Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Apr 1.
Published in final edited form as: J Surg Res. 2024 Feb 14;296:654–664. doi: 10.1016/j.jss.2024.01.037

Anatomy vs Biology: What Guides Chemotherapy Decisions in Older Patients with Breast Cancer?

Sydney M Record 1, Samantha M Thomas 2,3, William M Tian 1, Astrid Botty van den Bruele 1,2, Akiko Chiba 1,2, Gayle DiLalla 1,2, Maggie L DiNome 1,2, Gretchen Kimmick 2,4, Laura H Rosenberger 1,2, Hannah E Woriax 1,2, E Shelley Hwang 1,2, Jennifer K Plichta 1,2,5
PMCID: PMC10947834  NIHMSID: NIHMS1961912  PMID: 38359680

Abstract

Background and Objectives:

With the increasing utilization of genomic assays, such as the Oncotype DX Recurrence Score (RS), the relevance of anatomic staging has been questioned for select older patients with breast cancer. We sought to evaluate differences in chemotherapy receipt and/or survival among older patients based on RS and sentinel lymph node biopsy (SLNB) receipt/result.

Methods:

Patients aged ≥65 diagnosed with pT1–2/cN0/M0 hormone-receptor-positive (HR+)/HER2− breast cancer (2010–2019) were selected from the National Cancer Database. Logistic regression was used to identify factors associated with chemotherapy receipt. Cox proportional hazards models were used to estimate the association of RS/SLNB group with OS. A cost-benefit study was also performed.

Results:

Of the 75,428 patients included, the majority had an intermediate RS (58.2% vs 27.9% low, 13.8% high) and were SLNB− (85.1% vs 11.6% SLNB+, 3.3% none). Chemotherapy was recommended for 13,442 patients (17.8%). After adjustment, chemotherapy receipt was more likely with higher RS and SLNB+. After adjustment, SLNB receipt/result was only associated with OS among those with an intermediate RS. However, returning to the OR for SLNB is not cost-effective.

Conclusions:

SLNB receipt/result was associated with survival for those with an intermediate RS, but not a low or high RS, suggesting that a SLNB may indeed be unnecessary for select older patients with breast cancer.

Keywords: breast cancer, elderly, survival, genomic testing, oncotype, sentinel lymph node biopsy

INTRODUCTION

The population of the United States is aging,1 and increasing age is one of the strongest risk factors for developing cancer, including breast cancer.2 Furthermore, the proportion of breast cancer cases among older patients (age ≥70y) is predicted to rise from 24% to 35% by the next decade.3 However, the population of older patients with breast cancer is heterogenous, and while therapy de-escalation may be appropriate for select patients, it may not be advisable for others. While some studies related to axillary de-escalation, including CALGB 9343, have used an age restriction of ≥70,4 others have used 65y for investigations in older breast cancer patients, such as the PRIME-II trial5, and as such was used to define our study population. Martelli et al. showed in their clinical trial of patients 65y-80y with cT1cN0 breast cancer that there were no significant survival differences between patients who underwent axillary dissection and those who did not.6 The IBCSG 10–93 trial also showed similar survival between patients who did and did not undergo axillary dissection for patients as young as 60y with cN0 hormone receptor (HR)-positive breast cancer who were prescribed adjuvant endocrine therapy.7 However, these studies were conducted prior to routine use of the sentinel lymph node biopsy (SLNB) for axillary staging, which is often still performed and used to guide adjuvant therapy decision-making in older patients with breast cancer.8

More recently, studies have demonstrated that genomic assays may be used to identify specific subgroups of patients less likely to benefit from chemotherapy.9 For example, the 21-gene recurrence score (RS) assay (Oncotype DX, Exact Sciences, Madison, WI) is applicable to patients with HR-positive, HER2-negative, early-stage breast cancer, and it has been prospectively validated for use in prognostication and in the prediction of degree of benefit from adjuvant chemotherapy.9–12 These findings highlight the importance of tumor biology, which has become increasingly recognized as a key component in predicting outcomes rather than anatomic staging.9,10,13–15 With this shift in perspective, the ongoing relevance of axillary lymph node assessment for anatomic staging has been brought into question. Therefore, we sought to evaluate differences in chemotherapy recommendation/receipt and/or survival among older patients (age ≥65y) with breast cancer based on RS and SLNB receipt/result. A cost-benefit analysis was also performed to evaluate the potential financial impact of obtaining a RS and performing a SLNB.

METHODS

Study Population

Using the National Cancer Database (NCDB, 2020 PUF), patients aged ≥65y diagnosed in 2010–2019 with pT1–2/cN0/M0, HR-positive/HER2-negative breast cancer were identified. Patients with cN1–3 or cT3–4 disease were excluded, as were patients with missing data for cT, cN, pT, pN, tumor grade, estrogen receptor (ER) status, progesterone receptor (PR) status, HER2 status, chemotherapy, radiation therapy, endocrine therapy, or survival (including all patients diagnosed in the last reporting year, 2019, per NCDB administrative censoring requirements). Patients who did not have a numeric RS were also excluded, as were those with an unknown number or >5 lymph nodes removed/examined, and those who did not undergo surgery. Patients who received any type of neoadjuvant treatment, including chemotherapy, radiation therapy, endocrine therapy, or immunotherapy, were excluded.

RS were categorized as low (<11), intermediate (11–25), or high (>25). Since the SLNB variable was introduced into the NCDB in 2012 and our study includes patients prior to this date, patients were also categorized into SLNB groups as follows:

No lymph nodes removed → SLNB = None

1–5 lymph nodes removed, but 0 lymph nodes positive → SLNB-negative

1–5 lymph nodes removed and 1–5 lymph nodes positive → SLNB-positive Patients who could not be categorized into a SLNB group were excluded. This study was deemed exempt by our Institutional Review Board.

Statistical Analysis

For our primary analyses, the dependent variables were recommendation and receipt of chemotherapy, as well as overall survival (OS). Independent variables included SLNB receipt/result and RS. Patient characteristics were summarized with N (%) for categorical variables and median (interquartile range, IQR) for continuous variables. Differences were tested with the chi-square test or t-test for categorical or continuous variables, respectively. Data were summarized by recommendation for treatment with chemotherapy and receipt of chemotherapy. Logistic regression was used to identify factors associated with recommendation for treatment with chemotherapy or receipt of chemotherapy. All logistic regression models were built in the generalized estimating equations framework and included an exchangeable covariance structure to account for the correlation of patients treated at the same facility. Odds ratios (ORs) and 95% confidence intervals (CIs) were reported. Adjustment covariates in the logistic models included age, sex, race/ethnicity, year of diagnosis, patient insurance type, Charlson/Deyo comorobidity score, zip code education level, zip code income level, community type, facility type and location, histology, grade, tumor size, RS group, SLNB group, breast surgery type, treatment with radiation, and treatment with endocrine therapy.

OS was defined as the time from diagnosis to death or last follow up. Unadjusted OS was estimated using the Kaplan-Meier method, and groups were compared with the log-rank test. Cox proportional hazards models were used to estimate the association of SLNB group with OS after adjustment for covariates, which included age, sex, race/ethnicity, year of diagnosis, patient insurance type, Charlson/Deyo comorbidity score, zip code education level, zip code income level, community type, facility type and location, histology, grade, tumor size, RS group, SLNB group, surgery type, treatment with chemotherapy, treatment with radiation therapy, and treatment with endocrine therapy. All adjusted survival models included a robust sandwich covariance estimator to account for the correlation of patients treated at the same facility. Hazard ratios (HRs) and 95% CIs were reported. No adjustments were made for multiple comparisons. All statistical analyses were conducted using SAS version 9.4 (SAS Institute, Cary NC).

Cost-Benefit Analysis

Based on our findings, we conducted a cost-benefit analysis comparing two differential algorithmic approaches to SLNB using data from the patients included in our original analyses. We used the observed rates of SLNB receipt in our overall study cohort to model the rate of SLNB receipt for these analyses. We conducted a healthcare system perspective analysis. Patients’ costs, lost productivity, pain, and suffering were not included.

Our two treatment and testing algorithms were based on our finding that patients with an intermediate RS score (11–25) were the only population that saw improved OS with a SLNB. A RS is typically assessed on the surgical specimen, and thus, if we implemented a policy where only patients with an intermediate RS underwent SLNB, this would require a delayed return to the operating room (OR). Our algorithms were structured with the following rules and assumptions:

  1. Upfront SLNB. All patients aged ≥65y who undergo breast surgery undergo SLNB. A RS is assessed from the surgical specimen for all patients.

  2. Delayed SLNB. All patients aged ≥65y who undergo breast surgery do no undergo SLNB during initial surgery. A RS is assessed from the surgical specimen for all patients, and for those whose RS score is in the intermediate range, a delayed SLNB is performed.

Variable values used in the baseline model were obtained from the literature (Supplemental Table 1). For example, Gemignani et al. conducted a study that incorporated operating room, hospital stay, and pathologic examination costs from 100 patients at a single institution to calculate average healthcare system costs of breast conserving surgery (BCS) without SLNB, BCS with SLNB, and BCS with return to OR for SLNB.16 These costs were adjusted to January 2023 dollars using the medical care component of the Consumer Price Index and were used for our baseline model.17

All breast surgery in the cost-benefit analysis was modeled as BCS. This approach was chosen to avoid pulling price data from separate references, introducing unnecessary variability. Since patients undergo initial breast surgery in either algorithm, there would likely be minimal benefit to incorporating both mastectomy and BCS. Additionally, both models incorporate RS testing on surgical samples from all patients, meaning no cost difference between the two algorithms would be introduced by RS testing, so this cost was excluded from the model. The incremental cost-benefit was defined as the net cost per patient spared from SLNB using the Delayed SLNB algorithm compared to the Upfront SLNB algorithm, based on the prevalence of each RS score range in our study cohort.

RESULTS

Patient Characteristics

After applying our defined inclusion and exclusion criteria (Supplemental Figure 1), our cohort included 75,428 patients, and the median follow-up for all patients was 54.0 months (95% CI 53.7–54.3). The majority were aged <75y (37413/75428; 49.6% age 65–69y, 24165/75428; 32% age 70–74y, 10631/75428; 14.1% age 75–79y, 3219/75428; 4.3% age 80y+), had an intermediate RS (21081/75428; 58.2% vs 43907/75428; 27.9% low, 10440/75428; 13.8% high) and had negative nodes on SLNB (64202/75428; 85.1% vs 8735/75428; 11.6% SLNB+, 2491/75428; 3.3% none). Of those who had positive nodes on SLNB (n=8735), most were pN1 (n=8646; 98.3%). Performance of SLNB was similar across RS groups (96.4–96.8%), as was SLNB-positivity rate (low: 11.9%; intermediate: 11.9%; high: 9.6%). Most (66638/75428; 88.3%) patients did not receive chemotherapy (Table 1).

Table 1.

Patient characteristics of those aged ≥65y with pT1–2/cN0/M0, HR+/HER2− breast cancer diagnosed 2010–2019 in the National Cancer Database, stratified by chemotherapy receipt.

All Patients (N=75428) Treatment with Chemotherapy P-Value
No (N=66638) Yes (N=8690)
Age (Years) <0.001
 65–69 37413 (49.6%) 32320 (48.5%) 5049 (58.1%)
 70–74 24165 (32%) 21512 (32.3%) 2622 (30.2%)
 75–79 10631 (14.1%) 9776 (14.7%) 839 (9.7%)
 80–84 2670 (3.5%) 2494 (3.7%) 169 (1.9%)
 85+ 549 (0.7%) 536 (0.8%) 11 (0.1%)
 Median (IQR) 70 (67 – 73) 70 (67 – 73) 69 (66 – 72) <0.001
Sex 0.88
 Female 74714 (99.1%) 66006 (99.1%) 8609 (99.1%)
 Male 714 (0.9%) 632 (0.9%) 81 (0.9%)
Race/Ethnicity <0.001
 Non-Hispanic White 63980 (84.8%) 56672 (85%) 7225 (83.1%)
 Non-Hispanic Black 4816 (6.4%) 4109 (6.2%) 702 (8.1%)
 Non-Hispanic Asian 1728 (2.3%) 1530 (2.3%) 196 (2.3%)
 Hispanic 2391 (3.2%) 2113 (3.2%) 272 (3.1%)
 Other 566 (0.8%) 512 (0.8%) 54 (0.6%)
Insurance <0.001
 Private 11141 (14.8%) 9655 (14.5%) 1472 (16.9%)
 Government 63445 (84.1%) 56239 (84.4%) 7123 (82%)
 None 238 (0.3%) 208 (0.3%) 28 (0.3%)
Charlson/Deyo Comorbidity Score 0.24
 0 59267 (78.6%) 52363 (78.6%) 6819 (78.5%)
 1 12169 (16.1%) 10720 (16.1%) 1438 (16.5%)
 ≥2 3992 (5.3%) 3555 (5.3%) 433 (5%)
ER Status <0.001
 ER+ 75353 (99.9%) 66596 (99.9%) 8657 (99.6%)
 ER− 75 (0.1%) 42 (0.1%) 33 (0.4%)
PR Status <0.001
 PR+ 66976 (88.8%) 60809 (91.3%) 6084 (70%)
 PR− 8452 (11.2%) 5829 (8.7%) 2606 (30%)
Histology <0.001
 Ductal 54884 (72.8%) 47819 (71.8%) 6992 (80.5%)
 Lobular 18071 (24%) 16532 (24.8%) 1514 (17.4%)
 Other 2473 (3.3%) 2287 (3.4%) 184 (2.1%)
Grade <0.001
 1 21448 (28.4%) 20506 (30.8%) 922 (10.6%)
 2 42720 (56.6%) 38665 (58%) 3991 (45.9%)
 3 11260 (14.9%) 7467 (11.2%) 3777 (43.5%)
Clinical T-Stage <0.001
 T0 902 (1.2%) 797 (1.2%) 103 (1.2%)
 T1 61261 (81.2%) 54861 (82.3%) 6317 (72.7%)
 T2 13265 (17.6%) 10980 (16.5%) 2270 (26.1%)
Pathologic T-Stage <0.001
 T1 57152 (75.8%) 51409 (77.1%) 5666 (65.2%)
 T2 18276 (24.2%) 15229 (22.9%) 3024 (34.8%)
Pathologic N-Stage <0.001
 N0 64918 (86.1%) 57661 (86.5%) 7181 (82.6%)
 N1 8646 (11.5%) 7289 (10.9%) 1336 (15.4%)
 N2 23 (0%) 13 (0%) 10 (0.1%)
 N3 2 (0%) 1 (0%) 1 (0%)
 NX 1839 (2.4%) 1674 (2.5%) 162 (1.9%)
RS Group <0.001
 RS <11 21081 (27.9%) 20857 (31.3%) 214 (2.5%)
 RS 11–25 43907 (58.2%) 41448 (62.2%) 2390 (27.5%)
 RS >25 10440 (13.8%) 4333 (6.5%) 6086 (70%)
Sentinel Lymph Node Biopsy Result <0.001
 None 2491 (3.3%) 2263 (3.4%) 223 (2.6%)
 Negative 64202 (85.1%) 57030 (85.6%) 7098 (81.7%)
 Positive 8735 (11.6%) 7345 (11%) 1369 (15.8%)
Breast Surgery Type <0.001
 Lumpectomy 58191 (77.1%) 51616 (77.5%) 6507 (74.9%)
 Mastectomy 17237 (22.9%) 15022 (22.5%) 2183 (25.1%)
Recommended for Treatment with Chemotherapy -
 No 61986 (82.2%) 61986 (93%) 0 (0%)
 Yes 13442 (17.8%) 4652 (7%) 8690 (100%)
Treatment with Endocrine Therapy 0.25
 No 5174 (6.9%) 4533 (6.8%) 620 (7.1%)
 Yes 70254 (93.1%) 62105 (93.2%) 8070 (92.9%)
Open in a new tab

Data presented as N (%) unless otherwise specified. Percentages may not add up 100 due to rounding or missing values. Abbreviations: IQR=interquartile range, ER=estrogen receptor, PR=progesterone receptor, RS=recurrence score

Chemotherapy Recommendation/Receipt

Chemotherapy was recommended for 17.8% of patients (13442/75428), and 64.6% of those patients received chemotherapy (8,690/13442). Most patients who received chemotherapy were younger (5049/8690; 58.1% aged 65–69y), whereas 48.5% (32320/66638) of those who did not receive chemotherapy were aged 65–69y. No significant differences in Charlson-Deyo comorbidity scores were noted between those who did and did not receive chemotherapy (p=0.24). Patients with pT2, grade 3, pN+ disease, and/or high RS received chemotherapy more frequently (all p<0.001).

Rates of chemotherapy receipt increased with increasing RS (Figure 1). After adjustment, chemotherapy receipt was still more likely with a higher RS [low: REF; intermediate: OR 5.3 (95% CI 4.5–6.3); high: OR 118.3 (95% CI 99.5–140.6)] and SLNB positivity [(none: REF; SLNB–: OR 1.2 (95% CI 1.0–1.5); SLNB+: OR 3.0 (95% CI 2.4–3.8); Table 2]. Similarly, rates of chemotherapy recommendation increased with increasing RS (Figure 1). Once again, after adjustment, chemotherapy recommendation was more likely with a higher RS [low: REF; intermediate: OR 4.7 (95% CI 4.2–5.3); high: OR 99.2 (95% CI 87.2–112.9)] and SLNB+ [(none: REF; SLNB–: OR 1.0 (95% CI 0.8–1.1); SLNB+: OR 2.4 (95% CI 2.0–3.0)].

Figure 1.

Figure 1.

Open in a new tab

Rates of chemotherapy (A) recommendation and (B) receipt by RS group and SLNB receipt/result, among patients aged ≥65y with pT1–2/cN0/M0, HR+/HER2− breast cancer diagnosed 2010–2019 in the National Cancer Database.

Abbreviations: RS=Oncotype DX recurrence score; SLNB=sentinel lymph node biopsy

Table 2. Adjusted logistic regression model for chemotherapy receipt for patients aged ≥65y with pT1–2/cN0/M0, HR+/HER2− breast cancer diagnosed 2010–2019 in the National Cancer Database (N=61,539).

In addition to the variables listed, the model was also adjusted for age, sex, race/ethnicity, year of diagnosis, patient insurance type, Charlson/Deyo comorbidity score, zip code education level, zip code income level, community type, facility type, facility location, histology, tumor grade, tumor size, breast surgery type, radiation therapy receipt, and endocrine therapy receipt.

Covariates Odds Ratio (95% CI) P-Value Overall P-Value
RS Group <0.001
 RS < 11 REF
 RS 11–25 5.31 (4.48–6.3) <0.001
 RS >25 118.28 (99.54–140.56) <0.001
SLNB Receipt/Result <0.001
 None REF
 Negative 1.2 (0.98–1.46) 0.08
 Positive 3.02 (2.42–3.75) <0.001
Open in a new tab

Abbreviations: CI=confidence interval, RS=recurrence score, SLNB=sentinel lymph node biopsy.

Survival Analysis

Survival was greatest in those with a low RS score who were treated with chemotherapy (Figure 2). Of note, this observation is based on the data available, and does not imply treatment concordance with standard guidelines. In our unadjusted analysis, we saw significantly greater survival for patients with a negative SLNB in all three RS groups (Figure 3). After adjustment, SLNB receipt/result was only associated with OS among those with an intermediate RS [none: REF; SLNB–: HR 0.7 (95% CI 0.6–0.8); SLNB+: HR 0.9 (95% CI 0.8–1.2); Table 3]. Of patients who underwent SLNB, node positivity was associated with compromised survival overall, and this result was similar across all RS groups [(RS < 11: HR 1.3 (95% CI 1.1–1.6); RS 11–25: HR 1.4 (95% CI 1.3–1.6); RS > 25: HR 1.4 (95% 1.1–1.7)].

Figure 2.

Figure 2.

Open in a new tab

Overall survival stratified by RS group and treatment with/without chemotherapy, among patients aged ≥65y with pT1–2/cN0/M0, HR+/HER2− breast cancer diagnosed 2010–2019 in the National Cancer Database.

Abbreviations: RS=Oncotype DX recurrence score

Figure 3.

Figure 3.

Figure 3.

Open in a new tab

Overall survival stratified by SLNB receipt/result for patients with a (A) low RS, (B) intermediate RS, and (C) high RS; based on patients aged ≥65y with pT1–2/cN0/M0, HR+/HER2− breast cancer diagnosed 2010–2019 in the National Cancer Database.

Abbreviations: RS=Oncotype DX recurrence score; SLNB=sentinel lymph node biopsy

Table 3. Multivariate adjusted overall survival model for the overall cohort and stratified by RS Group; based on patients age ≥65y with pT1–2/cN0/M0 HR+/HER2− breast cancer diagnosed 2010–2019 in the National Cancer Database.

In addition to the variables listed, the models were also adjusted for age, sex, race/ethnicity, year of diagnosis, education, income, community type, insurance type, Charlson/Deyo comorbidity score, facility type, facility location, histology, tumor grade, tumor size, and breast surgery type.

Covariates All Patients
(N=61539)		 RS <11
(N=17221)		 RS 11–25
(N=35850)		 RS >25
(N=8468)
HR (95 % CI) P1 P2 HR (95 % CI) P1 P2 HR (95 % CI) P1 P2 HR (95 % CI) P1 P2
RS Group <0.001
RS <11 REF - - - - - - - - -
RS 11–25 1.09 (1.02–1.17) 0.02 - - - - - - - - -
RS >25 1.61 (1.45–1.79) <0.001 - - - - - - - - -
SLNB Receipt/Result <0.001 0.001 <0.001 0.01
None REF REF REF REF
Negative 0.75 (0.65–0.86) <0.001 0.77 (0.59–1.01) 0.06 0.67 (0.55–0.8) <0.001 0.95 (0.7–1.3) 0.74
Positive 1.04 (0.89–1.21) 0.63 1.03 (0.76–1.39) 0.87 0.94 (0.76–1.16) 0.55 1.3 (0.92–1.85) 0.14
Treatment with Chemotherapy <0.001 0.31 0.17 <0.001
No REF REF REF REF
Yes 0.82 (0.74–0.92) <0.001 0.76 (0.45–1.29) 0.31 0.89 (0.75–1.05) 0.17 0.74 (0.65–0.86) <0.001
Treatment with Radiation Therapy <0.001 0.001 <0.001 <0.001
No REF REF REF REF
Yes 0.69 (0.63–0.76) <0.001 0.74 (0.62–0.89) 0.001 0.71 (0.63–0.8) <0.001 0.58 (0.48–0.7) <0.001
Treatment with Endocrine Therapy <0.001 <0.001 <0.001 <0.001
No REF REF REF REF
Yes 0.59 (0.53–0.65) <0.001 0.65 (0.55–0.78) <0.001 0.57 (0.5–0.66) <0.001 0.54 (0.45–0.66) <0.001
Open in a new tab
1

Covariate-level p-value.

2

Overall p-value.

Abbreviations: HR=hazard ratio, CI=confidence interval, RS=recurrence score, SLNB=sentinel lymph node biopsy

Cost-Benefit Analysis Baseline Results

If the estimated cost of BCS without SLNB is $8,809 and BCS with upfront SLNB is $12,617 (Supplemental Table 1), then omitting SLNB would save $3,808 per patient who did not undergo upfront SLNB. However, if performing a delayed SLNB cost $19,866, then delayed SLNB would cost an additional $7,249 per patient, compared to those undergoing upfront SLNB. Based on our finding that SLNB receipt/result was only associated with survival outcomes for those with an intermediate RS, then 31,521 out of 75,428 patients (42% of the original cohort) could potentially be spared the morbidity associated with a SLNB, and it would result in a cost savings of $120,031,968 (assuming everyone with a low or high RS was previously receiving a SLNB and SLNB was instead omitted for all of those patients). However, if all patients with an intermediate RS (n=43,907) underwent a delayed SLNB instead of an upfront SLNB, this would cost an additional $318,281,843. Taken together with the savings of omitting SLNB for low and high RS patients, this would still result in a net increase in cost of $198,249,875, or $2,718 per patient, suggesting that delaying surgical decisions regarding SLNB is not currently cost-effective when the RS is obtained from the surgical breast specimen.

DISCUSSION

Given the heterogeneity observed among older patients with breast cancer, studies have sought to determine when certain therapies can be de-escalated,4,10,18 including surgery.19,20 However, recent publications have consistently demonstrated the importance of tumor biology as a key factor in determining outcomes,9,10,13–15 thus causing controversy in the continued utility of assessing anatomic risk (T and N grouping).21 The older breast cancer population serves as the ideal cohort to further examine this balance of risk factors, as there is significant heterogeneity in the management of these patients.22 In the current study, we sought to explore the potential differences in chemotherapy receipt and/or survival among older patients with breast cancer based on RS and SLNB receipt/result. In this study of >75,000 patients aged ≥65y with early-stage HR+/HER2− breast cancer, we demonstrated that factors associated with chemotherapy receipt, including tumor characteristics and younger age, were similar to those previously reported in the literature to be associated with chemotherapy receipt in younger patients.23 Interestingly, chemotherapy was recommended for 17.8% of patients, yet only 64.6% of those who were recommended chemotherapy proceeded to receive it. In addition, SLNB receipt/result was associated with survival for those with an intermediate RS, but not for those with a low or high RS. While 11.6% of patients who underwent SLNB were node positive, 98.3% of those patients only had one positive node. Furthermore, nodal positivity rates were similar regardless of RS, suggesting that RS may not necessarily reflect nodal burden, similar to findings from other groups.24 The association of nodal positivity with compromised survival was also similar across all RS groups, suggesting that nodal positivity may not provide further prognostic information than RS alone.

Recent studies of older patients with both a RS and SLNB found chemotherapy was more likely to be guided by Oncotype DX score than SLNB results,25,26 suggesting that SLNB may have little impact on chemotherapy decisions in older patients with early breast cancer and a RS. For example, Lee et al. demonstrated that SLNB changed the recommendation for/against chemotherapy, or the regimen, in only 4.8% of postmenopausal women with early-stage, ER-positive/HER2-negative breast cancer, and a sonographically negative axilla, thus supporting de-escalation of axillary surgery using a RS score.26 Similarly, Davey et al. found in a study of 253 patients aged >65y with ER-positive/HER2-negative, clinically node negative breast cancer, that a RS score >25 was an independent predictor of adjuvant chemotherapy receipt, despite all patients undergoing SLNB.25

In this large, retrospective study, we found that chemotherapy recommendation and receipt were higher with a higher RS, even after adjustment for relevant covariates. This is in line with the above findings that a RS can help guide chemotherapy decision making, potentially allowing for further de-escalation of axillary surgery in older patients and omitting a SLNB. However, we found that SLNB performance was significantly associated with survival for patients with an intermediate RS, but not a low or a high RS. While this supports omitting SLNB for those with a low or high RS, this suggests that there may still be a role for SLNB in patients with an intermediate RS.

This study must be considered in light of the RxPONDER prospective trial data recently published by Kalinsky et al., who showed that postmenopausal women with one to three positive LN and an RS ≤25, who received chemotherapy and endocrine therapy, did not have a longer invasive disease-free survival and distant relapse-free survival than those who received endocrine therapy alone, whereas premenopausal patients with the same characteristics did benefit from adjuvant chemotherapy.10

Our finding of an association with survival for SLNB in older patients with an intermediate RS shows that the patients who underwent a SLNB and had negative nodes had better survival than those who did not undergo SLNB, whereas patients who underwent a SLNB and had positive nodes had the same survival as patients who did not undergo SLNB. Thus, while the findings from Kalinsky et al. suggest that even those with an intermediate RS would not benefit from adjuvant chemotherapy, and therefore may not require a SLNB to aid in adjuvant chemotherapy decision making, our findings show that older patients with an intermediate RS may still benefit from undergoing SLNB for a different purpose: providing prognostic survival information, which is in line with the purpose of SLNB as a prognostic procedure. While prognostic information may not be important to all patients, there are likely some patients that would appreciate the improved estimation derived from having this information. In addition, some providers may still use the information obtained from a SLNB as an additional data point for consideration, among other important variables, when making adjuvant systemic therapy decisions.

These observations of an association with survival for SLNB in patients with an intermediate RS led to our cost-benefit analysis of performing SLNB only for patients with an intermediate RS (11–25). When considering de-escalation from a cost-benefit perspective, it is unsurprising that the Delayed SLNB approach was more costly, given that it required return to the OR for nearly half of our cohort. However, the majority of this additional cost could be avoided if the RS were sent from the biopsy specimen instead of the surgical specimen, thus having the RS result (low/intermediate/high) prior to performing the first surgery. Given that prior studies have shown a 72% concordance in RS group classification between biopsy and surgical specimens,27 this could represent a feasible approach. Alternatively, recent studies have suggested that a lower intermediate RS (11–17) may have similar prognostic implications to a low RS (<11).28 As such, if the value of a SLNB was not significantly associated with survival for those with a RS of 11–17, then a SLNB may only be required or those with a RS of 18–25, which could dramatically shift the cost-benefit analysis in favor of the Delayed SLNB approach. One limitation of our cost-benefit analysis is the assumption that all patients would receive RS testing, which is not the case in real-world practice, as evidenced by our need to exclude patients without a RS score.

Regardless, the financial cost of a SLNB is only one consideration, and a SLNB can be associated with other morbidities, such as chronic pain, decreased range of motion, and lymphedema.29 Lymphedema in particular may impact body image, quality of life, physical, psychological, and social function.30 In addition, lymphedema treatment can be costly. The lifetime costs associated with a diagnosis of lymphedema has been estimated at nearly $14,000, greatly exceeding our estimate of the cost of avoiding SLNB in eligible patients.31 Thus, while upfront surgery costs should be a consideration, there is evidence from a lifetime cost perspective, as well as a quality-of-life perspective, to avoid SLNB in select older patients.

Additional studies offer further support that SLNB may be unnecessary in older patients. A cost-effectiveness study by McEvoy et al. found that postmenopausal women with clinical T1-T2, sonographically N0-N1, HR-positive, HER2-negative breast cancer had a higher number of quality-life-years with axillary observation compared to SLNB.32 Another study examining pathological node positivity in clinically node negative patients found that among 1229 women aged >65y with unifocal tumors ≤ 2 cm, only 1% (n = 12) had pN2–3 disease on final pathology, suggesting that pN2–3 nodal burden is uncommon among these patients, similar to our findings, and omitting SLNB would not miss a significant number of patients who might benefit from adjuvant chemotherapy.33

Excluding patients who underwent axillary lymph node dissection (ALND), defined here as those who had greater than 5 nodes removed, was another deliberate choice in this investigation. While the findings of Donker et al. from the AMAROS trial34 and Giuliano et al. from the Z11 trial35 were not published until 2014 and 2017, respectively, meaning some patients in the NCDB from 2010–2019 were treated prior to the publication of these studied and likely underwent ALND for only 1–2 positive nodes, we chose to include only patients who axillary surgical management reflect current practice.

Limitations

Some limitations in our study result from the use of the NCDB for our analysis. Although it is the largest national tumor registry in the US and includes >80% of patients diagnosed with breast cancer each year,36 it is prone to human errors with data entry and other potential inaccuracies.37 In addition, survival outcomes may actually be worse than we found, as patients with missing data were excluded, and these patients have been shown to have worse survival outcomes.38 Additionally, cancer-specific survival is not recorded in the NCDB, and thus survival data in this study reflect mortality from all causes, although this is likely an acceptable end-point for older patient populations.

Also, our analyses may be limited by the fact that patients without a RS were excluded. Patients with no indication for chemotherapy would not need to receive RS testing, and therefore, the rate of chemotherapy in our study population may be overstated. Finally, the patients who had the greatest survival were those who had a low RS and received chemotherapy. While we recognize that this does not align with current treatment practices, it may reflect previous standards or factors unique to these individuals.

Conclusions

Factors associated with chemotherapy receipt in older patients with breast cancer are similar to those of younger patients in the literature. Notably, SLNB receipt/result was associated with survival for those with an intermediate RS, but not for those with a low or high RS, suggesting that a SLNB may indeed be unnecessary for select older patients based on RS. However, using the RS from a preoperative core needle biopsy to guide operative decision-making would benefit from prospective trials. Further study is needed to better characterize which older patients would most benefit from avoiding versus undergoing SLNB.

Supplementary Material

1

ACKNOWLEDGEMENTS

The National Cancer Data Base (NCDB) is a joint project of the Commission on Cancer (CoC) of the American College of Surgeons and the American Cancer Society. The CoC’s NCDB and the hospitals participating in the CoC’s NCDB are the source of the de-identified data used herein; they have not verified and are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors.

FUNDING:

  • Dr. Plichta is supported by the National Institutes of Health Office of Women’s Research Building Interdisciplinary Research Careers in Women’s Health K12AR084231 (PI: Amundsen).

  • This work was in part supported by Duke Cancer Institute through NIH grant P30CA014236 (PI: Kastan) for the Biostatistics Core.

DISCLOSURES

  • The authors report no proprietary or commercial interest in any product mentioned or concept discussed in this article.

  • Dr. J.K. Plichta was the recipient of research funding by the Color Foundation and Earlier.org (PI: Plichta). As the chair of the AAS Publications Committee, she serves on the JSR editorial board, but was not involved in the editorial decision-making process. She is also a member of Editorial Committee for the American Joint Committee on Cancer (AJCC), the Breast Panel for the AJCC, and the National Comprehensive Cancer Network (NCCN) Breast Cancer Diagnosis and Screening panel.

  • Dr. E.S. Hwang serves on the National Cancer Institute (NCI) Breast Cancer Steering Committee and the NCCN Breast Cancer Prevention Committee.

  • Dr L.H. Rosenberger serves on the NCCN Breast Cancer Panel.

  • The content of this manuscript is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health (NIH).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

REFERENCES

  • 1.Vespa J The Graying of America: More Older Adults Than Kids by 2035. United States Census Bureau. Updated 10/August/2019. Accessed 4/6/20253, https://www.census.gov/library/stories/2018/03/graying-america.html [Google Scholar]
  • 2.Key Statistics for Breast Cancer: How common is breast cancer? website. American Cancer Society. Updated 1/December/2023. Accessed 2/9/2023, https://www.cancer.org/cancer/breast-cancer/about/how-common-is-breast-cancer.html [Google Scholar]
  • 3.Ma SJ, Oladeru OT, Singh AK. Outcome of adjuvant chemotherapy in elderly patients with early-stage, hormone receptor-positive, HER-2-negative breast cancer. The Breast Journal. 2020 2020;26(10):2026–2030. doi: 10.1111/tbj.14054 [DOI] [PubMed] [Google Scholar]
  • 4.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. Journal of clinical oncology: official journal of the American Society of Clinical Oncology. 2013;31(19):2382–2387. doi: 10.1200/JCO.2012.45.2615 [doi] [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Kunkler IH, Williams LJ, Jack WJL, Cameron DA, Dixon JM. Breast-conserving surgery with or without irradiation in women aged 65 years or older with early breast cancer (PRIME II): a randomised controlled trial. The Lancet Oncology. 2015/March/01/ 2015;16(3):266–273. doi: 10.1016/S1470-2045(14)71221-5 [DOI] [PubMed] [Google Scholar]
  • 6.Martelli G, Boracchi P, De Palo M, et al. A Randomized Trial Comparing Axillary Dissection to No Axillary Dissection in Older Patients With T1N0 Breast Cancer. Annals of Surgery. 2005/July// 2005;242(1):1–6. doi: 10.1097/01.sla.0000167759.15670.14 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.International Breast Cancer Study G, Rudenstam C-M, Zahrieh D, et al. Randomized trial comparing axillary clearance versus no axillary clearance in older patients with breast cancer: first results of International Breast Cancer Study Group Trial 10–93. J Clin Oncol. 2006/January/20/ 2006;24(3):337–344. doi: 10.1200/JCO.2005.01.5784 [DOI] [PubMed] [Google Scholar]
  • 8.Tamirisa N, Thomas SM, Fayanju OM, et al. Axillary Nodal Evaluation in Elderly Breast Cancer Patients: Potential Effects on Treatment Decisions and Survival. Ann Surg Oncol. 2018/October//2018;25(10):2890–2898. doi: 10.1245/s10434-018-6595-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Sparano JA, Gray RJ, Makower DF, et al. Adjuvant Chemotherapy Guided by a 21-Gene Expression Assay in Breast Cancer. N Engl J Med. Jun 3 2018;doi: 10.1056/NEJMoa1804710 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Kalinsky K, Barlow WE, Gralow JR, et al. 21-Gene Assay to Inform Chemotherapy Benefit in Node-Positive Breast Cancer. New England Journal of Medicine. 2021;385(25):2336–2347. doi: 10.1056/NEJMoa2108873 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Syed YY. Oncotype DX Breast Recurrence Score®: A Review of its Use in Early-Stage Breast Cancer. Mol Diagn Ther. 2020/October/01/ 2020;24(5):621–632. doi: 10.1007/s40291-020-00482-7 [DOI] [PubMed] [Google Scholar]
  • 12.Sparano JA, Gray RJ, Makower DF, et al. Prospective Validation of a 21-Gene Expression Assay in Breast Cancer. The New England journal of medicine. 2015;373(21):2005–2014. doi: 10.1056/NEJMoa1510764 [doi] [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Cardoso F, van’t Veer LJ, Bogaerts J, et al. 70-Gene Signature as an Aid to Treatment Decisions in Early-Stage Breast Cancer. N Engl J Med. Aug 25 2016;375(8):717–29. doi: 10.1056/NEJMoa1602253 [DOI] [PubMed] [Google Scholar]
  • 14.Sparano JA, Gray RJ, Ravdin PM, et al. Clinical and Genomic Risk to Guide the Use of Adjuvant Therapy for Breast Cancer. New England Journal of Medicine. 2019/June/20/2019;380(25):2395–2405. doi: 10.1056/NEJMoa1904819 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Sparano JA, Crager MR, Tang G, Gray RJ, Stemmer SM, Shak S. Development and Validation of a Tool Integrating the 21-Gene Recurrence Score and Clinical-Pathological Features to Individualize Prognosis and Prediction of Chemotherapy Benefit in Early Breast Cancer. Journal of Clinical Oncology. 2021/February/20/ 2021;39(6):557–564. doi: 10.1200/JCO.20.03007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Gemignani ML, Cody HS, Fey JV, Tran KN, Venkatraman E, Borgen PI. Impact of sentinel lymph node mapping on relative charges in patients with early-stage breast cancer. Ann Surg Oncol. 2000/September/01/ 2000;7(8):575–580. doi: 10.1007/BF02725336 [DOI] [PubMed] [Google Scholar]
  • 17.Bureau of Labor Statistics Data.
  • 18.Kunkler IH, Williams LJ, Jack WJ, Cameron DA, Dixon JM. 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. Mar 2015;16(3):266–73. doi: 10.1016/s1470-2045(14)71221-5 [DOI] [PubMed] [Google Scholar]
  • 19.Marks CE, Ren Y, Rosenberger LH, et al. Surgical Management of the Axilla in Elderly Women With Node-positive Breast Cancer. J Surg Res. May 29 2020;254:275–285. doi: 10.1016/j.jss.2020.04.036 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Pepping RMC, Portielje JEA, van de Water W, de Glas NA. Primary Endocrine Therapy in Older Women with Breast Cancer. Curr Geriatr Rep. 2017;6(4):239–246. doi: 10.1007/s13670-017-0223-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Plichta JK, Campbell BM, Mittendorf EA, Hwang ES. Anatomy and Breast Cancer Staging: Is It Still Relevant? Surg Oncol Clin N Am. Jan 2018;27(1):51–67. doi: 10.1016/j.soc.2017.07.010 [DOI] [PubMed] [Google Scholar]
  • 22.Plichta JK, Thomas SM, Vernon R, et al. Breast cancer tumor histopathology, stage at presentation, and treatment in the extremes of age. Breast Cancer Research and Treatment. 2020/February/01 2020;180(1):227–235. doi: 10.1007/s10549-020-05542-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Griggs JJ, Hawley ST, Graff JJ, et al. Factors Associated With Receipt of Breast Cancer Adjuvant Chemotherapy in a Diverse Population-Based Sample. Journal of Clinical Oncology. 2012/September/01/2012;30(25):3058–3064. doi: 10.1200/JCO.2012.41.9564 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Bello DM, Russell C, McCullough D, Tierno M, Morrow M. Lymph Node Status in Breast Cancer Does Not Predict Tumor Biology. Ann Surg Oncol. Oct 2018;25(10):2884–2889. doi: 10.1245/s10434-018-6598-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Davey MG, Ryan ÉJ, Burke D, et al. Evaluating the Clinical Utility of Routine Sentinel Lymph Node Biopsy and the Value of Adjuvant Chemotherapy in Elderly Patients Diagnosed With Oestrogen Receptor Positive, Clinically Node Negative Breast Cancer. Breast Cancer(Auckl). 2021/January/01/2021;15:11782234211022203. doi: 10.1177/11782234211022203 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Lee E, Tweed C, Mylander C, et al. The Impact of Genomic Profiling on Adjuvant Therapy Recommendation in Postmenopausal Women with ER-Positive, T1–2 Breast Cancer: Can Genomic Profiling Eliminate the Need for Sentinel Lymph Node Biopsy? Clinical Breast Cancer. 2021/December/01/2021;21(6):e731–e737. doi: 10.1016/j.clbc.2021.02.011 [DOI] [PubMed] [Google Scholar]
  • 27.Qi P, Yang Y, Bai Q-m, et al. Concordance of the 21-gene assay between core needle biopsy and resection specimens in early breast cancer patients. Breast Cancer Research and Treatment. 2021/April/01 2021;186(2):327–342. doi: 10.1007/s10549-020-06075-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Kantor O, King TA, Shak S, et al. Expanding Criteria for Prognostic Stage IA in Hormone Receptor-Positive Breast Cancer. J Natl Cancer Inst. Nov 29 2021;113(12):1744–1750. doi: 10.1093/jnci/djab095 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Wilke LG, McCall LM, Posther KE, et al. Surgical complications associated with sentinel lymph node biopsy: results from a prospective international cooperative group trial. Ann Surg Oncol. Apr 2006;13(4):491–500. doi: 10.1245/aso.2006.05.013 [DOI] [PubMed] [Google Scholar]
  • 30.Pusic AL, Cemal Y, Albornoz C, et al. Quality of life among breast cancer patients with lymphedema: a systematic review of patient-reported outcome instruments and outcomes. J Cancer Surviv. Mar 2013;7(1):83–92. doi: 10.1007/s11764-012-0247-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Squitieri L, Rasmussen PW, Patel KM. An economic analysis of prophylactic lymphovenous anastomosis among breast cancer patients receiving mastectomy with axillary lymph node dissection. J Surg Oncol. 2020/June//2020;121(8):1175–1178. doi: 10.1002/jso.25902 [DOI] [PubMed] [Google Scholar]
  • 32.McEvoy AM, Poplack S, Nickel K, et al. Cost-effectiveness analyses demonstrate that observation is superior to sentinel lymph node biopsy for postmenopausal women with HR + breast cancer and negative axillary ultrasound. Breast Cancer Res Treat. 2020/September/01/2020;183(2):251–262. doi: 10.1007/s10549-020-05768-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Lee MK, Montagna G, Pilewskie ML, Sevilimedu V, Morrow M. Axillary Staging Is Not Justified in Postmenopausal Clinically Node-Negative Women Based on Nodal Disease Burden. Ann Surg Oncol. 2023/January/01/ 2023;30(1):92–97. doi: 10.1245/s10434-022-12203-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Donker M, van Tienhoven G, Straver ME, et al. Radiotherapy or surgery of the axilla after a positive sentinel node in breast cancer (EORTC 10981–22023 AMAROS): a randomised, multicentre, open-label, phase 3 non-inferiority trial. Lancet Oncol. 2014/11//2014;15(12):1303–1310. doi: 10.1016/S1470-2045(14)70460-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Giuliano AE, Ballman KV, McCall L, et al. Effect of Axillary Dissection vs No Axillary Dissection on 10-Year Overall Survival Among Women With Invasive Breast Cancer and Sentinel Node Metastasis: The ACOSOG Z0011 (Alliance) Randomized Clinical Trial. JAMA. 2017/September/12/2017;318(10):918–926. doi: 10.1001/jama.2017.11470 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Mallin K, Browner A, Palis B, et al. Incident Cases Captured in the National Cancer Database Compared with Those in U.S. Population Based Central Cancer Registries in 2012–2014. Ann Surg Oncol. Feb 8 2019;doi: 10.1245/s10434-019-07213-1 [DOI] [PubMed] [Google Scholar]
  • 37.Boffa DJ, Rosen JE, Mallin K, et al. Using the National Cancer Database for Outcomes Research: A Review. JAMA Oncology. 2017;3(12):1722–1728. doi: 10.1001/jamaoncol.2016.6905 [DOI] [PubMed] [Google Scholar]
  • 38.Plichta JK, Rushing CN, Lewis HC, et al. Implications of missing data on reported breast cancer mortality. Breast Cancer Res Treat. 2023/January/01/ 2023;197(1):177–187. doi: 10.1007/s10549-022-06764-4 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1961912-supplement-1.docx (38.9KB, docx)

RESOURCES