Psychosomatics 48:205-210, June 2007
doi: 10.1176/appi.psy.48.3.205
© 2007 Academy of Psychosomatic Medicine
Tamoxifen Treatment and New-Onset Depression in Breast Cancer Patients
Kelly C. Lee, Pharm.D., BCPP,
G. Thomas Ray, M.B.A.,
Enid M. Hunkeler, M.A., and
Patrick R. Finley, Pharm.D., BCPP
Received March 22, 2006; revised July 28, 2006; accepted August 8, 2006. From Loma Linda Univ.; Kaiser Permanente Division of Research; Univ. of California at San Francisco. Send correspondence and reprint requests to Kelly C. Lee, Pharm.D., BCPP, Assistant Professor, Pharmacy Practice, Loma Linda Univ. School of Pharmacy, 11262 Campus St., West Hall 1333, Loma Linda, CA 92350. e-mail: kclee{at}llu.edu
© 2007 The Academy of Psychosomatic Medicine
|
ABSTRACT
|
The authors conducted a retrospective cohort study of female patients diagnosed with breast cancer (BRCA), evaluating the risk of new-onset depression associated with tamoxifen treatment among those with estrogen receptor-positive (ER+) tumors, versus estrogen receptor-negative (ER–) tumors, who were not receiving tamoxifen. A total cohort of 2,943 patients was identified. The hazard-ratio for new-onset depression in the tamoxifen group was nonsignificant. A post-hoc analysis revealed that chemotherapy and ER+ status were significantly and independently associated with an increased risk for developing depression.
|
INTRODUCTION
|
The incidence of breast cancer (BRCA) has stabilized for the first time since the steady rise over the past 20 years; however, BRCA remains one of the top four cancers among all races and ethnicities.1 The impact of BRCA on the quality of life in this patient population has been studied extensively, but questions still remain as to the safety and tolerability of medications commonly prescribed to patients with BRCA.2–5 Current NIH recommendations call for initiation of adjuvant hormonal therapy in women found to have tumors containing estrogen-receptor protein (ER+ status). Tamoxifen has been shown to decrease tumor progression and recurrence and has also been approved for prevention of BRCA in high-risk patients.
Although medications such as tamoxifen may slow the progression of cancer or eradicate tumors, such interventions may also have a negative impact on a patient’s quality of life. A recent survey of women at risk for BRCA discovered that side effects are the main reason why they opted to avoid tamoxifen treatment.6 Tamoxifen has been shown to readily penetrate the blood–brain barrier, and the high incidence of "hot flashes" may serve as indirect evidence of its anti-estrogenic properties in the central nervous system.7 According to the manufacturer, tamoxifen has been "infrequently" associated with depression in clinical trials, but mood disorders may also be a product of a BRCA diagnosis, and a causal relationship of depression with adjuvant hormonal therapy has been difficult to demonstrate.8
Clinical observations and studies suggest that patients who receive tamoxifen for treatment of BRCA may also be at increased risk for developing depression. Several medications, such as beta-blockers, corticosteroids, and interferon, have been theorized to contribute to depressive symptoms, but the evidence linking tamoxifen and depression seems to be conflicting.9 The anti-estrogenic effects of tamoxifen at postsynaptic receptors have been theorized to contribute to symptoms of depression.10–12 The main objective of the present study is to investigate the rate of new-onset depression among patients receiving tamoxifen for BRCA treatment. We will also analyze additional risk factors for depression in BRCA patients.
|
METHOD
|
Setting
Kaiser Permanente of Northern California (KPNC) is a nonprofit, integrated healthcare delivery system providing care to over 3 million members. The KPNC Cancer Registry (KPNCCR) records approximately 13,000 incident cases per year, of which about 17% are breast cancer.13 Medication use by members is tracked in KPNC’s Pharmacy Information System (PIMS), an automated clinical database that records prescription drugs dispensed at KPNC pharmacies. About 95% of KPNC members have pharmacy coverage and would therefore be expected to have prescriptions filled at a KPNC pharmacy. Diagnoses for all hospitalizations and outpatient visits are captured by the KPNC automated databases.14
This study was approved by the Committee on Human Research (CHR) of the University of California at San Francisco (UCSF) and the Institutional Review Board (IRB) of Kaiser Foundation Research Institute. Because this was a retrospective analysis of administrative databases, subjects were not reimbursed for participation, and consent was not required.
Study Subjects
Women diagnosed with BRCA between January 1, 1997 and December 31, 2001 were identified from the KPNCCR. Exclusion criteria were the following: 1) unknown BRCA staging or estrogen-receptor (ER) status; 2) any patient with ER+ tumors who did not receive tamoxifen during the study period (i.e., between the date of BRCA diagnosis and the end of follow-up); 3) any patient with ER– tumors who received tamoxifen during the study period; 4) less than 12 months’ membership in the KP health plan before BRCA diagnosis; 5) any patient who had Stage III or IV BRCA at the time of diagnosis; 6) any patient who had a documented history of depression (ICD: 296, 300.4, 311), cerebrovascular accident (ICD: 430–438, except 435), alcohol dependence/abuse (ICD: 291–292, 303–304, 305), or cancers other than BRCA during the 12 months before BRCA diagnosis; 7) any patient who had a history of antidepressant use in the 6 months before their diagnosis of BRCA; and 8) any patient who received medications (clonidine, reserpine, methyldopa, interferon) that might induce or worsen depressive symptoms during the study period. The follow-up period for each patient was from the date of their BRCA diagnosis until either death, leaving the health plan, a diagnosis of depression, or December 31, 2001.
Measuring Exposure to Tamoxifen
During each day of the follow-up, a patient was considered to be either exposed or not exposed to tamoxifen. Tamoxifen exposure was assumed to begin the day after the drug was dispensed to the patient, and patients were assumed to use all quantities dispensed to them over the prescribed period of time as indicated by the days’ supply of the drug given to the patient. Exposure to tamoxifen was assumed to continue for 45 days after the patient stopped use, and small gaps in use, such as those caused by delays in picking up refills every month, were not considered clinically relevant. Thus, a patient was considered to be using tamoxifen from the day after the drug was dispensed to her, throughout the days’ supply dispensed, plus 45 days. Refills or new prescriptions were added to the end of previous prescriptions in a similar manner. Once the patient was considered to have run out of her supply of tamoxifen, and 45 days had passed without any additional drug being dispensed to her, the patient was considered to be no longer exposed to the drug.
New-Onset Depression
Patients were considered to have new-onset depression if they received a diagnosis of depression (ICD-9: 296, 300.4, 311) during a KP hospitalization or outpatient visit, or received an antidepressant during their follow-up. Only the first instance of a depression diagnosis or antidepressant use during follow-up was counted.
Statistical Analyses
We used the counting-process specification of the Cox model to estimate the increased hazard of new-onset depression for subjects who used tamoxifen. The Cox model allows adjustment for differential follow-up and the censoring of individuals, as well as adjustment for other confounders. The time to new-onset depression was scaled as the number of days since the BRCA diagnosis. Because a patient’s exposure to tamoxifen could change over the period of follow-up (i.e., they could go on and off medication), tamoxifen use was treated as a time-changing covariable. We also included age at BRCA diagnosis, ethnicity, tumor type, and the dichotomous indicators of radiation therapy, chemotherapy, and surgery as other covariables. A post-hoc analysis was also conducted that provided a comprehensive analysis of all patients with breast cancer, regardless of tamoxifen or ER status, and these results were compared with the findings from the original cohorts (i.e., with exclusion criteria). In this post-hoc analysis, we added women who 1) had ER+ tumors but did not receive tamoxifen; and 2) had ER– tumors but did receive tamoxifen. These patients were not included in the main analysis because of the inherent differences in tumor type and risk of depression that may exist in this population.
|
RESULTS
|
Subject Characteristics
In all, 2,943 patients met study inclusion criteria, of which 2,329 had ER+ tumors and received tamoxifen treatment and 614 had ER– tumors and were not treated with tamoxifen (Table 1). Patients receiving tamoxifen were older than those not receiving tamoxifen (62.7 years versus 56.6 years; p <0.05), and, on average, had 87 more days of follow-up (881 days versus 794 days). Most of the patients in the total cohort were white (p <0.05) and had localized tumors (p <0.05), and over 95% of the subjects had surgical treatment for their BRCA. Patients receiving tamoxifen were more likely to have undergone radiation (p <0.05) and less likely to have received chemotherapy (p<0.05) than patients who did not receive tamoxifen. A total of 771 patients were diagnosed with depression (or prescribed an antidepressant) during the follow-up. Before adjusting for follow-up time and other covariables, patients receiving tamoxifen were more likely to have been diagnosed with depression after their BRCA diagnosis than patients who did not receive tamoxifen (27% versus 22%; p <0.05). However, this difference was largely due to a number of patients (135 of 771) who were diagnosed with depression before starting tamoxifen. When these patients were removed from analysis, we found no difference in the rate of depression between those who used tamoxifen and those who did not.
Patterns of Tamoxifen Use
Among the 2,329 persons who used tamoxifen after their BRCA diagnosis, the average number of days from the BRCA diagnosis to the start of tamoxifen treatment was 118.8 days, or about 4 months (Figure 1). During the study period, the average duration of tamoxifen treatment was 668.2 days, and, among patients who were diagnosed with depression after their BRCA diagnosis, the average time from tamoxifen initiation to the first diagnosis of depression was 247.9 days. The mean days of follow-up for the entire cohort was 862.8 days.
Adjusted Hazard Ratios for New-Onset Depression
After adjusting for follow-up time, age of the patient, and other covariables, we found no statistically significant relationship between tamoxifen use and development of depression (Table 2; hazard ratio [HR]: 1.1; 95% confidence interval [CI]: 0.89 – 1.2). There were also no statistically significant differences in risk of developing depression with age, radiotherapy, or surgery. Chemotherapy, however, was associated with an increased risk of developing depression (HR: 1.2; 95% CI: 1.0 – 1.5).
In the post-hoc analysis, we included patients with ER+ tumors who did not use tamoxifen (N=933) and patients with ER– tumors who used tamoxifen (N=115). Again, we found no statistically significant relationship between tamoxifen use and developing depression (HR: 1.1; 95% CI: 0.94 – 1.3). In this post-hoc analysis, we also looked at the potential impact of ER+ status, itself (independent of tamoxifen exposure) and found that it was associated with a significant increase in risk of developing new-onset depression (HR: 1.2; 95% CI: 1.0 – 1.5). Chemotherapy was associated with a higher risk of developing depression, as well (HR: 1.5; 95% CI: 1.3 – 1.7). Age, radiotherapy, and surgery were not associated with this increased risk.
|
DISCUSSION
|
Previous attempts at elucidating the depressive effects of tamoxifen have arrived at different conclusions. Cathcart et al.11 found that 15% of ER– BRCA patients (N=301) exhibited clinical depression after starting tamoxifen, as opposed to 3% among BRCA control subjects. Despite the temporal association of tamoxifen and depression, the study was limited by small sample size, absence of a validated depression scale, and the participation of a single intervention oncologist. Anelli et al.15 evaluated the psychosocial effects of tamoxifen use, but in male BRCA patients (N=24). Mood alterations (20.8%), depression (16.6%), and insomnia (12.5%) were the leading psychologic symptoms reported, although the severity of symptoms could not be determined. Shariff et al.16 observed the effects of tamoxifen on various psychosocial scales from baseline to 8 months after tamoxifen was initiated, and found that the score were significantly elevated in a subset (17%) of women (p <0.05). A small pilot study examining the impact of estrogen deficiency on depressive symptoms in 21 BRCA patients found that 38% of the population satisfied diagnostic criteria for major depressive disorder (MDD), and four of these patients had received antidepressant treatment.10 Of the 14 study subjects receiving tamoxifen, 5 (36%) were subsequently diagnosed with MDD.
More recently, Breuer and Anderson reviewed 1,385 matched patient records from the state’s minimal data sets (MDS) and found that postmenopausal women exposed to tamoxifen were 42% more likely to carry a diagnosis of depression (odds ratio [OR]: 1.420; 95% CI: 1.212–1.663).17
Baseline data from the large, multicenter National Surgical Adjuvant Breast and Bowel Project (NSABP) trial included over 11,000 patients who were at high risk for BRCA.18 After stratifying for depression risk, the authors reported that differences between study groups were minimal on their depression rating scale. This investigation, however, did not assess the actual incidence of new-onset clinical depression, and findings may have been further compromised by the underrepresentation of vulnerable high-risk patients and the use of concurrent antidepressants.
In another tamoxifen prevention trial, 488 women with high familial risk were assessed by a questionnaire.12 Investigators reported that 22%–30% of all study participants exceeded the depression threshold during any assessment period, and 75% of subjects exceeded this threshold at least once, attesting to the high incidence of depression in these patients overall.
The third study that failed to find an association was a small, randomized, placebo-controlled trial of postmenopausal women (N=140) with axillary node-negative BRCA in remission.19 At 12 months, more patients in the placebo group than the tamoxifen group reported depression symptoms. The authors did mention, however, that a relatively high percentage of patients dropped out because of tamoxifen side effects (10% during the first year), which likely included subjects developing iatrogenic depression.
In the current study, we retrospectively identified a cohort of 2,943 patients (ER+, receiving tamoxifen: N=2,329; ER–, not receiving tamoxifen: N=614). We limited our analysis to include only Stage I and II BRCA because of potential differences in the treatment regimens between early and advanced tumors. A Cox proportional-hazards model was used to account for the time-changing variables such as tamoxifen initiation and duration of exposure. By using this model, tamoxifen was found to be associated with a small, but nonsignificant, risk for developing new-onset depression.
We also performed a more inclusive post-hoc analysis, using all patients who could be at risk for developing depression. As mentioned previously, these patients were excluded from the main analysis because of the possibility that they may be inherently different in their risk of developing depression. Chemotherapy and ER+ status were the only factors associated with an increased risk for developing depression. Previous studies also suggest that chemotherapy may be associated with cognitive impairment, depression, anxiety, and physical pain.20–22 Women with high-risk BRCA who received high-dose chemotherapy and tamoxifen were evaluated for cognitive functioning, health-related quality of life (QOL), depression, and anxiety symptoms.21 These women were compared with women who received standard-dose chemotherapy and tamoxifen, and women in Stage I BRCA not receiving tamoxifen (control group). Women receiving high-dose adjuvant chemotherapy (and tamoxifen) had higher scores on the Depression subscale of the Hopkins Symptom Checklist–25 (HSCL–25) than the placebo group (p=0.041). High-dose chemotherapy also seemed to be associated with decreased cognitive functioning (versus controls) as well as decreased QOL scores and greater fatigue.
ER+ status was found to significantly increase the risk of new-onset depression in the present study, independent of tamoxifen exposure. Conflicting reports have been published regarding ER status and psychosocial wellbeing. Razavi et al.23 showed that there was a correlation between ER status and anxiety levels among patients with primary BRCA. ER– patients had significantly higher levels of self-reported anxiety and global distress. Rosenqvist et al.,24 however, could not confirm this. In 92 patients with operable Stage II–III BRCA, there was no correlation between ER status and anxiety levels among these patients.
The major limitation of this study is that it was a non-randomized, retrospective analysis, using claims databases to examine the apparent manifestation of new-onset depression in actual practice. The study design allowed for the inclusion of a large number of subjects, as compared with case–control, open-label, or pilot studies. However, this retrospective approach also necessitated the use of surrogate measures for depression (e.g., new ICD-9 codes and/or antidepressant prescriptions) that would be considerably less sensitive than periodic surveys or diagnostic screening. Although the confirmation of depressive symptoms and severity would have been desirable, confidentiality issues prevented investigators from conducting systematic chart reviews. Also, because diagnostic screening tools are not utilized by all providers and often not captured in claims records, the investigators could not reliably obtain this information.
An additional limitation could be apparent in the menopausal status of these patients. There are well-documented differences in the menopausal status of women with ER+ and ER– tumors. Approximately two-thirds of women with ER+ tumors are postmenopausal, whereas two-thirds of women with ER– tumors are premenopausal.25 Because depression appears to be less common in postmenopausal women, this demographic variable could have affected our outcomes.26 Unfortunately, menopausal status was not consistently captured in claims databases and, therefore, not included in our analysis. Conversely, our methodology represents a relatively conservative method of capturing depression rates, and the increased risk of depression found in ER+ patients becomes more compelling when one considers the lower risk inherently found in older women.
Because the psychosocial impact of tamoxifen treatment has been difficult to ascertain, the health-policy ramifications of this body of research remain unclear. The high incidence of depression with BRCA itself makes it particularly difficult to isolate and quantify the influence of tamoxifen on mood and well-being. Although future research endeavors may identify genetic or environmental predeterminants for tamoxifen-induced depression, the most pertinent message at the moment may reside in the high incidence of depression found in the BRCA population, in general. In the present study, we discovered a prevalence of 20%–27% for depression among our cohort, which is very consistent with previous investigations and significantly higher than the 5%–6% prevalence found in the general population. For now, clinicians and institutions are well advised to implement routine depression screening in this vulnerable population and ensure that coordinated-care models are in place to manage patients suffering from mood disorders.
|
CONCLUSION
|
This retrospective analysis of administrative data failed to find a statistically significant association between the initiation of tamoxifen treatment and the subsequent development of depressive symptoms. Among other factors that were analyzed in this study, chemotherapy and ER+ status appeared to increase the risk for developing new-onset depression in BRCA patients. Future investigations may shed light on the magnitude of these independent risk factors and ultimately guide treatment decisions for BRCA patients in the years to come.
|
ACKNOWLEDGMENTS
|
This study was presented in part at the 156th Annual Meeting of the American Psychiatric Association, San Francisco on May 17–22, 2003, and the 15th Annual Meeting of the Academy of Managed Care Pharmacy, Minneapolis, MN, on April 9–13, 2003.
|
REFERENCES
|
- Howe HL, Wu X, Ries LA, et al: Annual Report to the Nation on the Status of Cancer, 1975-2003, Featuring Cancer Among U.S. Hispanic/Latino Populations. North American Association of Central Cancer Registries, Springfield, IL, (62704-6495); hhowe@naaccr.org
- Feting D: Depression in patients with breast cancer: prevalence, diagnosis, and treatment. Breast J 1997; 3:292–302[CrossRef]
- McDaniel J, Musselman D, Porter M: Depression in patients with cancer: diagnosis, biology, and treatment. Arch Gen Psychiatry 1995; 52:89–99[Abstract]
- Moyer A, Salovery P: Psychosocial sequelae of breast cancer and its treatment. Behav Med 1996; 16:110–125
- Thompson D, Spanier C, Vogel V: The relationship between tamoxifen, estrogen, and depressive symptoms. Breast J 1999; 5:375–382[CrossRef][Medline]
- Bober SL, Hoke LA, Duda RB, et al: Decision-making about tamoxifen in women at high risk for breast cancer: clinical and psychological factors. J Clin Oncol 2004; 22:4951–497[Abstract/Free Full Text]
- Lien EA, Wester K, Lonning PE, et al: Distribution of tamoxifen and metabolites into brain tissue and brain metastases in breast cancer patients. Br J Cancer 1991; 63:641–645[Medline]
- AstraZeneca Pharmaceuticals LP Nolvadex®: Tamoxifen Product Information. Wilmington, DE, 2005
- Patten SB, Barbui C: Drug-induced depression: a systematic review to inform clinical practice. Psychother Psychosom 2004; 73:207–215[CrossRef][Medline]
- Duffy L, Greenberg D, Younger J, et al: Iatrogenic acute estrogen deficiency and psychiatric syndromes in breast cancer patients. Psychosomatics 1999; 40:304–308[Abstract/Free Full Text]
- Cathcart CK, Jones SE, Pumroy CS, et al: Clinical recognition and management of depression in node-negative breast cancer patients treated with tamoxifen. Breast Cancer Res Treat 1993; 27:277–281[CrossRef][Medline]
- Fallowfield L, Fleissig A, Edwards R, et al: Tamoxifen for the prevention of breast cancer: psychosocial impact on women participating in two randomized, controlled trials. J Clin Oncol 2001; 19:1885–1892[Abstract/Free Full Text]
- Oehrli MD, Quesenberry CP, Leyden W, Northern California Cancer Registry, 2005: Annual Report on Trends, Incidence, and Outcomes. Kaiser Permanente Northern California Cancer Registry, Nov. 2005
- Selby JV: Linking automated databases for research in managed-care settings. Ann Intern Med 1997; 127:719–724[Medline]
- Anelli TF, Anelli A, Tran KN, et al: Tamoxifen administration is associated with a high rate of treatment-limiting symptoms in male breast cancer patients. Cancer 1994; 74:74–77[CrossRef][Medline]
- Shariff S, Cumming C, Lees A: Mood disorder in women with early breast cancer taking tamoxifen, an estradiol receptor-antagonist: an expected or unexpected effect? Ann N Y Acad Sci 1995; 761:365–368[Medline]
- Breuer B, Anderson R: The relationship of tamoxifen with dementia, depression, and dependence in activities of daily living in elderly nursing home residence. Women Health 2000; 31:71–85[CrossRef][Medline]
- Day R, Ganz PA, Constantino JP: Tamoxifen and depression: more evidence from The National Surgical Adjuvant Breast and Bowel Project’s breast cancer prevention (P-1) randomized study. J Natl Cancer Inst 2001; 93:1615–1623[Abstract/Free Full Text]
- Love RR, Cameron L, Connell BL, et al: Symptoms associated with tamoxifen treatment in postmenopausal women. Arch Intern Med 1991; 151:1842–1847[Abstract]
- Hughson AV, Cooper AF, McArdle CS, et al: Psychological impact of adjuvant chemotherapy in the first two years after mastectomy. BMJ 1986; 293:1268–1271[Medline]
- van Dam FS, Schagen SB, Muller MJ, et al: Impairment of cognitive function in women receiving adjuvant treatment for high-risk breast cancer: high-dose versus standard-dose chemotherapy. J Natl Cancer Inst 1998; 90:210–218[Abstract/Free Full Text]
- Nystedt M, Berglund G, Bolund C, et al: Side effects of adjuvant endocrine treatment in premenopausal breast cancer patients: a prospective, randomized study. J Clin Oncol 2003; 21:1836–1844[Abstract/Free Full Text]
- Razavi D, Farvacques C, Delvaux N, et al: Psychosocial correlates of oestrogen and progesterone receptors in breast cancer. Lancet 1990; 335:931–933[CrossRef][Medline]
- Rosenqvist S, Berglund G, Bolund C, et al: Lack of correlation between anxiety parameters and oestrogen-receptor status in early breast cancer. Eur J Cancer 1993; 29A:1325-1326
- Bernoux A, de Cremoux P, Laine-Bidron C, et al: Estrogen-receptor negative and progesterone-receptor positive primary breast cancer: pathological characteristics and clinical outcome: Institut Curie Breast Cancer Study Group. Breast Cancer Res Treat 1998; 49:219–225[CrossRef][Medline]
- Bebbington PE, Dunn G, Jenkins R, et al: The influence of age and sex on the prevalence of depressive conditions: report from the National Survey of Psychiatric Morbidity. Psychol Med 1998; 28:9–19[CrossRef][Medline]
This article has been cited by other articles:
|
|
|
|
 
J. T. PALMER and J. L. PAYNE
Stabilization of Hypomania Following Initiation of Tamoxifen
Am J Psychiatry,
May 1, 2008;
165(5):
650 - 651.
[Full Text]
[PDF]
|
|
|
Get information about faster international access.
Privacy Policy
Copyright © 2007
Academy of Psychosomatic Medicine.
All rights reserved.
Home
| Search
| Current Issue
| Past Issues
| Subscribe
| All APPI Journals
| Help
| Contact Us
|
TOP
ABSTRACT
INTRODUCTION
