Antidepressants and Breast Cancer Treatment Interactions

Institution: Kaiser Foundation Research Institute
Investigator(s): Reina Haque, Ph.D. -
Award Cycle: 2008 (Cycle 14) Grant #: 14IB-0086 Award: $156,068
Award Type: IDEA
Research Priorities
Etiology and Prevention>Etiology: the role of environment and lifestyle



Initial Award Abstract (2008)

Thousands of American women with breast cancer are taking tamoxifen to reduce their chances of developing a recurrence or a second primary cancer. Tamoxifen has been on the market for nearly 25 years and the manufacturer claims an exposure history of 10 million patient years. Despite its success in reducing cancer risk and recurrence, tamoxifen's serious side effects include stroke and venous thrombosis. Non-fatal but notable side effects that diminish a woman's quality of life and quality of sleep include vasomotor effects (hot flashes and night sweats) and depression. Because hormone replacement therapy commonly used to alleviate these symptoms is contraindicated for women with breast cancer, antidepressants increasingly have been used to relieve clinical depression and the hot flashes caused by tamoxifen.

Our interest is in the relationship between combined tamoxifen and antidepressant medication use and breast cancer recurrence. A few recent laboratory-based studies suggest that certain common antidepressants, such as Paxil, may interfere with tamoxifenís effectiveness in preventing breast cancer recurrence. Some recent reports caution physicians not to prescribe common antidepressant medications with tamoxifen. Nevertheless, no compelling studies on health outcomes have been published to demonstrate the ability of antidepressant drugs to diminish tamoxifenís effectiveness.

To address whether women are at an elevated risk for subsequent breast cancer (recurrence or second primary) when using certain antidepressants, we will assemble a large group of women with diverse ages, incomes, and racial/ethnic backgrounds from the membership of Kaiser Permanente Southern California (KPSC) health plan who were diagnosed with their first breast cancer lesion from 1996 to 2006. We will follow them over time to determine how many of these patients developed subsequent breast cancer. We plan to examine breast cancer risk in two main groups of women: (1) those who were treated with both antidepressants and tamoxifen after their initial disease, and (2) women who were exposed to tamoxifen only (this will be the comparison group). We will analyze whether breast cancer recurrence varies by types of antidepressants, and identify factors that may confound or modify this association such as demographics, tumor characteristics, comorbidities (additional disorders or diseases), or the use of other medications.

As both depression and hot flashes are extremely common in breast cancer patients, information about the safety or the adverse effects of combined tamoxifen and antidepressant use is crucial. Given the new concerns of this potential adverse drug interaction, this study is well-timed.




Final Report (2010)

Despite tamoxifen's success in reducing the risk of breast cancer recurrence, its notable side effects include hot flashes and night sweats. Antidepressants have been used to relieve some of these symptoms. However, few laboratory studies suggest that certain common antidepressants may interfere with tarnoxifen's effectiveness. Our goal was to determine whether concomitant tamoxifen and antidepressant use is associated with breast cancer mortality as well as all cause mortality.

We assembled a large cohort of women who were diagnosed with their first breast cancer in 1996-2006 & treated with tamoxifen and followed through 12/31 /07 using the KPSC-SEER affiliated tumor registry. We then examined rates of mortality by antidepressant groups (paroxetine, fluoxetine, other selective serotonin re-uptake inhibitors (SSRls), combined: tricyclics, and multiple types). The comparison group included women who did not use antidepressants (i.e., only tamoxifen).

We identified 7,548 ER+ early stage (0-11) breast cancer survivors. A total of 1,253 deaths were observed (394 due to breast cancer). In the multi variable Cox models, no association between paroxetine use and breast cancer mortality was found (HR=1.27. 95% CI: 0.63 - 2.58). However. we found associations between fluoxetine (HR=2.08, 95% CI I1.48 Ė 2.93) and other SSRls (HR=2.82. 95% CI: 1.59 -4.99) and all-cause mortality. The associations for these antidepressants and all-cause mortality were even stronger among the subset of women with good adherence to tamoxifen. These estimates were based on small numbers of deaths and should be interpreted with caution.

Our results suggest that concomitant use of tamoxifen and specific antidepressants may affect survival, but these preliminary data need to be evaluated in larger samples of breast cancer survivors. Thus, we were able to leverage this initial IDEA project into a larger full-scale study. This new project will investigate whether combined tamoxifen and antidepressant medication use increases the risk of recurrence and new primaries in women with early stage breast cancer. This multi-site study will include over 16,000 breast cancer survivors living in both regions of the state. The new collaborative project includes investigators from Kaiser Permanente Southern California, Kai ser Permanente Northern California, and Harvard Medical School, and was recently funded by the NlH/NCI (RO1 CA 136743-01A2).

Publications:
Taking antidepressants with cancer drug does not increase breast-cancer recurrence
Tamoxifen and Antidepressant Drug Interaction in a Cohort of 16 887 Breast Cancer Survivors




Symposium Abstract (2010)

Reina Haque (PI), Chantal Avila, Jiaxiao Shi, Joanie Chung, Craig Cheetham, Virginia P. Quinn
Kaiser Permanente Southern California

Overview of the research topic and relevance to breast cancer:† Thousands of women diagnosed with breast cancer (BCa) are taking tamoxifen to improve their chances of survival. Despite its success in reducing the risk of future BCa, tamoxifen's side effects include stroke and blood clots, hot flashes, and night sweats. Increasingly, antidepressants have been used to relieve some of these symptoms. Of concern, a few recent laboratory-based studies suggest that certain common antidepressants called selective serotonin re-uptake inhibitors (SSRIs) interfere with tamoxifen's effectiveness. Recent reports even caution physicians not to prescribe these antidepressant medications with tamoxifen. Very few studies have attempted to assess the effect of the interaction between antidepressants with tamoxifen on long-term outcomes.

The research question: The overarching goal of the study is to determine whether women are at an elevated risk for overall death or BCa death when they use antidepressants while receiving tamoxifen therapy.

General methods: We assembled a large group of socioeconomically diverse women from the membership of the Kaiser Permanente Southern California health plan who were diagnosed with their first BCa in 1996-2006 and followed through 12/31/07 to determine how many of these patients died. We then compared the mortality rates among mutually exclusive groups of antidepressant users: (1) paroxetine, (2) fluoxetine, (3) other SSRIs (sertaline & citalopram), (4) tricyclics, (5) other types, (6) multiple types, and (7) a comparison group of non-users of antidepressants (i.e., only tamoxifen).

Preliminary results: We identified 12,835 ER+ early stage (0-II) BCa survivors from the SEER-affiliated tumor registry. After excluding ineligibles (not continuously enrolled (n=1343) or no exposure to tamoxifen (n=3944)), we had 7,548 women. Cause of death through 12/31/07 included BCa (n=394, 5.2%); cardiovascular disease (n=183, 2.4%); other causes (n=408, 5.4%); and unknown causes (n=268, 3.6%); leaving 6,295 (83.4%) women alive by end of follow-up. We examined crude and adjusted rates of Bca and overall mortality using Cox proportional hazards modeling. We adjusted for age at diagnosis, year, race/ethnicity, income, medical utilization, comorbidity, stage, primary and adjuvant cancer therapy, and other tumor characteristics (histology, grade, lymph nodes, PR). We did not observe a significant association between paroxetine use and mortality (HR=1.29, 95% CI: 0.91-1.82). However, we did find significant associations between fluoxetine (HR=1.47, 95% CI: 1.15-1.89) and other SSRIs (HR=2.36, 95% CI: 1.39-4.01) with overall mortality. Among women with good adherence to tamoxifen (80% medication possession ratio), the association between fluoxetine and other SSRIs with all cause mortality became stronger. Of note, these estimates were based on small numbers of deaths and need to be interpreted with caution.

Impact & next steps: Our results suggest that concomitant use of tamixofen and some antidepressants may affect survival. Although we were able to adjust for a numerous potential confounders, these preliminary data need to be evaluated in larger samples of BCa survivors.



Tamoxifen and Antidepressant Drug Interaction in a Cohort of 16 887 Breast Cancer Survivors 10.1093/jnci/djv337
Periodical:Journal of the National Cancer Institute
Index Medicus: J Natl Cancer Inst
Authors: Reina Haque, Jiaxiao Shi, Joanne E. Schottinger, Syed A. Ahmed, et al.
Yr: 2015 Vol: 108 Nbr: 3 Abs: Pg: