Article

The Association Between Hormone Receptor Status and End-of-Life Care Among Patients With Metastatic Breast Cancer

Vivian Hui

Rachel Brazee

Margaret Rosenzweig

Young Ji Lee

metastatic breast cancer, hormone receptors, end-of-life care, palliative care
CJON 2022, 26(2), 198-203. DOI: 10.1188/22.CJON.198-203

Background: In metastatic breast cancer (MBC), positive estrogen receptor (ER) and human epidermal growth factor receptor 2 (HER2) status allow for more long-term, sequential treatment options compared to ER-negative and HER2-negative diseases. It is unclear if end-of-life care (timely integration of palliative care, discontinuation of chemotherapy, and enrollment into hospice) in MBC is now tailored to the ER and HER2 status.

Objectives: This article explores the association between ER and HER2 status and the quality of end-of-life care received among patients with MBC.

Methods: A 20-year MBC clinical database captured demographics, tumor characteristics, and treatment histories of deceased patients with MBC (N = 1,258) at a tertiary hospital located in Pittsburgh, Pennsylvania. Descriptive and inferential statistics were used.

Findings: Patients with ER-positive MBC had greater odds of receiving quality end-of-life care than those with ER-negative MBC. HER2 status was not associated with differences in the quality of end-of-life care.

Jump to a section

    One in eight U.S. women will develop invasive breast cancer during the course of her lifetime (Howlader et al., 2019). Of those women, approximately 20%–40% will develop metastatic disease (Metastatic Breast Cancer Network, 2019). In the metastatic breast cancer (MBC) population, tumor subtypes—specifically estrogen receptor (ER) and human epidermal growth factor receptor 2 (HER2)—are predictive of treatment response and metastatic spread, and prognostic for life expectancy. The use of anti-estrogen hormone therapy has allowed for several sequential treatment options, including therapies directed at ER blockade, ER production, cyclin-dependent kinases 4 and 6 inhibition, and others (Silwal-Pandit et al., 2017). Patients with HER2-positive tumors undergo successful MBC treatment with several options for trastuzumab-based therapies. In addition to these therapies, chemotherapy may be offered in a sequential fashion. With all the available treatment options, discussions about end-of-life goals of care may be delayed, with a focus on treatment rather than the trajectory of illness (Brazee et al., 2021; Christakis, 2000; Glare et al., 2003).

    In recent years, many new MBC clinical treatment agents have become available, extending life expectancy beyond two years (Caswell-Jin et al., 2018). Goals-of-care conversations are not consistently conducted and may occur only after several treatment options have already been exhausted.

    Patients with ER-negative and HER2-negative status comprise approximately 15%–25% of all breast cancer cases (Bauer et al., 2007). Because of the inability of treatment focused specifically on unique cellular targets, these patients have fewer treatment options and a worse prognosis (American Cancer Society, 2020; Collignon et al., 2016).

    Despite variance in subtype treatment options, one concern is that the standard of MBC care, including sequential treatment options for most patients with ER-positive and/or HER2-positive tumors, has created an expectation that there is always another treatment option for all tumor subtypes. This could create a tendency to delay conversations about end-of-life care for patients, including those with triple-negative diseases who have fewer treatment options and a shorter prognosis.

    Indicators for poor-quality end-of-life care have been established in advanced cancer care (Earle et al., 2003, 2005). Some of these indicators include receiving any chemotherapy within 14 days of death, lack of palliative care referrals, and lack of hospice care (Earle et al., 2005). These standards were updated in 2017 to include integrated palliative care into metastatic cancer treatments based on a 2010 article demonstrating important end-of-life outcomes with integrated palliative care for patients with non-small cell lung cancer (Ferrell et al., 2017; Temel et al., 2010). The relationship between ER and HER2 with poor-quality indicators for end-of-life care is not well documented in the literature. Without this knowledge, ensuring the provision of consistent, quality end-of-life care for all women with MBC is a challenge.

    The purpose of this study is to explore the association between ER and HER2 status and the presence of poor-quality indicators of end-of-life care experienced among patients with MBC.

    Methods

    Design and Setting

    The study design was a secondary data analysis of a clinical database from a breast cancer program in a National Cancer Institute–designated cancer center in western Pennsylvania. This database is a large, in-use clinical database of patients with MBC that was created in 1999 (Jung et al., 2011). The MBC database was designed to capture demographics, tumor characteristics, and clinical and symptom information of women undergoing routine MBC clinical care at the Magee-Womens Cancer Program of the University of Pittsburgh Medical Center Hillman Cancer Center in Pittsburgh. The MBC database captures 201 variables, which include 38 baseline, sociodemographic, and historic items. This study was approved by the University of Pittsburgh Institutional Review Board at the time of data collection.

    Sample

    Women diagnosed with MBC from January 1999 through December 2020 were recorded in the database. Inclusion criteria were a diagnosis of MBC and having received primary MBC treatment at the Magee-Womens Cancer Program of the University of Pittsburgh Medical Center Hillman Cancer Center. Because the complete course of each patient’s end-of-life care was central to this analysis, only deceased patients were included.

    Data Collection

    Data collection for the ongoing database was obtained through a protocolized assessment and data capture of all patients with MBC. The abstraction is performed by an RN or nurse practitioner from the cancer center. Baseline variables are captured through retrospective chart reviews with regular six-month updates to paper forms. Paper forms are verified and entered in REDCap data storage system (Patridge & Bardyn, 2018). When the patient is deceased, additional variables are captured. The data collection from the primary medical record is run through a double-verification system for both medical record abstraction and entry into REDCap. Any question in data accuracy is resolved by the principal investigator of the MBC database, a certified registered nurse practitioner from the clinic.

    Variables

    Demographic and social variables: Data collected from the MBC database included date of birth, date of metastatic diagnosis, race, and time between the last clinical visit and the patient’s death. The listed zip code was used to calculate the Neighborhood Deprivation Index (NDI), a measure of socioeconomic status based on five census tract-level measures of education, income/poverty, employment, housing, and occupation (Marcus et al., 2018; Messer et al., 2006). The NDI score includes neighborhoods’ state score on a scale of 1–10 and national scores on a scale of 1–100. Higher scores indicate greater socioeconomic deficiencies.

    Quality of end-of-life care: Any chemotherapy administered within 14 days of death, lack of palliative care, and the lack of hospice care within 14 days of death were categorized as poor-quality end-of-life care. The quality of end-of-life care was operationalized by grouping the following two indicators: palliative care and aggressive chemotherapy. Patients with no history of chemotherapy within the last 14 days of life were categorized as no chemotherapy. Patients with no history of hospice care or a palliative care referral were categorized as no palliative care received. There are four ordinal categories to describe the differences in quality of end-of-life care: Patient did not receive palliative care but received aggressive chemotherapy within 14 days prior to death; patient received both palliative care and aggressive chemotherapy within 14 days prior to death; patient received neither palliative care nor aggressive chemotherapy within 14 days prior to death; and patient received palliative care without aggressive chemotherapy within 14 days prior to death. Hormone receptor status was obtained as positive or negative, and then each hormone receptor itself was dichotomized as positive or negative (ER-positive or ER-negative and HER2-positive or HER2-negative). Because of occasional ambiguity regarding HER2 status, particularly because testing for HER2 began in the late 1990s (Zurrida & Veronesi, 2015), the variable was whether anti-HER2 therapy was used in the metastatic setting. This determined if the patient was treated as HER2-positive or HER2-negative, which was the key question of this analysis. 

    Analysis

    To explore the relationship of each hormone receptor and the quality of end-of-life care, a chi-square test was performed independently between each hormone receptor and the patient’s quality of end-of-life care received. For all significant results generated between these two variables independently, ordinal logistic regression was applied to develop the model for prediction.

    Prior to the primary statistical analysis to address study aims, preliminary analyses to screen the data were conducted on the univariate and multivariate levels. First, data were assessed for correct entry through frequency distribution. If there were any data anomalies, the entry data were double checked against the data point recorded in the MBC database to ensure congruence with the original data entry.

    Simultaneously, missing values were evaluated as random or nonrandom, whereas extreme values were reassessed regardless of data entry errors. A missing value analysis in IBM SPSS Statistics, version 25.0, was conducted to determine the amount and pattern of missing data. Dummy variables were created to compare the different qualities of end-of-life care for any missing data. Hot-deck imputation was used to address the missing variables for data that appeared to be missing at random by averaging similar variables’ scores in the sample (Andridge & Little, 2010).

    Univariate analyses were used to describe the frequency and proportion of demographic characteristics (race, age, and state and national NDI scores). Bivariate analyses were conducted with chi-square tests to assess the association between different hormone receptors’ reactivity and the quality of end-of-life care received. Different hormone receptors’ reactivity was compared for quality and poor-quality end-of-life care received by univariate analyses.

    Ordinal logistic regression was used to estimate the odds ratios and 95% confidence interval (CI) for hormone receptors associated with the type of end-of-life care received. The predictors of good- or poor-quality end-of-life care with a p value of less than 0.05 in the bivariate analyses were entered into the ordinal logistic regression. All statistical analyses were conducted using IBM SPSS Statistics, version 25.0.

    Results

    There were 1,789 patients in the database, with 1,258 deceased individuals. Demographics showed a sample size of N = 1,258. The sample included African American (n = 107; 9%), other (n = 31; 2%), and White (n = 1,120; 89%) patients. The mean age was 69.1 years (SD = 13.43). The mean state NDI score was 9.63, and the mean national NDI score was 65.76. A summary of the different hormone receptors in patients with MBC is shown in Table 1, along with sociodemographic characteristics. Overall, 57% (n = 706) of patients were ER-positive, 33% (n = 413) were ER-negative, 16% (n = 202) were HER2-positive, and 62% (n = 775) were HER2-negative.

    Among ER status, there were slight differences between races. White patients comprised 92% of patients who were ER-negative, and 7% of patients who were ER-negative were African American. For patients who were ER-positive, 89% were White and 10% were African American. Patients who were ER-negative were slightly older than patients who were ER-positive (mean = 70.5 years versus mean = 68.7 years, respectively). Patients who were ER-positive and those who were ER-negative shared similar state and national NDI scores. Patients treated as HER2-positive had a lower mean national NDI score than patients who were HER2-negative (mean = 59.5, SD = 25.3 versus mean = 62.95, SD = 24.77, respectively).

    Table 2 displays the univariate, bivariate, and ordinal regression analyses between hormone receptors and the quality of end-of-life care received by patients with MBC. For the univariate comparison, the association between ER status and quality of end-of-life care was significant (p = 0.02). More than 80% (n = 420) of patients who were ER-positive received better-quality end-of-life care, and more than 28% (n = 98) of patients who were ER-negative received poor-quality end-of-life care. Treatment for HER2-positive status was not associated with different quality of end-of-life care (p > 0.05).

    Ordinal logistic regression showed a significantly higher odds ratio to explain the association between different hormone receptor status and the quality of end-of-life care received. Patients with ER-positive MBC had 1.74-fold greater odds of receiving good-quality end-of-life care (exp [0.556] = 1.74; 95% CI [0.306, 0.805]) than those with ER-negative disease. Patients with MBC who were not treated as HER2-positive had slightly better odds of receiving good-quality end-of-life care (exp [0.019] = 1.02; 95% CI [–0.405, 0.443]) than those treated as HER2-positive, although not statistically significant.

    Discussion

    MBC continues to be a progressive, incurable disease with an ever-growing array of treatment options (Sledge, 2016). To extend the length of survival, a treatment paradigm has evolved to include multiple sequential therapies over time, usually with lessening probability of success. In addition, the extensive treatment options available for some disease subtypes can delay prognostic discussions, end-of-life care preparation, and honest communication around the severity of the disease (Bergqvist & Strang, 2019; Schofield et al., 2006). The extensive treatment for patients with MBC is further demonstrated when even at the end of life, patients with metastatic disease continue to receive aggressive, life-prolonging care (Huang et al., 2017; Nipp & Temel, 2017; Smith et al., 2019). Instead, tailored and individualized care should be provided to meet the unique challenges in accordance with the patient’s goals-of-care wishes in the context of the disease subtypes’ clinical course of disease.

    In the total MBC cohort, 16% (n = 201) of patients received palliative care, yet 27% (n = 339) still received chemotherapy within 14 days of death, and 27% (n = 339) received hospice care only 14 days before death. However, because of the evolving treatment paradigm over the course of the total dataset, it is important to examine a smaller, more recent subset. Therefore, a three-year period of patients with MBC (N = 167) from November 2016 to November 2019 showed that 14% (n = 23) received chemotherapy in the last 14 days of life, 58% (n = 97) had three days or less of hospice care prior to death, 37% (n = 61) had death occurring in the acute care setting, and 80% (n = 133) participated in palliative care. Ultimately, results showed that many patients were still receiving aggressive chemotherapy before death. This study found that patients with ER-positive MBC were more likely to have better-quality end-of-life care (receiving palliative care and no aggressive chemotherapy within the last 14 days of life) than patients with ER-negative MBC. Although different MBC subtypes have variable treatment responses and differing lengths of survival, the findings of this study suggest that a blanket-treatment protocol of waiting for the next treatment is being exercised without consideration for patient demographic differences. In addition, sequential therapies may not be in accordance with all the patient’s goals and wishes for quality of end-of-life care.

    Implications for Nursing

    Healthcare providers should assess preferences from patients with MBC for their goals and end-of-life care early in the course of the disease and at regular intervals throughout the disease trajectory. This is particularly important for those with a shorter survival prognosis, such as ER-negative subtypes. In addition, as the intricacies behind quality end-of-life care are recognized, a more personalized prognostic course needs to be taken into greater consideration in clinical application. For example, ER-positive breast cancer is the most common breast cancer subtype and has more treatment options than ER-negative breast cancer (Harb, 2015). A recent study illustrated the diversity of treatment options for patients with ER-positive MBC, such as targeted, endocrine, and chemotherapy treatments (Rozeboom et al., 2019). Patients with ER-negative MBC have fewer treatment options and do not have all the multiple sequential therapies. Healthcare providers may neglect to address their patients’ end-of-life care needs in a timely fashion because of limited treatment options and shorter life expectancy than ER-positive cases. Although patients with ER-negative MBC have a shorter life expectancy and fewer treatment options, they must have timely and tailored end-of-life care.

    In terms of clinical application, a personalized prognostic course should be advocated in hospitals for nurses to evaluate patients’ preferences and treatment options based on hormone subtypes. More practical training for nurses is needed to understand the difference in hormone subtypes and how these subtypes affect patients’ treatment options. Facilitating the communication between patient and clinical nurses is also important to improve the quality of end-of-life care (Temel et al., 2020). Clinical nurses are encouraged to teach and empower patients about their choices so they can make an informed decision before planning end-of-life care. Nurses are encouraged to work with medical social workers or counselors to understand the patient’s preference if the acute clinical setting is not allowed for personalized palliative care treatment.

    Strengths and Limitations

    The data used for this study came from a large sample size with an extensive range of variable types. The length of time covered in this study is more than 20 years, which captured a large amount of data for patients with MBC. Data were abstracted by clinic personnel with a strict data collection protocol. However, there are some limitations that should be noted. The data source is limited to women who underwent primary MBC clinical care at a large urban breast cancer center in western Pennsylvania, limiting the generalizability. Although this is a single-site study, it reflects the homogeneous community of western Pennsylvania’s demographic characteristics. Secondly, the proposed operationalization of quality end-of-life care is based on the literature, which will need to be evaluated on another clinical dataset for validation. Thirdly, the data have been collected since 1999, and there has been an evolution of the palliative care treatment and breast cancer oncology protocols since then. Palliative care has become better integrated into oncology throughout the past two decades. Variables affecting treatment decisions or the acceptance of palliative care, including insurance and mental health status, were not taken into consideration. Future studies can consider the changes of palliative care protocols, include more social factors, and compare the difference between two decades.

    Conclusion

    A relationship was found between hormone receptor status and reactivity and the quality of end-of-life care. Ultimately, ER status was associated with better-quality end-of-life care received among patients with MBC.

    About the Author(s)

    Vivian Hui, BSN, RN, is a PhD student in the Department of Health and Community Systems in the School of Nursing, Rachel Brazee, BSN, RN, is a PhD student, Margaret Rosenzweig, PhD, CRNP-C, AOCNP®, FAAN, is a distinguished service professor in the Department of Acute and Tertiary Care in the School of Nursing, and Young Ji Lee, PhD, RN, is an assistant professor in the Department of Health and Community Systems in the School of Nursing and the Department of Biomedical Informatics in the School of Medicine, all at the University of Pittsburgh in Pennsylvania. The authors take full responsibility for this content. This study was supported by a grant to Lee from the Center for Research and Evaluation Pilot/Feasibility Study Program in School of Nursing at the University of Pittsburgh, and the Central Research Development Fund at the University of Pittsburgh. The article has been reviewed by independent peer reviewers to ensure that it is objective and free from bias. Rosenzweig can be reached at mros@pitt.edu, with copy to CJONEditor@ons.org. (Submitted May 2021. Accepted November 16, 2021.)

     

    References

    American Cancer Society. (2020). How common is breast cancer? https://www.cancer.org/cancer/breast-cancer/about/how-common-is-breast-…
    Andridge, R.R., & Little, R.J.A. (2010). A review of hot deck imputation for survey non-response. International Statistical Review, 78(1), 40–64. https://doi.org/10.1111/j.1751-5823.2010.00103.x
    Bauer, K.R., Brown, M., Cress, R.D., Parise, C.A., & Caggiano, V. (2007). Descriptive analysis of estrogen receptor (ER)-negative, progesterone receptor (PR)-negative, and HER2-negative invasive breast cancer, the so-called triple-negative phenotype: A population-based study from the California cancer Registry. Cancer, 109(9), 1721–1728. https://doi.org/10.1002/cncr.22618
    Bergqvist, J., & Strang, P. (2019). Breast cancer patients’ preferences for truth versus hope are dynamic and change during late lines of palliative chemotherapy. Journal of Pain and Symptom Management, 57(4), 746–752. https://doi.org/10.1016/j.jpainsymman.2018.12.336
    Brazee, R.L., Sereika, S.M., & Rosenzweig, M.Q. (2021). Prevalence, pattern, and probability for goals of care discussions among women diagnosed with metastatic breast cancer. Journal of Cancer Survivorship, 15(3), 375–379. https://doi.org/10.1007/s11764-021-01022-w
    Caswell-Jin, J.L., Plevritis, S.K., Tian, L., Cadham, C.J., Xu, C., Stout, N.K., . . . & Kurian, A.W. (2018). Change in survival in metastatic breast cancer with treatment advances: Meta-analysis and systematic review. JNCI Cancer Spectrum, 2(4), pky062. https://doi.org/10.1093/jncins/pky062
    Christakis, N.A. (2000). Extent and determinants of error in doctors’ prognoses in terminally ill patients: Prospective cohort study. BMJ, 320(7233), 469–473. https://doi.org/10.1136%2Fbmj.320.7233.469
    Collignon, J., Lousberg, L., Schroeder, H., & Jerusalem, G. (2016). Triple-negative breast cancer: Treatment challenges and solutions. Breast Cancer, 2016(8), 93–107. https://doi.org/10.2147/BCTT.S69488
    Earle, C.C., Neville, B.A., Landrum, M.B., Souza, J.M., Weeks, J.C., Block, S.D., . . . Ayanian, J.Z. (2005). Evaluating claims-based indicators of the intensity of end-of-life cancer care. International Journal for Quality in Health Care, 17(6), 505–509. https://doi.org/10.1093/intqhc/mzi061
    Earle, C.C., Park, E.R., Lai, B., Weeks, J.C., Ayanian, J.Z., & Block, S. (2003). Identifying potential indicators of the quality of end-of-life cancer care from administrative data. Journal of Clinical Oncology, 21(6), 1133–1138. https://doi.org/10.1200/JCO.2003.03.059
    Ferrell, B.R., Temel, J.S., Temin, S., Alesi, E.R., Balboni, T.A., Basch, E.M., . . . Smith, T.J. (2017). Integration of palliative care into standard oncology care: American Society of Clinical Oncology clinical practice guideline update. Journal of Clinical Oncology, 35(1), 96–112. https://doi.org/10.1200/jco.2016.70.1474
    Glare, P., Eychmueller, S., & Virik, K. (2003). The use of the palliative prognostic score in patients with diagnoses other than cancer. Journal of Pain and Symptom Management, 26(4), 883–885. https://doi.org/10.1016%2Fs0885-3924%2803%2900335-x
    Harb, W. (2015). Management of patients with hormone receptor-positive breast cancer with visceral disease: Challenges and treatment options. Cancer Management and Research, 7, 37–46. https://doi.org/10.2147/CMAR.S72592
    Howlader N., Noone, A.M., Krapcho, M., Miller, D., Brest, A., Yu, M., . . . Cronin, K.A. (Eds.). (2019). SEER cancer statistics review, 1975-2016. National Cancer Institute. https://seer.cancer.gov/csr/1975_2016/
    Huang, H.K., Hsieh, J.G., Hsieh, C.J., & Wang, Y.W. (2017). Do cancer patients with dementia receive less aggressive treatment in end-of-life care? A nationwide population-based cohort study. OncoTarget, 8(38), 63596–63604. https://doi.org/10.18632/oncotarget.18867
    Jung, S.Y., Sereika, S.M., Linkov, F., Brufsky, A., Weissfeld, J.L., & Rosenzweig, M. (2011). The effect of delays in treatment for breast cancer metastasis on survival. Breast Cancer Research and Treatment, 130(3), 953–964. http://doi.org/10.1007/s10549-011-1662-4
    Marcus, J.L., Hurley, L.B., Chamberland, S., Champsi, J.H., Gittleman, L.C., Korn, D.G., . . . Silverberg, M.J. (2018). Disparities in initiation of direct-acting antiviral agents for hepatitis C virus infection in an insured population. Public Health Reports, 133(4), 452–460. https://doi.org/10.1177/0033354918772059
    Messer, L.C., Laraia, B.A., Kaufman, J.S., Eyster, J., Holzman, C., Culhane, J., . . . O’Campo, P. (2006). The development of a standardized neighborhood deprivation index. Journal of Urban Health, 83(6), 1041–1062. https://doi.org/10.1007/s11524-006-9094-x
    Metastatic Breast Cancer Network. (2019). Statistics for metastatic breast cancer. http://mbcn.org/statistics-for-metastatic-breast-cancer
    Nipp, R.D., & Temel, J.S. (2017). Multimorbidity and aggressiveness of care at the end-of-life among older adults with cancer. Journal of Geriatric Oncology, 8(2), 82–83. https://doi.org/10.1016/j.jgo.2017.01.005
    Patridge, E.F., & Bardyn, T.P. (2018). Research Electronic Data Capture (REDCap). Journal of the Medical Library Association, 106(1), 142–144. https://doi.org/10.5195/jmla.2018.319
    Rozeboom, B., Dey, N., & De, P. (2019). ER+ metastatic breast cancer: Past, present, and a prescription for an apoptosis-targeted future. American Journal of Cancer Research, 9(12), 2821–2831.
    Schofield, P., Carey, M., Love, A., Nehill, C., & Wein, S. (2006). “Would you like to talk about your future treatment options?” Discussing the transition from curative cancer treatment to palliative care. Palliative Medicine, 20(4), 397–406. https://doi.org/10.1191/0269216306pm1156oa
    Silwal-Pandit, L., Nord, S., von der Lippe Gythfeldt, H., Møller, E.K., Fleischer, T., Rødland, E., . . . Engebraaten, O. (2017). The longitudinal transcriptional response to neoadjuvant chemotherapy with and without bevacizumab in breast cancer. Clinical Cancer Research, 23(16), 4662–4670. https://doi.org/10.1158/1078-0432.CCR-17-0160
    Sledge, G.W., Jr. (2016). Curing metastatic breast cancer. Journal of Oncology Practice, 12(1), 6–10. http://doi.org/10.1200/JOP.2015.008953
    Smith, C.E.P., Kamal, A.H., Kluger, M., Coke, P., & Kelley, M.J. (2019). National trends in end-of-life care for veterans with advanced cancer in the Veterans Health Administration: 2009 to 2016. Journal of Oncology Practice, 15(6), e568–e575. https://doi.org/10.1200/JOP.18.00559
    Temel, J.S., Greer, J.A., Muzikansky, A., Gallagher, E.R., Admane, S., Jackson, V.A., . . . Lynch, T.J. (2010). Early palliative care for patients with metastatic non–small-cell lung cancer. New England Journal of Medicine, 363(8), 733–742. https://doi.org/10.1056/nejmoa1000678
    Temel, J.S., Moy, B., El-Jawahri, A., Jackson, V.A., Kamdar, M., Jacobsen, J., . . . Greer, J.A. (2020). Randomized trial of a collaborative palliative and oncology care intervention to improve communication about end-of-life care in patients with metastatic breast cancer [Abstract 1008]. Journal of Clinical Oncology, 38(15, Suppl.). http://doi.org/10.1200/JCO.2020.38.15_suppl.1008
    Zurrida, S., & Veronesi, U. (2014). Milestones in breast cancer treatment. Breast Journal, 21(1), 3–12. https://doi.org/10.1111%2Ftbj.12361