abandoned the study because of the side effects (e.g., somnolence, tachycardia, depersonalization, pain). However, no comparison for tolerability was made between the groups treated with the antidepressants and the placebo group. Finally, amitryptyline and clomipramine, two tricyclic antidepressants, were found to be effective in treating major depression and reducing suicidal ideations and the desire for premature death in a limited number of patients (n = 6) in the terminal phase of cancer (Kugaya et al., Reference Kugaya, Akechi and Nakano1999).
In sum, antidepressants seem to be effective for the treatment of depression in cancer patients. However, additional studies are needed to investigate their efficacy among specific populations, like women with metastatic breast cancer. In fact, patients with advanced cancer may present a different response because of the physical symptoms that could arise from the illness and treatments and because of their poorest prognosis. Furthermore, the long-term efficacy and tolerability of antidepressants are currently not well known in the context of advanced cancer. Antidepressants appear to be associated with a number of negative side effects (e.g., dryness of mouth, nausea, constipation, anorexia, insomnia) that would be added to the symptoms of advanced cancer and its treatments. Finally, although statistically significant findings have been reported, little information is available on their clinical significance. Given their side effects and that a number of patients are reluctant to take antidepressants for various reasons (Sharpe et al., Reference Sharpe, Strong and Allen2004; Brown et al., Reference Brown, Battista and Bruehlman2005), nonpharmacological interventions should be considered as a possible alternative in the treatment of depression in women with metastatic breast cancer.
Psychological
In the general population, certain forms of psychotherapy, notably cognitive–behavioral and interpersonal therapy, are recognized as empirically supported treatments for depression (Chambless & Ollendick, Reference Chambless and Ollendick2001). Several studies have assessed the efficacy of psychotherapy in reducing depressive symptoms in patients with early-stage cancer. Most of these studies suggest that psychotherapy, particularly cognitive–behavioral therapy, is effective in reducing depressive symptoms in cancer patients (Marchioro et al., Reference Marchioro, Azzarello and Checchin1996; Antoni et al., Reference Antoni, Lehman and Kilbourn2001; Winzelberg et al., Reference Winzelberg, Classen and Alpers2003; Given et al., Reference Given, Given and Rahbar2004). Less numerous studies have looked at the efficacy of psychological interventions specifically in women with metastatic breast cancer. The psychotherapies that were most frequently investigated were support therapy, supportive–expressive group therapy, and cognitive–behavioral therapy, administered individually or in groups.
Supportive–Expressive Therapy
Spiegel et al. (Reference Spiegel, Bloom and Yalom1981) conducted a randomized and controlled study among women with metastatic breast cancer in order to evaluate the efficacy of supportive–expressive therapy (n = 34) compared to an untreated control group (n = 24). Compared to the control group, the total score of psychological distress on the POMS, as well as scores on subscales for tension–anxiety, vigor, fatigue, and confusion, improved significantly after 1 year of treatment. However, no significant difference in depressive symptoms could be observed between the groups.
The results of another study conducted in the same population by Edmonds et al. (Reference Edmonds, Lockwood and Cunningham1999) indicated that long-term supportive–expressive therapy (8 months), which included some cognitive–behavioral strategies, was associated with a reduction in feelings of helplessness and an increase in the quality of life. Again, no significant difference in depressive symptoms was observed between the treated and control groups as assessed by the POMS at the 4-, 8-, and 14-month follow-ups.
Two other randomized and controlled studies evaluated the efficacy of long-term supportive–expressive therapy on depressive symptoms in women with metastatic breast cancer (Classen et al., Reference Classen, Butler and Koopman2001; Goodwin et al., Reference Goodwin, Leszcz and Ennis2001). After controlling for between-group differences at baseline, no difference was detected at posttreatment between the groups on the different scales of the POMS in these two studies. However, Goodwin et al. (Reference Goodwin, Leszcz and Ennis2001) observed that, among women who presented an elevated level of psychological distress at baseline, those who received the intervention showed a significantly greater reduction in depressive symptoms than those in the control group, whereas no difference was observed in women who showed little psychological distress at baseline. Alternatively, the results of the study conducted by Classen et al. (Reference Classen, Butler and Koopman2001) suggest that the absence of difference between the treatment and control groups may be due to the presence of women in the terminal stage of cancer. In fact, when the women who were in their final year of life were withdrawn from their analyses, the group receiving supportive–expressive therapy showed a significantly greater decrease in their psychological distress compared to the control group. Together, these results suggest that long-term supportive–expressive therapy may only be effective in reducing psychological distress for women with certain characteristics, such as a high baseline level of distress (possibly because of a floor effect) and a life expectancy of more than 1 year.
Cognitive–Behavioral Therapy
A first randomized study aimed to assess the effect of a combination of cognitive and behavioral strategies on the pain and mood of 24 women with metastatic breast cancer (Arathuzik, Reference Arathuzik1994). The therapy helped improve these women's capacity to control pain, but no difference was observed between the intervention and control groups as regards mood.
In another randomized and controlled study conducted in women with metastatic breast cancer, cognitive–behavioral therapy was associated with a significant improvement at posttreatment in depression, overall mood, and self-esteem as measured by the POMS subscales (n = 43), compared to the untreated control group (n = 49; Edelman et al., Reference Edelman, Bell and Kidman1999a, Reference Edelman, Lemon and Bell1999b). However, no difference was observed between the groups at the 3- and 6-month follow-ups. Among other possible explanations, the authors explained this weak sustaining of therapeutic gains over time by the fact that treatment was administered in a group and could not be individually tailored to each patient.
The efficacy of individual psychotherapy was first assessed in a pilot study conducted using a single-case experimental study (Lévesque et al., Reference Lévesque, Savard and Simard2004). This study showed that individual cognitive therapy using Beck's model (Beck et al., Reference Beck, Rush and Shaw1979) was effective both in the short and medium term for treating depressive symptoms in women with metastatic breast cancer. Six participants with clinical levels of depression, that is, with a score of 17 or greater on the Beck Depression Inventory (BDI; Beck et al., Reference Beck, Ward and Mendelson1961) or of 7 or greater on the HADS-D, were enrolled in the study. Intervention time-series analyses conducted on daily mood data revealed a statistically significant improvement of depression symptoms and, more importantly, anhedonia and associated features (i.e., anxiety, fatigue) for each participant. The visual inspection of HADS-D scores suggested that the introduction of cognitive therapy was associated with a rapid reduction in depressive symptoms for three of the four completers, and depression scores began to decline at the end of the intervention for the fourth one. In addition, the improvements were judged to be clinically significant and were maintained at the 3- and 6-month follow-up evaluations.
These results were replicated in a randomized controlled trial conducted by the same research team (Savard et al., Reference Savard, Simard and Giguere2006). Forty-five women with metastatic breast cancer and with depressive symptoms (score of 7 or more on the HADS-D or 15 or more on the BDI) took part in this study. The results showed that women who received 8 weeks of cognitive therapy scored significantly lower on the HDRS, compared to the waiting-list control group at posttreatment. When both groups were pooled together after they all received therapy, a significant reduction in depressive symptoms from pre- to posttreatment as well as a significant decrease in associated symptoms like anxiety, fatigue, and insomnia were found. The better sustaining of therapeutic gains obtained in these two studies during follow-up can be explained by the fact that the treatment was administered individually or by the administration of three booster sessions following posttreatment.
In conclusion, it appears that psychotherapy can be effective at reducing depressive symptoms in women with metastatic breast cancer. Although no comparative study has been conducted, Beck's cognitive therapy appears to be the most cost-effective approach for treating women with metastatic breast cancer, given its short-term nature and the magnitude of effects observed. Further studies are needed to identify the best therapeutic approaches for depression in this population and to study the characteristics of women who best respond to psychological interventions.
CONCLUSION
Numerous women with metastatic breast cancer appear to suffer from depression, but further large-scale studies using diagnostic interviews are needed to determine its prevalence with more precision (Hotopf et al., 2002). Many factors can increase the risk that women will develop depressive symptoms, but additional longitudinal studies are needed to establish a causal link between some of these factors and depression in women with metastatic breast cancer and also to look at the evolution of depressive symptoms until the end of life (Fulton, 1997; Butler et al., 2003). Nonetheless, available data suggest that particular attention should be paid to women displaying certain characteristics (e.g., visceral metastases, severe physical symptoms, weak social support, antecedents of depression; Greenberg, 2004). Despite the fact that depression appears to be associated with numerous negative consequences, it remains underdiagnosed and undertreated. There is, therefore, a need to offer treatment to women with metastatic breast cancer who suffer from depression. Currently, it appears that pharmacotherapy is associated with a decrease in depressive symptoms in this population. New studies are needed, however, to evaluate the long-term tolerability and efficacy of pharmacotherapy in cancer patients. Given its efficacy and short-term nature, cognitive-behavioral may be the most cost-effective treatment option for treating depression in women with metastatic breast cancer. However, further studies are needed to determine the specific characteristics of pharmacological or psychological treatments as well as of the cancer patients themselves in order to optimize the beneficial effects. In addition, there does not appear to be any data currently available on the efficacy of combining pharmacological and psychological therapies, although both treatments are commonly combined in clinical practice.
ACKNOWLEDGMENTS
Preparation of this article was supported in part by a studentship from the Canadian Cancer Society through an award from the National Cancer Institute of Canada awarded to the first author, by an operating grant from the Canadian Breast Cancer Research Alliance, and research scientist awards from the Canadian Institutes of Health Research and the Fonds de la recherche en santé du Québec held by the second author. The authors thank Fred Sengmueller for his contribution.
