|Home | About | Journals | Submit | Contact Us | Français|
Up to one-quarter of breast cancer patients suffer clinically significant depression in the year after diagnosis, which may respond to intervention. About half may be prescribed a psychotropic medication, such as a selective serotonin reuptake inhibitor (SSRI), while completing breast cancer therapy. Cytochrome P-450 2D6 (CYP2D6) metabolizes SSRIs and also metabolizes tamoxifen to more active forms. Therefore, concurrent use of SSRIs may reduce tamoxifen’s effectiveness at preventing breast cancer recurrence. The SSRI citalopram has limited potency to inhibit CYP2D6 activity, so has been recommended for breast cancer patients taking tamoxifen. This study provides epidemiologic evidence to support this recommendation.
We conducted a case-control study of breast cancer recurrence nested in the population of female residents of Denmark who were diagnosed with non-metastatic estrogen-receptor positive breast cancers between 1994 and 2001 and who took tamoxifen for at least one year. We ascertained complete prescription histories by linking cases’ and controls’ civil registration numbers to the Danish national prescription registry. We estimated the association between SSRI use while taking tamoxifen and risk of recurrent breast cancer.
About the same proportion of recurrent cases (37 of 366) and matched controls (35 of 366) received at least one prescription for citalopram or its s-stereoisomer while taking tamoxifen (adjusted odds ratio = 1.1, 95% confidence interval =0.7, 1.7). Breast cancer patients taking other SSRIs were also at no increased risk of recurrence (adjusted odds ratio = 0.9, 95% confidence interval = 0.5, 1.8).
Breast cancer patients with indications for an SSRI may be prescribed citalopram — and possibly other SSRI — without adversely affecting the outcome of adjuvant therapy with tamoxifen.
Almost all newly diagnosed breast cancer patients experience normal distress . Up to one-quarter, however, suffer clinically significant depression in the year after diagnosis, which may respond to interventions . Although no study has yet examined the effectiveness of psychotherapy and psychiatric drug therapy in cancer patients , the prevalence of prescriptions for psychotropic drugs among patients treated for breast cancer is high. For example, about half of the patients in a breast cancer waiting-room sample had received psychotropic medication during their breast cancer treatment . One class of psychotropic medications, selective serotonin reuptake inhibitors (SSRI), may reduce both depressive symptoms and menopausal symptoms [4, 5].
SSRIs are metabolized by cytochrome P450 2D6 (CYP2D6) , as is the selective estrogen-receptor modulator tamoxifen . Tamoxifen reduces the risk of breast cancer recurrence by about half in patients with estrogen-receptor positive tumors . Tamoxifen’s two 4-hydroxylated metabolites have the highest binding affinity for the estrogen receptor, and are the most important modulators of the estrogen receptor in the tamoxifen pathway [9, 10]. Because both tamoxifen and SSRIs are metabolized by CYP2D6, SSRI inhibition of CYP2D6 activity could reduce tamoxifen’s prevention of breast cancer recurrence. Citalopram is among the least potent SSRI inhibitors of CYP2D6 activity [6, 11], which led Henry et al. to recommend citalopram (or similar low-inhibiting venlafaxine) for treatment of breast cancer patients taking tamoxifen . Aside from a preliminary report from this research group , no clinical epidemiologic evidence has shown that these SSRI do not interfere with tamoxifen’s effect on breast cancer recurrence risk.
We previously reported that Danish breast cancer patients with estrogen-receptor positive tumors who were treated with tamoxifen had no higher rate of recurrence if they were simultaneously taking the SSRI citalopram or its s-stereoisomer than if they were not . This initial study was limited to four counties with local prescription registries. We have now extended the study to eight counties and use Danish national prescription data to ascertain exposure to SSRI and to control for exposure to a wide range of medications. The current study has substantially improved precision and has allowed more complete investigation of other CYP2D6-inhibiting SSRI medications.
The study was approved by the Boston University Medical Campus Institutional Review Board and the Regional Committee on Biomedical Research Ethics of Aarhus County, Denmark. Because the data are housed in medical registries, individual informed consent was not obtained.
The source population included female residents of eight Danish counties (Funen, South Jutland, Ribe, Vejle, Ringkøbing, Aarhus, Viborg, and North Jutland) 35–69 years old at diagnosis of stage I, II or III primary breast cancer between 1994–2001 and who were reported to the Danish Breast Cancer Cooperative Group (DBCG; ). We divided the source population into three groups: (a) ER+/TAM+ — estrogen-receptor positive and treated with tamoxifen for at least one year without recurrence in that year, (b) ER−/TAM− —estrogen-receptor negative, not treated with tamoxifen, and survived recurrence-free at least one year, and (c) group III women —all others, including patients who recurred in the first year, ER+ patients who did not receive tamoxifen, and ER− patients who did receive tamoxifen, all of whom were excluded from this analysis. Estrogen receptor expression was assayed at diagnosing hospitals by standard DBCG protocols. Clinical assay of estrogen receptor expression in pathology laboratories has shown high concordance with centralized testing in similar settings . ER+/TAM+ women were assigned to tamoxifen therapy protocols of 1 year, 2 years, or 5 years, depending on the guideline current in Denmark when they were diagnosed . Many of the women assigned to tamoxifen protocols shorter than five years took tamoxifen for much longer (unpublished validation data). Follow-up time began one year after breast cancer diagnosis and continued until the date of the first of breast cancer recurrence, death from any cause, loss to follow-up (e.g., emigration), ten years of follow-up, or 1 September 2006.
Cases were women with local or distant breast cancer recurrence during their follow-up time. We used the DBCG definition of breast cancer recurrence as any type of breast cancer subsequent to the initial course of therapy. Using risk-set sampling, we matched one control to each case on (a) group membership (ER+/TAM+ or ER−/TAM−), (b) menopausal status at diagnosis (premenopausal or postmenopausal), (c) date of breast cancer surgery (caliper matched +/− twelve months), (d) county of residence at time of diagnosis, and (e) stage at diagnosis (stage I, II, or III). Controls were free of breast cancer recurrence at the same duration of post-surgery follow-up as their matched case. It was not possible to match controls to cases on duration of tamoxifen therapy, but the calendar time matching induced by risk-set sampling afforded good balance between cases and controls with regard to the duration of assigned tamoxifen protocol.
We used the Danish civil registration number (CPR) assigned to all Danish citizens and residents to link data sets. We collected demographic information (age, menopausal status, and hospital of diagnosis), tumor characteristics (UICC stage, histologic grade, and estrogen-receptor expression), and therapy characteristics (primary surgical tumor management, receipt of radiation therapy, receipt of chemotherapy, and receipt of tamoxifen therapy) from the DBCG database. We collected data on receipt of citalopram prescriptions, prescriptions for other SSRIs, and prescriptions for other potential CYP2D6 inhibitors by linking the CPR numbers of cases and controls to the prescription database maintained by Statistics Denmark as a component of the Danish national health care system.
Prescription medications were coded by the Anatomical Therapeutic Chemical (ATC) classification system . We defined SSRIs as all those classified in ATC group N06AB. We combined prescriptions for citalopram with prescriptions for its s-enantiomer escitalopram because escitalopram inhibits CYP2D6 activity similarly to citalopram in vitro  and both have been shown to inhibit CYP2D6 activity in vivo [11, 19]. We classified cases and controls as those with no record of a citalopram prescription during their follow-up time (never citalopram) and those with any record of prescription for citalopram during their follow-up time (ever citalopram). We used a similar procedure to classify cases and controls as ever or never users of another SSRI or of another prescription medication that is a CYP2D6 inhibitor or substrate.
For ER+/TAM+ women who ever had a citalopram prescription, we calculated the percentage of time on tamoxifen during which they were simultaneously taking citalopram. We created categories of (a) intermittent citalopram use, defined as citalopram use overlapping tamoxifen use for more than 0% but less than 30% of the time on tamoxifen, and (b) regular citalopram use, defined as citalopram use overlapping tamoxifen use for 30% or more of the time on tamoxifen. For this analysis, we used the full duration of their tamoxifen use as recorded in the DGCG registry, which was often longer than the duration anticipated by their original protocol assignment.
We defined the following set of covariates: time period of breast cancer diagnosis, age at diagnosis, menopausal status at diagnosis, county of residence at diagnosis, UICC stage at diagnosis, histologic grade, surgery type, receipt of systemic adjuvant chemotherapy, and receipt of a prescription for another medication that is a CYP2D6 inhibitor or substrate while taking tamoxifen (aside from those used to treat breast cancer recurrence or its effects).
We performed all analyses within strata of ER+/TAM+ and ER−/TAM− women. We calculated the number of cases and controls ever receiving each SSRI, the number of total prescriptions for each SSRI summed over all cases or controls, and the range of the number of prescriptions for each SSRI received by each individual case or control. Table 1 gives a complete list of SSRI medications and the frequency of their use in the study population. We also classified cases and controls as ever or never users of another prescription medication that is a CYP2D6 inhibitor or substrate. Table 2 gives a complete list of these medications and the frequency of their use in the study population.
We then computed the frequency and proportion of cases and controls within categories of assigned protocol of tamoxifen duration, citalopram use, use of other SSRIs, use of other CYP2D6 inhibitors or substrates, and the covariates.
We estimated the rate ratio associating citalopram prescription with breast cancer recurrence as the odds ratio (OR) and its accompanying 95% confidence interval (CI) in a conditional logistic regression including only citalopram use as the exposure variable and conditioned on the matched factors. We then adjusted for additional confounding by covariates not included in the matching by including them as independent variables in a conditional logistic regression, retaining any covariate that affected the log odds ratio by more than ten percent. All analyses were performed using SAS version 9.
This research was supported by grants from the US National Cancer Institute (R01 CA118708), the Danish Cancer Society (DP06117), and the Karen Elise Jensen Foundation. The Department of Clinical Epidemiology at Aarhus University Hospital is involved in studies with funding from various companies as research grants to, and administered by, Aarhus University. These include a grant from the Lundbeck Foundation to study meningococcal disease and collaborations with the Centre for Registry Research, which receives grants from H. Lundbeck A/S (the manufacturer of citalopram and escitalopram). None of these studies have direct relation to the present study or supported any of the work reported herein. No author has any other potential competing interest.
Table 1 shows the pattern of SSRI prescriptions received by cases and controls. In both ER+/TAM+ and ER−/TAM− women, SSRI prescriptions were primarily written for citalopram or escitalopram. Table 2 shows the frequency and proportion of cases and controls who received prescriptions for other CYP2D6 substrates and inhibitors, within strata of ER+/TAM+ and ER−/TAM−. These frequencies and proportions were approximately the same among cases and controls, varying only as expected due to chance .
Table 3 shows the frequency and proportion of cases and controls, within strata of ER+/TAM+ and ER−/TAM−, in the categories of the covariates. Ten percent of ER+/TAM+ cases and 10% of ER+/TAM+ controls ever used citalopram while taking tamoxifen and about 6% of ER+/TAM+ cases and 6% of their controls ever used another SSRI while taking tamoxifen.
ER+/TAM+ women who ever used citalopram while taking tamoxifen had about the same rate of breast cancer recurrence as women who never used citalopram while taking tamoxifen (Table 4; adjusted OR=1.1, 95% CI=0.7, 1.7). These near-null results persisted within categories of intermittent and regular users of citalopram while taking tamoxifen. ER+/TAM+ women who ever used another SSRI (fluoxetine, paroxetine, or sertraline) while taking tamoxifen were also at no increased risk of breast cancer recurrence (Table 4; adjusted OR=0.9, 95% CI=0.5, 1.8). Neither citalopram use (adjusted OR=0.9, 95% CI=0.4, 2.2) nor use of another SSRI (adjusted OR=0.6, 95% CI=0.3, 1.6) had a substantial effect on recurrence in ER−/TAM− women, suggesting that these SSRI medications do not directly affect the risk of breast cancer recurrence.
The results of this study provide clinical epidemiologic support for the hypothesis that citalopram, taken concurrently with tamoxifen, does not reduce tamoxifen’s protective effect against breast cancer recurrence in early stage patients whose tumor cells express the estrogen receptor. This support is in agreement with recent recommendations that tamoxifen-treated breast cancer patients with indications for antidepressant medications may be safely prescribed citalopram or another SSRI with low potency to inhibit CYP2D6 activity , and fills a void in the evidence base identified by the US National Comprehensive Cancer Network’s treatment guidelines .
Most SSRI prescriptions in our study were for citalopram or its s-stereoisomer, which is a modest inhibitor of CYP2D6 compared with some other SSRI medications [11, 18, 19]. Use of other SSRI medications (fluoxetine, paroxetine, or sertraline) while taking tamoxifen, some of which are more potent inhibitors of CYP2D6 [11, 22], was also unassociated with recurrence risk in our results. The frequencies of prescriptions for these other SSRIs were, however, too low to say with confidence that they do not reduce the effectiveness of tamoxifen.
This study extends our earlier results  by including 366 ER+/TAM+ cases and their 366 matched controls, resulting in 37 cases and 35 controls who ever used citalopram while taking tamoxifen. The earlier study included only 184 ER+/TAM+ cases and their 184 matched controls, resulting in only 17 cases and 21 controls who ever used citalopram while taking tamoxifen. Fifty-six percent of ER+/TAM+ cases and controls in this study were included in the earlier study, and 46% of citalopram-exposed cases were included in the earlier study. The present study’s null result is, therefore, much more precisely measured than the null result of the earlier study. In addition, the large sample size and comprehensive prescription registry allowed investigation of, and control for, exposure to a wide range of prescription medications.
Despite the study’s size and methodologic strength as a population-based case-control study, the results should be interpreted with the following limitations in mind. First, we do not know the reasons why SSRIs were prescribed to the study participants. SSRIs may have been prescribed to treat either depression or hot flashes , but SSRI prescriptions for hot flashes are very rare in Danish breast cancer patients. Second, we do not know whether participants carried CYP2D6 variant alleles that reduce the enzyme’s activity. Genetic variation in CYP2D6 function, however, is not related to switching SSRI antidepressants or discontinuation of SSRI antidepressants , and does not affect response to, or tolerance of, citalopram in particular . If CYP2D6 genotype is unrelated to receipt or adherence to citalopram prescription, then the absence of genotyping data could not bias the results. Furthermore, clinicians caring for breast cancer patients who present with indications for SSRI antidepressants will seldom know the patient’s CYP2D6 genotype, so this study’s result applies directly to the typical clinical setting.
Third, we have not confirmed that patients actually took either tamoxifen or a prescribed SSRI. In Denmark, tamoxifen is dispensed by breast cancer physicians to breast cancer patients at follow-up visits. SSRI medications recorded in the prescription registry are paid for and retrieved by patients, and then partly reimbursed by the national health care system. Both of these systems should assure good adherence to the registered medications. Fourth, most women taking SSRI prescription medications did not take them for the full duration of their tamoxifen therapy. This pattern reflects the clinical practice in this population during the study period. It would be very difficult to find a population in which a substantial proportion of tamoxifen-treated breast cancer patients took SSRI medications for the full five years of their tamoxifen therapy. Indeed, no such study has been reported. Finally, breast cancer patients with estrogen-receptor positive tumors were assigned treatment protocols calling for one, two, or five years of tamoxifen therapy, whereas current guidelines recommend five years of tamoxifen therapy . Many of the women assigned to tamoxifen protocols shorter than five years took tamoxifen for much longer (unpublished validation data), and we recorded the full duration of their use in the analysis of intermittent and regular use. In addition, recurrence risks between tamoxifen-treated and placebo-treated women differ as early as one year after initiation of tamoxifen treatment , so inhibition of tamoxifen effectiveness by concurrent SSRI prescriptions should have been apparent among all women included in our study.
While these results may seem at odds with the strong biologic rationale and in vivo evidence supporting the hypothesis that any CYP2D6 inhibition would reduce tamoxifen’s effectiveness, this information may not be as compelling as it first seems . SSRI medications could reduce the plasma concentration of tamoxifen’s secondary metabolites without reducing its anti-tumorigenicity . Tamoxifen doses as low as 1 mg/day affect biomarkers of cardiovascular, bone, and tumor endpoints to about the same degree as the usual dose of 20 mg/day [28, 29], so the three-fold reduction in the concentration of tamoxifen’s secondary metabolites associated with receipt of the SSRI paroxetine  may have little consequence. Our results, combined with this emerging alternative view of the limited potential for CYP2D6 inhibition to interact with tamoxifen, suggest that breast cancer patients with indications for an SSRI may be prescribed citalopram while taking tamoxifen with little effect, if any, on their risk of breast cancer recurrence.