Abstract
Introduction:
Guidelines allow for the omission of sentinel lymph node biopsy (SLNB) and post-lumpectomy radiotherapy in women ≥70 years old with hormone receptor positive (HR+) breast cancer. Despite this, national data suggest these procedures have not been widely de-implemented. Our objectives were to evaluate trends in SLNB and post-lumpectomy radiotherapy utilization in patients who are eligible for omission and evaluate patient preferences as a target for de-implementation of low-value care.
Methods:
We performed a sequential explanatory mixed-methods study by first analyzing an institutional database of patients ≥70 years old with HR+ breast cancer who received surgical treatment from 2014-2018. Based on the quantitative data, we conducted semi-structured interviews with women identified as high or low utilizers of breast cancer treatments to elicit patient perspectives on de-implementation.
Results:
SLNB and post-lumpectomy radiotherapy were performed in 68% and 43% of patients who met criteria for omission, respectively. There was a significant decrease in SLNB rates from 2014-2018. Forty-nine percent of patients were classified as high utilizers and 26% were classified as low utilizers. Qualitative analysis found that the most important factors influencing decision-making regarding SLNB and post-lumpectomy radiotherapy omission for both high and low utilizers were trust in their provider and desire for peace of mind.
Conclusions:
Despite efforts to de-implement low-value care, older women with HR+ breast cancer remain at risk of overtreatment. Patient perspectives suggest that multi-level de-implementation strategies will need to target provider practice patterns and patient-provider communication to promote high quality decision-making and reduction in breast cancer overtreatment.
Introduction
Approximately 30% of new breast cancer patients are ≥70 years old, and this proportion is projected to grow as the population ages.1,2 The vast majority of these patients present with early stage tumors that express estrogen and/or progesterone receptors (hormone receptor positive, HR+). These cancers have a favorable tumor biology with a low risk of recurrence and greater than 98% 10-year breast cancer-specific survival rates.3 Despite this overall excellent prognosis, older women are often treated with the same aggressive strategies used in younger women or women with less favorable tumor biology. However, older patients are more vulnerable to the toxicities of common cancer treatments and may be at significant risk for overtreatment given their promising breast cancer-specific prognosis.4
Although surgical excision and endocrine therapy are known to improve survival in women ≥70 years old with HR+ cancer, mounting evidence suggests that other previously routine therapies, such as axillary staging and post-lumpectomy radiotherapy, offer marginal to no overall survival benefit in this patient population.3,5,6 Specifically, the Cancer and Leukemia Group B (CALGB) 9343 trial found that neither axillary evaluation nor post-lumpectomy radiotherapy provided additional survival benefit compared to lumpectomy and endocrine therapy alone.3,7 As a result, in 2016 the Society of Surgical Oncology (SSO) recommended against the routine use of axillary staging with sentinel lymph node biopsy (SLNB) in women ≥70 years old with clinically node negative, HR+ breast cancer.8 This trend of de-escalation followed long-standing NCCN guidelines that have supported the omission of post-lumpectomy radiotherapy for T1 tumors in these same women since 2004.9
Current national data on rates of SLNB in women ≥70 years old predate the 2016 SSO recommendation, but suggest that over 80% of older women with early-stage, HR+ cancer receive SLNB.10–12 Similarly, data have shown minimal de-implementation of post-lumpectomy radiotherapy in older women since 2004, with more than 65% of women who are candidates for omission continuing to receive adjuvant radiotherapy.13,14 The reasons for persistent use of these low-value procedures is unclear, but some evidence suggests that potential patient drivers of overtreatment include difficulty understanding the rationale for treatment omission to be a favorable, rather than poor, prognosis and perceived value of SLNB for adjuvant therapy decision-making.15 In order to understand the barriers and facilitators to the de-implementation of low-value breast cancer care, we performed a mixed methods study in which we first evaluated institutional quantitative data on breast cancer treatment utilization in older women to 1) evaluate trends in SLNB and post-lumpectomy radiotherapy use overtime and 2) understand the impact of SLNB omission on adjuvant therapy receipt. We then performed qualitative interviews of patients who were high and low utilizers of breast cancer treatment options to 1) evaluate patient perspectives on the de-escalation of breast cancer treatment and 2) identify potential targets for de-implementation.
Methods
Study Overview
We used a sequential explanatory mixed methods approach, a contemporary research technique in which quantitative and qualitative research methods are mixed in a way that allows for understanding beyond what would be possible with either method alone.16 We first performed a retrospective analysis of a registry of breast cancer patients who received multidisciplinary care at our institution. All patients were evaluated by a surgical oncologist and referred to a medical and/or radiation oncologist based on tumor board discussion and patient preference. All women ≥70 years old with clinically node negative, ER/PR+, HER2− invasive breast cancer who received surgical treatment between 2014 and 2018 were included in this study. These findings were used to identify a cohort of high and low utilizers of breast cancer treatment options. High utilizers were defined as mastectomy patients who received SLNB and lumpectomy patients who received both SLNB and adjuvant radiotherapy, while low utilizers were defined as patients who received neither SLNB nor post-lumpectomy radiotherapy. We then conducted qualitative interviews with a sample of high and low utilizers to help explain decision-making for patients eligible for SLNB and radiotherapy omission in order to identify potential patient-level determinants of de-implementation. Data were mixed in the sampling stage and integrated in a joint display. All research was performed after review board approval and per institutional ethical guidelines. This study is reported in accordance with the Good Reporting of a Mixed Methods Study (GRAMMS) guidelines.17
Quantitative Methodology
The primary outcome was utilization of axillary staging by SLNB and post-lumpectomy radiotherapy among patients eligible for omission. Based on current clinical guidelines, all women in this cohort were eligible for SLNB omission, and all women with T1 tumors who received a lumpectomy were eligible for adjuvant radiotherapy omission. Secondary outcomes were rates of other therapies including complete axillary lymph node dissection (ALND), Oncotype Dx testing, chemotherapy, and endocrine therapy. For all patients, we evaluated demographic and clinical information, including age, race, type of breast surgery (lumpectomy versus mastectomy), tumor size, T-stage, tumor grade, histology, and nodal status. A comorbidity score was determined based on a modified Charlson Comorbidity Index.
Demographic and clinical characteristics of patients who received SLNB were compared to those who did not using Fisher’s exact test, a χ2 test, or the Mann-Whitney U test as appropriate. Similar analyses were performed comparing lumpectomy patients with T1 tumors who received or omitted adjuvant radiotherapy. We performed multivariate logistic regression to examine patient and tumor factors that were independently associated with receipt of SLNB or post-lumpectomy adjuvant radiotherapy. To examine which clinical factors, including axillary evaluation, were associated with the receipt of Oncotype Dx testing, chemotherapy, and endocrine therapy, we performed separate multivariate logistic regression models. All statistical analyses were performed using STATA 16.0 and a two-sided p-value <0.05 was considered statistically significant.
Qualitative Methodology
In order to better understand potential reasons for treatment differences in patients’ own words, we screened, recruited, and interviewed a maximum variation sample of 8 high utilizers and 4 low utilizers (defined above) from among those identified in the quantitative analysis. High utilizers were intentionally oversampled to provide insight into patient-level facilitators of overtreatment. Participants were recruited until the sample size was judged sufficient based on preliminary analyses and study aims. Eligible patients received treatment between 2017 to 2018 to ensure they were diagnosed after the 2016 SSO recommendation for SLNB omission. Patients were excluded if found to be currently hospitalized, with altered mental status, or requiring an English interpreter. All participants verbally consented to be interviewed, and a $20 gift card was offered as an incentive.
The semi-structured interview guide was developed based on factors hypothesized to drive breast cancer treatment decision-making in older women. It was piloted with two female volunteers over the age of 70 and further refined for clarity. First, participants were asked to reflect on their initial visit with a surgeon to discuss their breast cancer diagnosis. They were then asked questions to elicit information about the various treatment options provided by their physicians, including their understanding of their prognosis and whether they considered alternative treatments. Participants were next asked to reflect specifically on the decision to undergo or omit SLNB, reasons for their preferences regarding SLNB, and any recollection of being told about guidelines for omission in women ≥70 years of age. Participants were similarly asked to reflect on the decision to receive or omit radiotherapy. Finally, women were asked more general questions including their thoughts on breast cancer treatment options for older women and whether they would change their treatment decisions if diagnosed today. After the interview, a brief demographic survey was administered.
One qualitative researcher (JM) conducted all interviews. Interviews were audio-recorded, transcribed verbatim, and de-identified. We followed the inductive and iterative approach of interpretive description, a qualitative methodology that interprets participants’ subjective experiences to improve understanding of clinical problems.18,19 Transcripts were imported to MAXQDA 2020 software (VERBI Software) to support coding and further analysis. The research team developed a codebook of structural and descriptive codes deductively applied, later supplemented by inductively-derived codes. Analysis proceeded via data abstraction, case comparison, and memo-writing to develop the thematic description presented here.20 The study was determined exempt from ongoing review via the University of Michigan Institutional Review Board.
Results
Quantitative Results
Characteristics of women ≥70 years old with clinically node negative, HR+/HER2− invasive breast cancer by receipt of SLNB and radiotherapy are shown in Table 1. In total, 263 women meeting all inclusion criteria received surgical treatment at our institution during the study period.
Table 1.
Characteristics of women ≥70 years old with clinically node negative, HR+, HER2− breast cancer in patients eligible for treatment omission by SLNB and post-lumpectomy radiotherapy (RT) receipt.
| SLNB n (%) N=179 | No SLNB n (%) N=84 | p-value1 | RT2 n (%) N=70 | No RT2 n (%) N=88 | p-value3 | |
|---|---|---|---|---|---|---|
| Year of Diagnosis | <0.01 | 0.91 | ||||
| 2014 | 45 (77.6%) | 13 (22.4%) | 17 (42.5) | 23 (57.5) | ||
| 2015 | 44 (74.6%) | 15 (25.4%) | 14 (43.8) | 18 (56.3) | ||
| 2016 | 32 (65.3%) | 17 (34.7%) | 11 (39.3) | 17 (60.7) | ||
| 2017 | 33 (76.7%) | 10 (23.3%) | 14 (51.9) | 13 (48.2) | ||
| 2018 | 25 (46.3%) | 29 (53.7%) | 14 (45.2) | 17 (54.8) | ||
| Age (mean ± SD) | 74.4 ± 3.9 | 79.0 ± 6.1 | <0.01 | 73.5 ± 3.0 | 76.6 ± 5.1 | <0.01 |
| Race | 0.24 | 0.90 | ||||
| White | 162 (68.4%) | 75 (31.7%) | 65 (45.5%) | 78 (54.6%) | ||
| Black | 13 (81.3%) | 3 (18.8%) | 3 (37.5%) | 5 (62.5%) | ||
| Other | 2 (40.0%) | 3 (60.0%) | 2 (50.0%) | 2 (50.0%) | ||
| Comorbidity Score | 0.01 | <0.01 | ||||
| 0 | 108 (76.1%) | 34 (23.9%) | 48 (55.8%) | 38 (44.2%) | ||
| 1 | 42 (58.3%) | 30 (41.7%) | 19 (39.6%) | 29 (60.4%) | ||
| 2+ | 29 (59.2%) | 20 (40.8%) | 3 (12.5%) | 21 (87.5%) | ||
| Breast Surgery | <0.01 | --- | ||||
| Lumpectomy | 135 (63.1%) | 79 (36.9%) | --- | --- | ||
| Mastectomy | 44 (89.8%) | 5 (10.2%) | --- | --- | ||
| Tumor size (mean +SD) | 17.3 ±12.1 | 14.3 ± 9.9 | 0.06 | 11.7 ±5.0 | 10.7 ±4.3 | 0.12 |
| T-Stage | 0.92 | --- | ||||
| T1 | 131 (67.2%) | 64 (32.8%) | --- | --- | ||
| T2 | 43 (70.5%) | 18 (29.5%) | --- | --- | ||
| T3+ | 5 (71.4%) | 2 (28.6%) | --- | --- | ||
| Grade | 0.25 | 0.41 | ||||
| 1 | 57 (62.0%) | 35 (38.0%) | 28 (45.9%) | 33 (54.1%) | ||
| 2 | 97 (72.4%) | 37 (27.6%) | 39 (46.4%) | 45 (53.6%) | ||
| 3 | 25 (69.4%) | 11 (30.6%) | 3 (25.0%) | 9 (75.0%) | ||
| Histology | 0.06 | 0.17 | ||||
| Ductal | 123 (65.1%) | 66 (34.9%) | 51 (42.9%) | 68 (57.1%) | ||
| Lobular | 36 (83.7%) | 7 (16.3%) | 13 (61.9%) | 8 (38.1%) | ||
| Other | 20 (64.5%) | 11 (35.5%) | 6 (33.3%) | 12 (66.7%) | ||
| SLNB | --- | <0.01 | ||||
| Yes | --- | --- | 58 (58.6%) | 41 (41.4%) | ||
| No | --- | --- | 12 (20.3%) | 47 (79.7%) | ||
| Nodal Status | --- | <0.01 | ||||
| Positive | 39 (21.8%) | --- | 13 (92.9%) | 1 (7.1%) | ||
| Negative | 135 (75.4%) | --- | 45 (54.9%) | 37 (45.1%) | ||
| Not identified | 5 (2.8%) | --- | 12 (19.4%) | 50 (80.7%) | ||
| ALND (if SLNB+) | --- | --- | ||||
| Yes | 12 (30.8%) | --- | --- | --- | ||
| No | 27 (69.2%) | --- | --- | --- | ||
| Oncotype Dx | <0.01 | <0.01 | ||||
| Yes | 78 (95.1%) | 4 (4.9%) | 33 (70.2%) | 14 (29.8%) | ||
| No | 101 (56.1%) | 79 (43.9%) | 37 (33.3%) | 74 (66.7%) | ||
| Oncotype Dx Score | 0.82 | 1.00 | ||||
| 0-25 | 74 (94.9%) | 4 (5.1%) | 32 (69.6%) | 14 (30.4%) | ||
| 26+ | 4 (100%) | 0 (0%) | 1 (100.0%) | 0 (0%) | ||
| Chemotherapy | 0.02 | 0.44 | ||||
| Yes | 12 (100%) | 0 (0%) | 1 (100%) | 0 (0%) | ||
| No | 166 (66.9%) | 82 (33.1%) | 69 (44.0%) | 88 (56.1%) | ||
| Radiation | <0.01 | --- | ||||
| Yes | 92 (86.8%) | 14 (13.2%) | --- | --- | ||
| No | 84 (55.3%) | 68 (44.7%) | --- | --- | ||
| Hormone therapy | <0.01 | 0.82 | ||||
| Yes | 159 (75.4%) | 52 (24.6%) | 61 (45.9%) | 72 (54.1%) | ||
| No | 16 (37.2%) | 27 (62.8%) | 9 (40.9%) | 13 (59.1%) |
p-value compares characteristics of patients who received SLNB versus patients who omitted SLNB
Patients with T1 tumors receiving lumpectomy only. Status of radiotherapy administration unknown in N=5 patients.
p-value compares characteristics of lumpectomy patients with T1 tumors who received radiotherapy versus patients who omitted radiotherapy
Trends in sentinel lymph node biopsy and radiotherapy
At our institution, there was moderate de-implementation of SLNB over time with the rate of SLNB decreasing from 78% in 2014 to 46% in 2018. Later year of surgery was an independent predictor of decreased odds of SLNB (Figure 1 A). Of the patients who received SLNB (n=179), 22% (n=39) were positive. Axillary lymph node dissection (ALND) was performed in 31% of node-positive patients. Of the patients with T1 tumors who received a lumpectomy (n=163), 43% received adjuvant radiotherapy. Rates of post-lumpectomy radiotherapy in patients eligible for omission were stable overtime (Figure 1B).
Figure 1.
A. Predicted probability of SLNB from 2014-2018 with multivariate adjustment. B. Predicted probability of post-lumpectomy radiotherapy for patients with T1 tumors from 2014-2018 after multivariate adjustment. Probably of SLNB is significantly lower in 2018 compared to 2014-2017 while probability of post-lumpectomy radiotherapy does not vary by year.
Adjusted odds ratios from multivariable logistic regression analysis predicting the odds of receiving SLNB and post-lumpectomy radiotherapy in patients eligible for omission are shown in Table 2. Older age and increased comorbidity scores were associated with significantly lower odds of receiving SLNB or post-lumpectomy radiotherapy while larger tumor size was associated with significantly increased odds of SLNB (Figure 2). Factors including race, tumor grade, and histology were not significantly associated with SLNB or post-lumpectomy radiotherapy receipt.
Table 2.
Adjusted odds ratios (OR) with 95% confidence interval (CI) from multivariable logistic regression analysis predicting the odds of receiving SLNB and post-lumpectomy radiotherapy for patients eligible for omission
| SLNB OR (95% CI) | Radiotherapy1 OR (95% CI) | |
|---|---|---|
| Year of Diagnosis | ||
| 2014 | 1.33 (0.44-4.01) | 0.71 (0.18-2.82) |
| 2015 | 1.76 (0.57-5.42) | 1.28 (0.28-5.71) |
| 2016 (ref) | 1.00 | 1.00 |
| 2017 | 1.35 (0.42-4.31) | 1.16 (0.28-4.85) |
| 2018 | 0.18 (0.06-0.56) | 1.29 (0.31-5.36) |
| Age | 0.74 (0.68-0.81) | 0.80 (0.70-0.91) |
| Race | ||
| White (ref) | 1.00 | 1.00 |
| Non-white | 1.94 (0.42-8.89) | 1.87 (0.38-9.20) |
| Comorbidity Score | ||
| 0 (ref) | 1.00 | 1.00 |
| 1 | 0.43 (0.19-0.98) | 0.58 (0.21-1.58) |
| 2+ | 0.30 (0.12-0.80) | 0.03 (0.00-0.26) |
| Breast Surgery | ||
| Lumpectomy (ref) | 1.00 | --- |
| Mastectomy | 5.71 (1.69-19.27) | --- |
| Tumor size (mm) | 1.06 (1.02-1.10) | 1.10 (1.00-1.22) |
| Grade | ||
| 1 (ref) | 1.00 | 1.00 |
| 2 | 1.65 (0.76-3.56) | 0.60 (0.24-1.50) |
| 3 | 1.17 (0.34-4.11) | 0.37 (0.05-2.53) |
| Histology | ||
| Ductal (ref) | 1.00 | 1.00 |
| Lobular | 2.62 (0.85-8.07) | 1.13 (0.35-3.65) |
| Other | 1.76 (0.61-5.06) | 0.45 (0.11-1.84) |
| Nodal Status | ||
| Negative (ref) | --- | 1.00 |
| Positive | --- | 93.2 (3.63-2395.0) |
| Not assessed | --- | 0.42 (0.13-1.31) |
Patients with T1 tumors receiving lumpectomy only
Figure 2.
A. Predicted probability of SLNB by patient age after multivariate adjustment. B. Predicted probability of post-lumpectomy radiotherapy for patients with T1 tumors by patient age after multivariate adjustment. Probability of SLNB and post-lumpectomy radiotherapy is significantly lower in older patients.
Trends in adjuvant therapy administration
Adjusted odds ratios from multivariable logistic regression analysis predicting the odds of receiving adjuvant therapy are shown in Table 3. Patients receiving SLNB had significantly greater odds of receiving Oncotype Dx testing. Nearly all patients (95%) who underwent Oncotype Dx testing also received SLNB. Ninety-five percent of patients who received Oncotype Dx testing had scores between 0-25 (low to intermediate recurrence risk); only 4 patients received high-risk scores of ≥26 and all 4 received chemotherapy. Patients who were older or received surgery in 2014 (compared to 2015-2018) had decreased odds of receiving Oncotype Dx testing.
Table 3.
Adjusted odds ratios (OR) with 95% confidence interval (CI) from multivariable logistic regression analysis predicting the odds of receiving Oncotype Dx testing, chemotherapy, and endocrine therapy.
| Oncotype Dx OR (95% CI) | Chemotherapy OR (95% CI) | Endocrine therapy OR (95% CI) | |
|---|---|---|---|
| Year of Diagnosis | |||
| 2014 | 0.19 (0.06-0.59) | 0.53 (0.04-7.40) | 2.06 (0.55-7.67) |
| 2015 | 0.57 (0.19-1.67) | 0.09 (0.01-1.45) | 1.67 (0.47-5.95) |
| 2016 (ref) | 1.00 | 1.00 | 1.00 |
| 2017 | 0.67 (0.22-2.03) | 0.94 (0.06-14.19) | 0.80 (0.22-2.93) |
| 2018 | 0.71 (0.22-2.33) | 0.50 (0.03-7.78) | 2.19 (0.55-8.67) |
| Age | 0.75 (0.66-0.85) | 0.61 (0.43-0.87) | 0.83 (0.76-0.91) |
| Race | |||
| White (ref) | 1.00 | --- | 1.00 |
| Non-white | 0.37 (0.09-1.55) | --- | 0.87 (0.16-4.72) |
| Comorbidity Score | |||
| 0 (ref) | 1.00 | 1.00 | 1.00 |
| 1 | 1.97 (0.86-4.54) | 0.12 (0.01-1.62) | 3.71 (1.16-11.94) |
| 2+ | 0.75 (0.27-2.09) | 0.02 (0.00-1.35) | 1.21 (0.40-3.68) |
| Breast Surgery | |||
| Lumpectomy (ref) | 1.00 | 1.00 | 1.00 |
| Mastectomy | 0.66 (0.28-1.58) | 0.11 (0.01-1.70) | 0.35 (0.12-1.04) |
| Tumor size (mm) | 1.04 (1.00-1.07) | 1.18 (1.06-1.30) | 1.04 (0.99-1.09) |
| Grade | |||
| 1 (ref) | 1.00 | --- | 1.00 |
| 2 | 2.09 (0.96-4.55) | --- | 1.38 (0.56-3.43) |
| 3 | 2.95 (0.89-9.81) | --- | 2.13 (0.41-11.18) |
| Histology | |||
| Ductal (ref) | 1.00 | 1.00 | 1.00 |
| Lobular | 1.47 (0.61-3.57) | 0.47 (0.05-4.29) | 0.90 (0.24-3.40) |
| Other | 0.66 (0.18-2.45) | 0.87 (0.04-19.0) | 0.51 (0.16-1.60) |
| Nodal Status | |||
| Negative (ref) | 1.00 | 1.00 | 1.00 |
| Positive | 0.60 (0.23-1.55) | 12.63 (1.68-94.7) | 5.81 (0.60-55.84) |
| Not assessed | 0.06 (0.01-0.22) | 0.34 (0.11-1.00) | |
| Oncotype Dx | |||
| Yes (ref) | --- | 1.00 | --- |
| No | --- | 0.31 (0.04-2.27) | --- |
Four percent of patients who underwent SLNB received chemotherapy based on SLN positivity. Although a positive SLNB result was associated with significantly increased odds of chemotherapy receipt, only 21% of SLN-positive patients received chemotherapy; 41% either declined or were not candidates for chemotherapy and 39% omitted chemotherapy after Oncotype Dx testing indicated a low to intermediate recurrence risk. On multivariate analysis, older age and smaller tumor size were independently associated with decreased chemotherapy receipt. The majority of patients (81%) received endocrine therapy and these rates were stable over time. Patients who were older had decreased odds of receiving endocrine therapy while patients with higher comorbidity scores had significantly increased odds of receiving endocrine therapy.
Qualitative Results
Overall, 49% of patients were classified as high utilizers of breast cancer treatment options (mastectomy patients who received SLNB and lumpectomy patients who received both SLNB and adjuvant radiotherapy), 26% of patients were classified as low utilizers (omitted both SLNB and adjuvant radiotherapy), and 24% of patients were classified as mixed utilizers (received either SLNB or adjuvant radiotherapy). From these two cohorts, we interviewed 12 women in total. The 8 high utilizers all received SLNB, adjuvant radiotherapy, and endocrine therapy, and all but one received Oncotype Dx testing. The 4 low utilizers did not receive SLNB, adjuvant radiotherapy, or Oncotype Dx testing; 2 low utilizers received endocrine therapy. Because guidelines are more restrictive for post-lumpectomy radiotherapy omission, 2 of the high utilizers with pathologic stage T2 tumors were excluded from qualitative analyses on decision-making regarding radiotherapy. A summary of participant characteristics and treatments received is presented in Table 4. Demographics of the two groups were similar with both tending to be highly educated and from metropolitan counties.
Table 4.
Participant characteristics of high and low utilizers of breast cancer treatment options
| High Utilizers N=8 | Low Utilizers N=4 | |
|---|---|---|
| Median age at diagnosis (range) | 71 (70 to 74) | 73.5 (72 to 74) |
| Median age at interview (range) | 72.5 (72 to 77) | 75.5 (73 to 76) |
| Self-reported race/ethnicity | ||
| White | 88% | 75% |
| African-American or Black | 13% | 0% |
| Declined | 0% | 25% |
| Education | ||
| High School | 38% | 25% |
| Two- or four-year degree | 13% | 25% |
| Master’s degree | 50% | 50% |
| Metropolitan county | 88% | 75% |
| Treatments received | ||
| Lumpectomy | 100% | 100% |
| SLNB | 100% | 0% |
| Oncotype | 88% | 0% |
| Radiotherapy | 100% | 0% |
| Endocrine therapy | 100% | 50% |
| Chemotherapy | 13% | 0% |
A key finding of this study is that we identified similar themes regarding patient treatment preferences and decision-making regardless of whether a patient was a high versus low utilizer of breast cancer treatment options. Mixed methods results are presented in a joint display in Table 5.
Table 5.
Representative quotes on factors important in decision-making for patients who are high versus low utilizers of breast cancer treatment options
| Participant recall of breast cancer treatment options | |
| High Utilizers |
On SLNB: I just went with recommendations and no, I don’t remember if that was an option. (Participant 03) On Radiotherapy: They gave me percentage choices, in terms of understanding. They spent a lot of time and my daughters and I discussed it a lot about whether I should have the radiation treatment. (Participant 12) |
| Low Utilizers |
On SLNB: I don’t think I was ever given the choice. (Participant 07) On Radiotherapy: He told me what my options were, and I was very slow giving him an answer, but I finally did. He wasn’t so happy with me when I gave him the final, “No, I would like to try it without.” (Participant 09) |
| Misunderstandings about breast cancer treatment options | |
| High Utilizers | Well, my recollection, it was just part of the process. I didn’t regard it as something additional to the lumpectomy. I thought it was all part of the same surgery. (Participant 08) |
| Low Utilizers | Because at that point there was no evidence I had been given that there was any cancer in those sentinel lymph nodes. And I understand that lymphatic system carries the cancer cells, I understand that, but there was no evidence, no tests had been done to show that there was cancer in those lymph nodes. So why take them out until we know we need to? (Participant 07) |
| Influence of patient-provider trust in decision-making | |
| High Utilizers |
They pretty much said, “we would like to do the radiation,” and so I did. I just trusted them… I didn’t fight it back and forth. I feel that I just trusted he knows better than I. (Participant 02) I guess I would go with what the doctor … I have my faith in the doctors. I would go with what they recommended. I wouldn’t feel that I knew more than they did. (Participant 11) |
| Low Utilizers |
He didn’t seem to think it was necessary, I guess. He didn’t talk to me as far as I know. He might’ve mentioned it as something that could be done, but he didn’t evidently think it was necessary to do it with me. (Participant 04) I might’ve done something different if I was in my forties, because if it helped me to have a longer life span, I might’ve done it. But the doctor saying he thought he got it all and that he would recommend me taking the pill. I thought ‘okay.‘ I went with his recommendation. (Participant 05) |
| Influence of age in decision-making | |
| High Utilizers |
Age is only a number. It’s how you feel and we have positive attitude and we live our life… Just to put an age on anything to me is not always the best way to go. (Participant 03) I know that life expectancy is different, but I would like people to treat them, us older people, as if our lives matter just as much as younger people. (Participant 08) |
| Low Utilizers |
I believe if they’re in good health physically and they can handle the side effects, and their quality of life is wonderful then I feel everyone regardless of age, should be treated the same. I don’t believe in death panels, which essentially to me is what that would almost be. (Participant 07) I think that you’re more vulnerable if you’re older, and I think that things sometimes happen faster, and you may not be able to catch it before it gets out of control, because the body is weaker. (Participant 09) |
| Other factors important in decision-making about breast cancer treatment | |
| High Utilizers |
It was not just peace of mind, but treatment factors. I wanted to know if it had spread or not, and the risk of it spreading, I wanted to know. (Participant 12) I think at that time just to make sure that I thought the radiation was the lesser of two evils, just to make sure to kill the cancer if there’s any cancer there I felt. (Participant 01) |
| Low Utilizers | They’d have to show me that there was really some reason for it: That it’s spreading or that it’s come back or we didn’t get it all or… I ain’t going to do [radiotherapy] just for preventative, forget about that. Probably doing more damage to the tissues. (Participant 04) |
Participant recall of breast cancer treatment options
The majority of participants, both high and low utilizers, did not recall discussing the option to receive or omit SLNB with their surgeon. Two participants did recall potentially being informed of age-based guidelines for SLNB omission. However, neither patient strongly considered omission. Participants in both the high and low utilizer groups reported significant misunderstandings about the role of SLNB in breast cancer treatment. For example, some women thought that SLNB was just another part of their breast cancer surgery and did not realize that opting for SLNB was a separate decision than the one to undergo surgery for their primary cancer. Additionally, women had misconceptions about why a SLNB is sometimes performed; some participants suggested that a SLNB was a therapeutic option to remove cancer cells or that bloodwork would be equivalent for detecting locoregional spread. Unlike SLNB, the majority of patients recalled weighing the option to receive or omit radiation, although most were heavily influenced by their provider’s opinion. Only two participants, both of whom ultimately elected to receive radiotherapy, recalled being informed of guidelines allowing for adjuvant radiotherapy omission in women ≥70 years old.
Influence of patient-provider trust in decision-making
The most significant factor influencing decisions regarding SLNB and radiotherapy was patients’ trust in their providers. All but one participant cited their doctor’s recommendation when asked why they elected to undergo or omit SLNB and/or radiotherapy, and most participants reported that they would make the same decisions again regardless of their age or being informed of current age-based recommendations because they trusted their providers. The importance of patient-provider trust was further illustrated by two women (both low utilizers) who sought second opinions because of a poor experience with physicians at other institutions. While asserting that they trusted their doctors, several participants reported the importance of their own research, Tumor Board recommendations, and the experience of their peers and family members as validating their physicians’ recommendations.
Factors influencing decision-making in older women
Another factor that encouraged women to pursue more treatment was the desire for peace of mind. For many women, knowing whether cancer had spread and informing any need for additional adjuvant therapy supported the importance of SLNB. Similarly, some women felt that radiotherapy would eradicate any remaining cancer and reduce the risk of disease spread. Overall, women did not appear to weigh the risks versus benefits of overtreatment or the increased risks of undergoing cancer treatment for older patients. For most participants, age was not a factor in deciding to undergo or omit SLNB and/or radiotherapy, asserting that age is just a number and functional status matters more. Some women therefore felt that age-based guidelines may be discriminatory. Despite these sentiments, several high utilizers allowed that they would be less inclined to pursue aggressive treatment at an older age. Conversely, 3 out of 4 low utilizers said they would have elected SLNB if they were much younger. Some considerations participants raised that are unique to the context of cancer treatment decision-making and older age included stamina, anticipated lifespan, quality of life, and time with family.
Discussion
Women ≥70 years old diagnosed with clinically node negative, HR+ breast cancers have an excellent cancer-specific prognosis.2,3 As a result, attention has turned to reducing overtreatment in these women. Our results found lower rates of SLNB and post-lumpectomy radiotherapy at our institution than reported in the literature, and the decrease in SLNB rates over time suggest practice-change in response to the 2016 SSO recommendation to avoid SLNB in these women. However, even with these improvements, approximately half of the women who are eligible for omission continue to receive SLNB and post-lumpectomy radiotherapy. Although the reasons for persistent utilization of low-value care are complex, our qualitative data of high and low utilizers of cancer treatment options suggest that multi-level interventions designed to target both patients and providers will be necessary in order to de-implement SLNB and post-lumpectomy radiotherapy in women ≥70 years with HR+ breast cancer.
The most consistently cited factor that influenced decision-making for all participants (including both high and low utilizers) was trust in their provider. In fact, most patients did not recall any information about recommendations allowing for SLNB and radiotherapy omission or remember weighing different treatment options. Rather, participants who were both high and low utilizers recalled being heavily influenced towards one decision or another. As a result, providers should focus on ensuring that patients are making a high-quality decision in which they understand their options, are meaningfully involved in the decision-making process, and receive treatment that reflects their goals.21,22 Potential strategies to support improved communication include patient-facing educational materials, tools such as pictographs for accurate risk communication, or decision aids.23–25
Another key patient-level driver of SLNB and post-lumpectomy radiotherapy utilization is perceived benefit such as eradicating cancer cells or informing adjuvant therapy decisions. Although it is tempting to assume that more treatment leads to improved survival, both prospective and retrospective evidence suggest this is not the case for women ≥70 years old with clinically node negative, HR+ breast cancer who already have greater than 98% cancer-specific 10-year survival rates.3,6 Simulations suggest that older women likely receive marginal, if any, benefit from chemotherapy in addition to endocrine therapy, and women >75 years old may in fact experience net losses in quality-adjusted life years (QALYs) if treated with chemotherapy.26 The concern that chemotherapy may result in more harm than benefit for older patients is likely reflected by the small number of patients in our study that actually received chemotherapy for SLN positivity; only 4% of patients who underwent SLNB received chemotherapy for a positive result. In fact, a high Oncotype Dx score better correlated with chemotherapy administration than SLNB positivity, as all 4 patients with a score 26+ received chemotherapy. Thus, the incorporation of Oncotype Dx into adjuvant decision-making further underscores the safety of omitting axillary staging. While routine Oncotype Dx testing in this patient population may not result in cost savings compared to SLNB, Oncotype Dx is non-invasive, can be performed on the core needle biopsy, and could avoid clinical harms related to SLNB.
Similar to axillary staging, post-lumpectomy radiotherapy has not been shown to improve survival in these women, and persistent use likely reflects a desire for modest improvements in locoregional recurrence rates or provider concerns that patients may not be compliant with endocrine therapy.3,5 Importantly, current guidelines do not endorse the substitution of endocrine therapy with radiotherapy alone, although retrospective data suggests that this may be safe and warrants evaluation with a randomized controlled trial.27 As a result, patients and providers need to participate in shared decision making in which the risks versus benefits of all therapy options are appropriately weighed. Strategies should focus on educating patients about their excellent cancer-specific prognosis and explaining that aggressive treatment of older women with multiple adjuvant therapies could potentially result in harm without significant improvement in cancer outcomes compared to tumor excision and endocrine therapy alone.
While improving patient decision-making is a key component of efforts to decrease overtreatment of breast cancer in women ≥70 years old with early stage, HR+ tumors, the reliance of patients on their provider’s expertise suggests that specific strategies are required to target what providers recommend to patients and why. Our study shows that there is a patient selection bias towards more adjuvant therapies in women who are younger among the cohort, have fewer comorbidities, and have larger tumors. This finding is consistent with other studies that both patients and providers are more willing to pursue aggressive treatment strategies for high functioning patients even when there is a lack of evidence to support a benefit.28,29 Such considerations are complex, and efforts to target persistent utilization of low-value care at the provider level will require de-implementation strategies that directly target specific drivers. For instance, a lack of guideline dissemination would suggest that provider education could be beneficial, a lack of self-efficacy on leading discussions with patients about treatment omission would suggest that written communication prompts could be successful, and financial incentives to offer more services may be addressed with alternative payment models. Ultimately, a combination of strategies and engagement by multidisciplinary providers will be necessary for SLNB and post-lumpectomy radiotherapy de-implementation in older women.
This study has several limitations. The quantitative data is limited by its retrospective and single institution nature; however, no national data are currently available on practice patterns after the 2016 SSO recommendations to omit SLNB. Additionally, quantitative data from a single institution were important for the identification of high and low utilizers of breast cancer treatment options to inform the qualitative aims of this study. The qualitative data are potentially biased by sampling, as we noted easier recruitment of high utilizers compared to low utilizers. This contrast may suggest a general tendency among low utilizers to avoid engagement with medical activities, although lack of further participation makes this interpretation speculative. We feel that this concern is mitigated because this study aims to identify facilitators of overtreatment, and thus high utilizers of breast cancer treatment options were intentionally oversampled. The qualitative data are limited by recall bias, as women were asked to reflect on discussions from up to 3 years prior. However, we feel that women’s long-lasting interpretations of treatment recommendations are reflective of key factors important to their decision-making. Finally, although the mixed methods design of this study offers several strengths including unique insights into potential strategies for the de-implementation of low-value care for older women with HR+ breast cancer, our institution is a large, academic medical center and practice patterns may not reflect those of other institutions. As a result, any de-implementation strategies will need to be designed and assessed in multiple practice settings for optimal impact.
In conclusion, despite recommendations allowing for omission of SLNB and post-lumpectomy radiotherapy in women ≥70 years old with clinically node negative, HR+ breast cancer, our results suggest that approximately half of patients eligible for omission continue to receive these therapies. Patients suggest that trust in their providers is one of the most important factors in their decision-making when deciding to pursue optional therapies. Therefore, efforts at de-implementation of low-value care should focus on patient-provider communication to improve high quality decision-making and target provider practice patterns.
Synopsis.
Older women with early stage, HR+ breast cancer are at risk of overtreatment. Multi-level de-implementation strategies will need to target provider practice patterns and patient-provider communication to promote high quality decision-making.
Acknowledgments
Disclosures: Dr. Dossett is supported by a grant from the Agency for Healthcare Research and Quality (AHRQ; 5 K08 HS026030-02). Dr. Wang is supported by a grant from the National Cancer Institute (NCI; T32 CA009672).
Footnotes
Publisher's Disclaimer: This Author Accepted Manuscript is a PDF file of an unedited peer-reviewed manuscript that has been accepted for publication but has not been copyedited or corrected. The official version of record that is published in the journal is kept up to date and so may therefore differ from this version.
References
- 1.Rosenberg PS, Barker KA, Anderson WF. Estrogen Receptor Status and the Future Burden of Invasive and In Situ Breast Cancers in the United States. J Natl Cancer Inst. 2015;107(9). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.American Cancer Society. Breast Cancer Facts & Figures 2019-2020. In. Atlanta: American Cancer Society, Inc; 2019. [Google Scholar]
- 3.Hughes KS, Schnaper LA, Bellon JR, et al. Lumpectomy plus tamoxifen with or without irradiation in women age 70 years or older with early breast cancer: long-term follow-up of CALGB 9343. J Clin Oncol. 2013;31(19):2382–2387. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Muss HB, Berry DA, Cirrincione C, et al. Toxicity of older and younger patients treated with adjuvant chemotherapy for node-positive breast cancer: the Cancer and Leukemia Group B Experience. J Clin Oncol. 2007;25(24):3699–3704. [DOI] [PubMed] [Google Scholar]
- 5.Kunkler IH, Williams LJ, Jack WJ, Cameron DA, Dixon JM, investigators PI. Breast-conserving surgery with or without irradiation in women aged 65 years or older with early breast cancer (PRIME II): a randomised controlled trial. Lancet Oncol. 2015;16(3):266–273. [DOI] [PubMed] [Google Scholar]
- 6.Chagpar AB, Hatzis C, Pusztai L, et al. Association of LN Evaluation with Survival in Women Aged 70 Years or Older With Clinically Node-Negative Hormone Receptor Positive Breast Cancer. Ann Surg Oncol. 2017;24(10):3073–3081. [DOI] [PubMed] [Google Scholar]
- 7.Hughes KS, Schnaper LA, Berry D, et al. Lumpectomy plus tamoxifen with or without irradiation in women 70 years of age or older with early breast cancer. N Engl J Med. 2004;351(10):971–977. [DOI] [PubMed] [Google Scholar]
- 8.Boughey JC, Haffty BG, Habermann EB, Hoskin TL, Goetz MP. Has the Time Come to Stop Surgical Staging of the Axilla for All Women Age 70 Years or Older with Hormone Receptor-Positive Breast Cancer? Ann Surg Oncol. 2017;24(3):614–617. [DOI] [PubMed] [Google Scholar]
- 9.NCCN Clinical Practice Guidelines in Oncology: Breast Cancer. https://www.nccn.org/professionals/physician_gls/pdf/breast.pdf. Accessed February 25, 2020.
- 10.Wang T, Baskin A, Dossett LA. De-Implementation of the Choosing Wisely® Recommendations for Low-Value Breast Cancer Surgery: A Systematic Review2020 In press, JAMA Surgery. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Chagpar AB, Horowitz N, Sanft T, et al. Does lymph node status influence adjuvant therapy decision-making in women 70 years of age or older with clinically node negative hormone receptor positive breast cancer? Am J Surg. 2017;214(6):1082–1088. [DOI] [PubMed] [Google Scholar]
- 12.Downs-Canner SM, Gaber CE, Louie RJ, et al. Nodal positivity decreases with age in women with early-stage, hormone receptor-positive breast cancer. Cancer. 2020;126(6):1193–1201. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Shumway DA, Griffith KA, Sabel MS, et al. Surgeon and Radiation Oncologist Views on Omission of Adjuvant Radiotherapy for Older Women with Early-Stage Breast Cancer. Ann Surg Oncol. 2017;24(12):3518–3526. [DOI] [PubMed] [Google Scholar]
- 14.Taylor LJ, Steiman JS, Anderson B, et al. Does persistent use of radiation in women > 70 years of age with early-stage breast cancer reflect tailored patient-centered care? Breast Cancer Res Treat. 2020;180(3):801–807. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Wang T, Mott N, Miller J, et al. Patient Perspectives on De-Escalation of Breast Cancer Treatment Options in Older Women with Hormone-Receptor Positive Breast Cancer: A Qualitative Study. 2020. Under Review. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Dossett LA, Kaji AH, Dimick JB. Practical Guide to Mixed Methods. JAMA Surg. 2020. [DOI] [PubMed] [Google Scholar]
- 17.O’Cathain A, Murphy E, Nicholl J. The quality of mixed methods studies in health services research. J Health Serv Res Policy. 2008;13(2):92–98. [DOI] [PubMed] [Google Scholar]
- 18.Thorne S, Kirkham SR, MacDonald-Emes J. Interpretive description: a noncategorical qualitative alternative for developing nursing knowledge. Res Nurs Health. 1997;20(2):169–177. [DOI] [PubMed] [Google Scholar]
- 19.Thorne S Interpretive Description: Qualitative Research for Applied Practice. Second ed: Routledge; 2016. [Google Scholar]
- 20.Sandelowski M, Barroso J. Classifying the findings in qualitative studies. Qual Health Res. 2003;13(7):905–923. [DOI] [PubMed] [Google Scholar]
- 21.Sepucha KR, Borkhoff CM, Lally J, et al. Establishing the effectiveness of patient decision aids: key constructs and measurement instruments. BMC Med Inform Decis Mak. 2013;13 Suppl 2:S12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Sepucha KR, Scholl I. Measuring shared decision making: a review of constructs, measures, and opportunities for cardiovascular care. Circ Cardiovasc Qual Outcomes. 2014;7(4):620–626. [DOI] [PubMed] [Google Scholar]
- 23.Schonberg MA, Freedman RA, Recht AR, et al. Developing a patient decision aid for women aged 70 and older with early stage, estrogen receptor positive, HER2 negative, breast cancer. J Geriatr Oncol. 2019;10(6):980–986. [DOI] [PubMed] [Google Scholar]
- 24.Zikmund-Fisher BJ, Fagerlin A, Ubel PA. Improving understanding of adjuvant therapy options by using simpler risk graphics. Cancer. 2008;113(12):3382–3390. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Fagerlin A, Zikmund-Fisher BJ, Ubel PA. Helping patients decide: ten steps to better risk communication. J Natl Cancer Inst. 2011;103(19):1436–1443. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Chandler Y, Jayasekera J, Schechter C, Isaacs C, Cadham C, Mandelblatt J. Simulation of Chemotherapy Effects in Older Breast Cancer Patients With High Recurrence Scores. J Natl Cancer Inst. 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Buszek SM, Lin HY, Bedrosian I, et al. Lumpectomy Plus Hormone or Radiation Therapy Alone for Women Aged 70 Years or Older With Hormone Receptor-Positive Early Stage Breast Cancer in the Modern Era: An Analysis of the National Cancer Database. Int J Radiat Oncol Biol Phys. 2019;105(4):795–802. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Jacobs BL, Lopa SH, Yabes JG, Nelson JB, Barnato AE, Degenholtz HB. Association of functional status and treatment choice among older men with prostate cancer in the Medicare Advantage population. Cancer. 2016;122(20):3199–3206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Smith ME, Vitous CA, Hughes TM, Shubeck SP, Jagsi R, Dossett LA. Barriers and Facilitators to De-Implementation of the Choosing Wisely((R)) Guidelines for Low-Value Breast Cancer. Ann Oncol. 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]