Patient Perspectives on Treatment Options for Older Women With Hormone Receptor–Positive Breast Cancer

Key Points Question What are older women’s perspectives on recommendations for treatment de-escalation for low-risk, early-stage hormone receptor–positive breast cancer? Findings In this qualitative study of 30 participants aged 70 years or older, women expressed skeptical views regarding age-based treatment guidelines, difficulty interpreting the rationale for reducing low-value care to be a favorable rather than poor prognosis, and perceived benefit of some low-value therapies for peace of mind. Meaning Emphasizing an overall favorable prognosis and improving patient education on the risks vs benefits of adjuvant therapies may help reduce overtreatment in older women with early-stage, hormone receptor–positive breast cancer.


Introduction
More than one-third of patients with newly diagnosed breast cancer are aged 70 years or older. 1 Most of these patients receive a diagnosis at an early stage, and their cancers have the favorable characteristic of expressing estrogen and progesterone receptors (ie, they are hormone receptor [HR] positive). These tumors carry an excellent long-term prognosis, and the probability of a woman aged 70 years or older dying from breast cancer is less than 1%. 2 At the same time, older patients are more vulnerable to the risks and toxicities of cancer treatments, placing these women at high risk of overtreatment where potential harms outweigh potential benefits. 3 Several studies [4][5][6][7][8] have tested the safety of omitting previously routine therapies in women aged 70 years or older with early-stage, HR-positive cancer. In 2013, the Cancer and Leukemia Group B 9343 trial 4 found that women who underwent postlumpectomy radiotherapy with or without axillary staging experienced no additional survival benefit compared with women treated with lumpectomy and endocrine therapy alone. On the basis of these and other data and as part of the Choosing Wisely campaign, the Society of Surgical Oncology recommended against routine axillary staging with sentinel lymph node biopsy (SLNB) in clinically node-negative women aged 70 years or older with HR-positive breast cancer in 2016. [5][6][7] Similarly, the National Comprehensive Cancer Network guidelines 8 have allowed for the omission of postlumpectomy radiotherapy in these patients since 2004.
Despite these recommendations, both SLNB and adjuvant radiotherapy continue to be used at high rates for women who are eligible for omission. In national samples, more than 80% of women aged 70 years or older with HR-positive breast cancer underwent SLNB, and more than 65% received adjuvant radiotherapy. 1,[9][10][11] Previous assessments of barriers to de-escalation of low-value breast cancer care have focused on clinician beliefs and attitudes. 12,13 Although clinicians cite patient preferences as a factor associated with overtreatment, evidence suggests that many older patients may actually prefer a less-aggressive treatment strategy. 14 To broadly understand this clinical decision-making scenario from the patient's perspective, we performed a qualitative study with women aged 70 years or older to understand patient perspectives on the omission of SLNB and radiotherapy for early-stage breast cancer and to identify strategies to facilitate evidence-based and values-concordant care for patients.

Methods
The study was determined to be exempt from ongoing review via the University of Michigan institutional review board. All participants verbally consented to be interviewed, and a $25 gift card was offered as an incentive. This study is reported in accordance with the Standards for Reporting Qualitative Research (SRQR) reporting guideline. 15 We conducted semistructured phone interviews with 30 participants in the Midwest from October 2019 to January 2020. Eligible participants were women aged 70 years or older, Englishspeaking, and without a previous diagnosis of breast cancer. Individuals who had never received a diagnosis of breast cancer were intentionally chosen for this study to eliminate influences from prior

JAMA Network Open | Oncology
Patient Perspectives on Treatment for Older Women With Hormone Receptor-Positive Breast Cancer experiences. One reason for this choice is that many women who are eligible are not offered omission of these therapies by their clinicians and, thus, are unable to comment on factors influencing their decision-making. Volunteers were recruited via the UMHealthResearch.org website. Purposive sampling was used to increase diversity with respect to age, education, and race/ethnicity. The semistructured interview guide (see the eAppendix in the Supplement) was developed in consultation with subject and methodological experts on the basis of factors hypothesized to be important in breast cancer treatment decision-making. It was piloted with 2 women and further refined for clarity. In response to topics raised in early interviews, questions were added to iteratively explore those topics in subsequent interviews.
Briefly, participants were asked to imagine a scenario in which their doctor gave them a diagnosis of early-stage, HR-positive breast cancer and recommended surgery to remove the cancer.
They were asked about surgery preferences, SLNB preferences in light of age-based guidelines, and chemotherapy preferences if they were found to have a positive lymph node. In a second scenario, participants were asked to imagine undergoing a lumpectomy for the same diagnosis. They were asked about their radiotherapy preferences in relation to the age-based guideline and hormone therapy preferences. Finally, a brief demographic survey was administered.
One surgeon-scientist trained in qualitative methods (T.W.) conducted all interviews. Interviews were audio-recorded, transcribed verbatim, and deidentified. We followed the inductive and iterative approach of interpretive description, [16][17][18] a qualitative method in the constructivist research paradigm that interprets participants' subjective experiences to improve understanding of clinical problems. After 10 interviews, transcripts were examined, and responses were summarized by topic to a framework matrix 19 to facilitate data immersion and preliminary analysis. On the basis of this analysis, the overall sample size of 30 was estimated to provide more than sufficient information power given the exploratory aims of the study, the analysis strategy, and the specificity of the sample.
Information power is a flexible, transparent, and auditable approach to both estimating and assessing sample size in qualitative studies. It relies on key methodological principles rather than any specific technique of analysis, and it aligns well with interpretive description. 20,21 Transcripts were imported to MAXQDA 2020 software (VERBI Software) to support coding and further analysis. The research team developed a codebook that initially contained structural and descriptive codes deductively applied for each question that were later supplemented by inductively derived codes Full analysis proceeded via data abstraction, case comparison, and writing memoranda on clusters of codes to develop the thematic description presented here. 22 In regular meetings, the research team discussed alternative interpretations, researcher biases, latent themes, prevalence, outlying cases, and clinical implications of findings. 16 Cross-tabulation and typology tables in MAXQDA were used to explore potential associations among codes and demographic variables. Data analysis was performed from January to March 2020.

Study Participants
Thirty women were interviewed, ranging in age from 70 to 84 years, with a median (interquartile  Table 1.

Views on Age-Based Guidelines
Approximately one-half of participants agreed that age-based guidelines for cancer treatment were acceptable and acknowledged the value of data to inform treatment decisions. Women commonly cited a biological basis for age-based guidelines, stating that older patients may not respond to treatments or may experience increased adverse effects due to reduced stamina. Additionally, some women acknowledged that the risks of recurrence in older individuals may be outweighed by other causes of death. Many women used a social lens to reflect on age-based guidelines, suggesting they may have pursued more-aggressive treatments when they were younger because of a desire to raise children and grandchildren, but less-aggressive treatments now reflected their personal fulfillment with major life goals.
In contrast, women who stated that age-based guidelines are unacceptable suggested that other factors, such as a person's overall health, family history, and anticipated life span, should primarily inform treatment recommendations. Some women even suggested that age-based guidelines were discriminatory, dismissive of the value of older women, and driven by financial greed.
Other participants accepted a change in treatment recommendations with age but suggested that an age cutoff of 70 years was too young and that guidelines should be further stratified by age (eg, 70 vs 80 vs 90 years old). Regardless of their views on age-based guidelines, all women strongly valued autonomy and defended the right of an individual to pursue any offered treatment regardless of age, even if they would not want a treatment themselves. Importantly, several women volunteered that age-based guidelines would support their autonomy by providing options and an impetus for patient-clinician discussions. For illustrative quotations, see Table 2.

Interpretation of Guidelines to Omit SLNB and Adjuvant Radiotherapy
Many women experienced difficulty interpreting the rationale for guidelines to omit SLNB and adjuvant radiotherapy. Although the scenarios described clinical situations with a good long-term prognosis, several women assumed that SLNB and radiotherapy were not recommended because death due to the disease was inevitable. Women who preferred to omit SLNB and radiotherapy were more likely to reference the anticipated good prognosis. They either stated that the expected outcome (with lumpectomy and endocrine therapy alone) was good enough or that receiving additional interventions would not change the prognosis.

Views and interpretations Representative quotations
Accept age-based guidelines Change in physiology "There is a difference in physiology…not just hormones, the general physiology in people as they age. I don't think the medical researchers are making subjective decisions about things like quality of life and prognosis. They're acting using science-based research to provide the information. Therefore, it makes sense to me. Let me put it this way. It doesn't seem weird to me or conspiratorial that the treatment options and recommendations are different based on age especially with women." (Participant 25) Patient autonomy "I believe older people have often been denied the ability to make their own decisions. Doctors say that's what you do when you find it, so you do it. Without talking with the patients. Some may not have that desire to go through that for the two years they had remaining." (Participant 18) Oppose age-based guidelines Importance of health status "It would depend on your overall health… If you were a healthy person you would make a different decision than if you were unhealthy for any reason. A younger person I think would opt for the treatment no matter what." (Participant 11) Improved longevity "I'm seeing more and more healthy people in their 70s and 80s and 90s, if people eat right and get exercise, they're healthier than they were, and people are living longer." (Participant 29) Genetics "How positive can they be, whoever they are, to say that because I'm 72, the information you learn from it, this procedure won't be helpful… It depends on my physical condition and genetics and everything else I would think." (Participant 14) Need for further age stratification "I think it's one thing if you're in your 70s. It's another thing if you're in your 80s, another if you're in your 90s and not just sort of lump it together as over 70. I mean, being in your 70s right now is like being in your, as far as I'm concerned, over 50s." (Participant 2) Patient autonomy "Well, I think it's kind of all about equality of everybody…Each person is an individual and a person should be treated as that and be able to have the benefits of the younger, as well as the older." (Participant 12) Financial greed "I think that a lot of times, insurance companies are calling the shots on a lot of this stuff, and that that's where the research is coming from. And that they don't want to spend the money on more testing or more procedures." (Participant 24) Discrimination "I don't get this age thing. That's just sort of a discrimination of some sort to me…I find that slightly offensive." (Participant 14) Interpretation of guidelines to omit sentinel lymph node biopsy and adjuvant radiotherapy Mistrust in interpretation of research findings "I would ask why did it not make much difference? Is it because that age group, the 70 plus for instance, has a weaker body and cannot respond very well to chemotherapy or whatever? Or is it because, did they die of some other cause? I mean there are many things that I don't understand about just that information." (Participant 7) "I guess maybe I wouldn't trust the study. I would trust my doctor and I wouldn't trust the study, but I would also do some more research on it, too, on my own." (Participant 12) Some women were wary of applying population-based data underlying the recommendations to their individual circumstances. For example, some participants speculated that they might be healthier than the average participant in the clinical trials. Thus, they expressed a belief that they were exceptions to these guidelines because of presumed longevity based on their good health and, therefore, might benefit from aggressive treatment. Additionally, some participants questioned the reliability of the research informing the guidelines. Illustrative quotations are provided in Table 2.

Factors in Decision-making for SLNB and Radiotherapy
Approximately one-half of participants ( family who underwent radiotherapy. Another factor was the inconvenience of daily travel to a radiotherapy center. The participants who would undergo radiotherapy expressed a desire for peace of mind by eradicating any remnants of cancer. Some women stated a preference for radiotherapy to avoid hormone therapy, which they viewed as potentially having significant adverse effects and lasting for several years as opposed to only a few weeks. For illustrative quotations, see Table 3. Other factors important to older women were their general health, quality of life, and particular life or family circumstances. Nearly all women believed that their overall health status should be of equal or greater weight than chronological age in influencing treatment decisions. Most participants agreed that a worse health status would sway them toward less-aggressive treatments. Some would omit adjuvant therapies because they would not want to risk compromising their current functional status that enables them to serve as a caregiver to their spouse or to not be burdensome to their children. See Table 4 for illustrative quotations. A key emergent theme is the importance of ensuring that older women are correctly interpreting their favorable breast cancer-specific prognosis. Participants frequently justified low-value treatments because of concerns for disease progression and death if they did not pursue aggressive action. In contrast, participants who accepted the expected good prognosis in the presented scenarios expressed comfort with treatment omission. Because older women may be more accustomed to conversations around end-of-life care in the context of cancer treatment, it could be counterintuitive to learn that recommendations for omission of certain therapies is based on a good, rather than poor, prognosis. This suggests that emphasizing the high likelihood of a positive breast cancer-specific prognosis could reassure older patients that avoiding overtreatment will reduce harms while providing equivalent benefit.

JAMA Network Open | Oncology
This approach relies on communicating statistical information and accurately presenting risks of various outcomes to patients. In our study, many participants claimed they were likely healthier than average and more likely to benefit from aggressive treatments. This feeling of personal exceptionalism and desire to pursue low-value services is likely associated with a tendency for patients and clinicians alike to overestimate the benefits and underestimate the risks of medical treatment. [31][32][33] For example, a study 13 of surgeons and radiation oncologists found that physicians overestimate patients' life expectancies, incorrectly attribute radiotherapy to an improvement in survival rather than an improvement in local recurrence, and overestimate the risk of recurrence with radiotherapy omission in older women. One strategy to improve accurate risk communication is to optimize the format in which clinicians provide estimations. 34 For example, although clinicians traditionally communicate risk either numerically or verbally, studies have found that pictographs significantly improve the ability to comprehend incremental risk reduction. 35 Similarly, positively reframing these guidelines may address women's concerns about age discrimination. Some women asserted that older patients were unfairly receiving less medical care but did not fully understand the motivations underlying these recommendations, in this case a good rather than poor prognosis from breast cancer. Conversely, other women thought age-based guidelines actually supported their autonomy by providing them an option to omit treatment. At times, clinicians may worry that their ethical obligation to reduce costs and harms associated with the provision of low-value services conflicts with the ethical principle of patient autonomy. 36 However, studies in behavioral economics demonstrate that patients often do not make choices that are logical, values-concordant, or in their best interest. 37 Thus, there is a potential role for nudging or beneficent persuasion, which is distinct from manipulation because of the transparency involved. 38 "When I was younger, I had two friends that passed away from breast cancer after going through all the treatments and everything." (Participant 21) "I hesitate to follow that route. My mother, and this was years ago, had breast cancer and she had…I don't know if she had removal of lymph nodes…Something done, but use of her arm was limited and it was often swollen." (Participant 28) found that older patients are interested in engaging in conversations with their clinician about reducing low-value cancer screening, particularly when this decision is framed in a way that emphasizes the patient's own health priorities.
Another key finding of this study is that reducing the use of SLNB and adjuvant radiotherapy may require separate strategies, because participants viewed the potential contributions of these treatments differently. SLNB was perceived as a low-risk prognostic test potentially providing peace of mind or informing future treatments. These perceptions are shared by clinicians; a qualitative study 12 of surgeons found that the belief that SLNB influences downstream treatment decisions is a major barrier to decreasing SLNB rates. However, although studies have shown that nodal positivity is associated with more adjuvant therapy, this additional therapy does not increase breast cancerspecific survival. 40 In fact, simulations that suggest older patients with HR-positive breast cancer may experience negative quality-adjusted life-years if they receive chemotherapy compared with endocrine therapy alone. 40,41 Therefore, research-informed education of both clinicians and patients on the risks, costs, and potential downstream consequences of SLNB is necessary. The Figure shows a conceptual model illustrating potential clinician strategies in response to patient-level challenges to the de-escalation of low-value breast cancer therapies in older women.
In contrast to SLNB, most participants in our study would prefer to omit radiotherapy, viewing it as additive and unnecessary. Women frequently cited fears of radiotherapy and concerns about the burden of attending daily appointments for radiotherapy; this contrasts with the perceived convenience of SLNB, which is performed with resection of their primary tumor. These findings suggest that determinants at the clinician or institution level, rather than the patient level, likely contribute to high rates of postlumpectomy radiotherapy in older women. Efforts to decrease radiotherapy use may benefit from clinician-facing strategies balancing the desire for a modest improvement in locoregional recurrence rates with older women's priorities, such as quality of life.

Limitations
This study has several limitations. Although we purposefully sampled to include a diverse group of women, our recruitment method tended to select for White, highly educated, and high-functioning participants. We recognize that the views of this patient population may differ from those of other demographic groups. However, research 42 on disparities in health care suggest that minority women and those of lower socioeconomic status are at risk of poorer outcomes and undertreatment for breast cancer. Conversely, a systematic review 43 on overtreatment found that in circumstances in which race was a significant factor associated with the use of low-value services, nearly all studies suggested that White patients were more likely to receive unnecessary care. In addition, the incidence of breast cancer, particularly in older women, is highest in White women. 2 Thus, we anticipate the patient population we have sampled may be at greater risk for overtreatment if diagnosed with breast cancer.
Furthermore, participants in this study were presented with hypothetical scenarios. Thus, it is possible that women would make different decisions when facing an actual diagnosis or when provided with a more extensive real-world medical consultation with more time to reflect on their decision. However, we think that the factors found to be associated with women's decision-making, such as their interpretation of guidelines, views on aging, and secondhand experiences, are static constructs related to their personal values and history and are relevant in both real-life and theoretical scenarios.

Conclusions
This qualitative study explored patient perspectives on low-value breast cancer treatment options for older women who receive a diagnosis of early-stage, HR-positive breast cancer. Positive reframing of recommendations to avoid SLNB and radiotherapy may be a strategy to reduce overtreatment while maintaining patient autonomy. Although patient preferences for SLNB could partially explain high rates of SLNB use nationally, persistent use of adjuvant radiotherapy may be associated with factors related to guidelines, health professionals, incentives or resources, and capacity for organizational change.