Axillary Staging Is Not Justified in Postmenopausal Clinically Node-Negative Women Based on Nodal Disease Burden

Abstract

Background:

RxPONDER showed no benefit of adjuvant chemotherapy in postmenopausal women with ER+/HER2− breast cancer and limited nodal burden (pN1) with a recurrence score≤25, suggesting that axillary staging could be omitted in cN0 patients if significant numbers of such women do not have pN2-3 disease. Here we evaluate the pN2-3 disease rate in a large cohort of postmenopausal women presenting with cN0 breast cancer.

Methods:

Consecutive postmenopausal cT1-2N0 patients who underwent axillary surgery from 02/2006-12/2011 were identified. Clinicopathologic characteristics associated with pN2-3 disease were examined using Chi-square or Fisher’s exact tests.

Results:

Of 3363 postmenopausal women with cT1-2N0 breast cancer(median age 58y,IQR 48-67), median tumor size was 1.3cm(IQR 0.90-1.90cm). Post-axillary staging, 2600(77.3%) were pN0,643(19.1%) pN1, and 120(3.6%) pN2-3. pN2-3 disease rate did not differ across subtypes(4.4% HER2+,3.5% HR−/HER2−,3.5% HR+/HER2−,P=0.70). In the subset with HR+/HER2− tumors, on multivariable analysis age<65y (OR 2.38,95%CI 1.32-4.49), lymphovascular invasion (OR 5.29,95%CI 2.72-11.2), multifocal/centric tumors (OR 3.08,95%CI 1.79-5.32), and tumor size >2cm (OR 5.51,95%CI 3.05-10.4) were significantly associated with pN2-3 nodal burden. Of 506 patients with tumors >2cm, 49(9.7%) had pN2-3 disease; in the subset of 90 patients age <65y who had multifocal/centric tumors >2cm; 23(25.6%) had pN2-3 disease.

Conclusions:

In postmenopausal women with cN0 disease, pN2-3 nodal burden is uncommon; omitting axillary staging would not miss a significant number of patients who might benefit from adjuvant chemotherapy. Information available preoperatively indicating a higher risk of nodal disease such as younger age and large, multifocal tumors, should be considered in the multidisciplinary management of the axilla.

INTRODUCTION

Surgical treatment of the axilla in patients with early-stage breast cancer has evolved significantly over time.^1–3 Sentinel lymph node biopsy (SLNB), rather than axillary lymph node dissection (ALND), is now the standard axillary staging procedure for patients with clinically node-negative (cN0), early-stage breast cancer. With multiple randomized trials failing to identify a survival benefit for more radical nodal surgery, SLNB has replaced ALND as therapy in patients with a limited nodal disease burden.^4–7 However, nodal staging has continued to provide key prognostic data and inform adjuvant systemic treatment recommendations for patients with early-stage breast cancer.⁸

While SLNB has significantly decreased morbidity compared to ALND, the procedure is associated with pain, lymphedema, and sensory nerve damage.⁹ Therefore, in patients with cN0 disease, axillary staging should be performed only if lymph node status is needed for adjuvant treatment decisions. The widespread use of genomic assays provides an opportunity to further de-escalate axillary surgery in patients with hormone receptor (HR) positive, human epidermal growth factor receptor 2 (HER2) negative breast cancer.

Until relatively recently, the presence of nodal metastases was considered an indication for chemotherapy. Large prospective studies have demonstrated the clinical validity of the 21-gene recurrence score to predict the benefit of adding chemotherapy to endocrine therapy in node-negative, HR+/HER2− breast cancer in both pre- and postmenopausal women.^10,11 The recent results of RxPONDER showed no benefit of adjuvant chemotherapy in postmenopausal women with HR+/HER2− breast cancer and 1-3 positive lymph nodes with a recurrence score ≤ 25. In contrast, for premenopausal women with similar characteristics, there was a significant benefit to chemotherapy regardless of recurrence score, and therefore axillary staging for premenopausal patients continues to inform adjuvant treatment decisions.¹²

In light of these findings, recurrence score results, rather than the presence or absence of nodal metastases, now guides adjuvant treatment recommendations in postmenopausal women with HR+ breast cancer with limited nodal burden (pN1), raising the possibility that in cN0 postmenopausal patients meeting RxPONDER criteria, routine SLNB could potentially be omitted if a significant proportion of patients do not have involvement of more than 3 lymph nodes (pN2-3). This study sought to evaluate the rate of pN2-3 nodal disease in postmenopausal patients with cN0 breast cancer and to identify factors associated with heavy nodal burden.

METHODS

Study Population

Upon Memorial Sloan Kettering Cancer Center (MSK) Institutional Review Board approval, consecutive postmenopausal patients with cT1-2N0 who underwent breast and axillary surgery at MSK between February 2006 and December 2011 were identified. Prior to 2010, it was routine practice to perform a completion ALND for any patient with a pathologically positive sentinel lymph node. After 2010 and the publication of the American College of Surgeons Oncology Group (ACOSOG) Z0011 trial, patients who met eligibility criteria underwent ALND only for 3 or more positive SLNs, or gross extracapsular extension of tumor in the SLNs. Axillary ultrasound was not routinely done as part of the diagnostic work-up, and patients were classified as cN0 by physical exam alone. Multicentricity/multifocality was determined radiographically. Patients undergoing neoadjuvant chemotherapy were excluded.

Statistical Analysis

Clinical and pathological characteristics associated with pN2-3 disease were compared using the Wilcoxon rank sum test or t-test for continuous variables, and the Chi-square or Fisher’s exact test for categorical variables. Univariate and multivariable logistic regressions were conducted to assess the association between pN2-3 status and clinicopathological factors. All covariates that were significant at the α=0.05 level in the univariate analysis were included in the multivariable analysis. The final model was determined using backward selection. Statistical analysis was performed using R 3.6.3 (R Core Development Team, Vienna, Austria).

RESULTS

Patient Characteristics

During the study period, 5142 consecutive patients with cT1-2N0 breast cancer were identified; 3363 were postmenopausal and comprised the study cohort. The mean patient age was 58 years (interquartile range [IQR] 48, 67). Most patients had ductal histology (n = 2734, 81.3%) with a median tumor size of 1.3 cm (IQR 0.90, 1.90 cm). The majority were HR+/HER2− (n = 2639, 78.5%). Of the remainder, 12% were HR−/HER2− (n = 402), and 297 (8.8%) were HER2+. Most patients underwent SLNB alone (n = 2707, 80.5%), 2381 (70.8%) had breast-conservation surgery, 982 (29.2%) had mastectomy, and only 656 (19.5%) had SLNB followed by ALND. On final pathology 2600 (77.3%) patients were pN0, 643 (19.1%) were pN1, and 120 (3.6%) had pN2-3 disease. The rate of pN2-3 nodal disease did not differ significantly by HR or HER2 status (3.5% HR+/HER2−, 4.4% HER2+, 3.5% HR−/HER2−, P = 0.70).

Factors Associated With pN2-3 Disease

Of the 2639 postmenopausal women with HR+/HER2− tumors, only 92 (3.5%) had pN2-3 nodal disease. Compared to patients with pN0-1 disease, those with pN2-3 nodal burden were more likely to be less than 65 years of age (65.2% versus 50.8%, P = 0.009), to have lobular histology (34.8% versus 15.1%, P < 0.001), to have lymphovascular invasion (LVI) (67.4% versus 23.0%, P < 0.001), and to have multifocal or multicentric tumors (51.1% versus 20.8%, P < 0.001). Those with pN2-3 nodal disease had significantly larger tumors, with most having a pathological size greater than 2 cm (68.5% versus 16.2%, P < 0.001) compared to pN0-1 patients (Table 1). On multivariable analysis, age less than 65 years (odds ratio [OR] 2.38, 95% confidence interval [CI] 1.32-4.49), LVI (OR 5.29, 95% CI 2.72-11.2), multifocal/centric tumors (OR 3.08, 95% CI 1.79-5.32), and tumor size greater than 2 cm (OR 5.51, 95% CI 3.05-10.4) were significantly associated with pN2-3 nodal burden (Table 2).

TABLE 1.

Clinical and pathological features of the ER+/HER2− subset

Variable	Overall (n = 2639)	pN0-1 (n = 2547)	pN2-3 (n = 92)	P value
Age, n (%)				0.009
< 65 years	1354 (51.3)	1294 (50.8)	60 (65.2)
≥ 65 years	1285 (48.7)	1253 (49.2)	32 (34.8)
Abnormal breast exam* (%)	1076 (40.8)	1010 (39.6)	66 (71.7)	< 0.001
Clinical T stage (%)				< 0.001
T1mic-T1a	156 (5.9)	156 (6.1)	0 (0)
T1b	884 (33.5)	876 (34.4)	8 (8.7)
T1c	1093 (41.4)	1058 (41.5)	35 (38.0)
T2	506 (19.2)	457 (17.9)	49 (53.3)
Histology (%)**				< 0.001
Ductal	2066 (78.3)	2009 (78.8)	57 (62.0)
Lobular	416 (15.8)	384 (15.1)	32 (34.8)
Differentiation (%)**				< 0.001
Well	193 (7.3)	193 (7.6)	0 (0)
Moderate	849 (32.2)	835 (32.8)	14 (15.2)
Poor	1155 (43.8)	1103 (43.3)	52 (56.5)
Lymphovascular invasion (%)	647 (24.5)	585 (23.0)	62 (67.4)	< 0.001
Multifocal/centric (%)	577 (21.9)	530 (20.8)	47 (51.1)	< 0.001
Tumor size (%)**				< 0.001
< 5 mm	268 (10.2)	268 (10.5)	0 (0)
5 mm-1 cm	747 (28.3)	742 (29.1)	5 (5.4)
1 cm-2 cm	1129 (42.8)	1107 (43.4)	22 (23.9)
> 2 cm	477 (18.1)	414 (16.2)	63 (68.5)

^*Abnormal breast exam includes palpable mass, skin/nipple changes

^**Unknowns not included

ER estrogen receptor, HER2 human epidermal growth factor receptor 2

TABLE 2.

Multivariable analysis of risk factors for pN2-3 nodal disease

Variable	OR	95% CI	P value*
Age
≥ 65 years	Ref
< 65 years	2.38	1.32-4.49	0.005
LVI
Absent	Ref
Present	5.29	2.72-11.2	< 0.001
Multifocal/centric
No	Ref
Yes	3.08	1.79-5.32	< 0.001
Tumor size
< 5 mm	0.00	NA	> 0.9
5 mm-1 cm	0.69	0.19-1.95	0.5
1 cm-2 cm	Ref
> 2 cm	5.51	3.05-10.4	< 0.001

^*Bold indicates values of significance

CI confidence interval, LVI lymphovascular invasion

Factors Associated With pN2-3 Disease Known Before Surgery

Of the factors that remained statistically significant on multivariable analysis, only age, multifocality/centricity, and an estimate of tumor size are commonly known preoperatively. Rates of pN2-3 were 4.1% (75/1821) among patients age < 65 years, 7.7% (56/728) among those with multifocal/centric tumors, and 12.6% (86/685) among those with tumors larger than 2 cm. The rate of pN2-3 increased significantly in patients when all 3 factors were present: among the 90 patients younger than 65 years of age with multifocal/centric tumors larger than 2 cm, 23 (25.6%) were found to have pN2-3 nodal burden. Conversely, among the 1229 women > 65 years of age with unifocal tumors ≤ 2 cm, only 1% (n = 12) had pN2-3 disease on final pathology.

DISCUSSION

The de-escalation of axillary surgery continues to expand. In 2016 the Society of Surgical Oncology Choosing Wisely recommendations included the omission of SLNB in patients over 70 years of age with early-stage, HR+ breast cancer.¹³ These guidelines were based on studies showing no difference in recurrence or overall survival when axillary surgery was deferred in this subset of older women.^14–16 Results from the RxPONDER trial now also confirm that axillary staging does not inform adjuvant chemotherapy decisions in postmenopausal women with HR+/HER2− breast cancer, 1-3 positive lymph nodes, and a recurrence score ≤ 25.¹² Although an important prognostic indicator, axillary lymph node status is no longer the critical factor in determining which patients need adjuvant chemotherapy. However, this is true only in patients with limited nodal burden, raising the question of whether SLNB is necessary in all cN0 postmenopausal HR+/HER2− patients to exclude the presence of pN2-3 disease. In this large cohort of postmenopausal patients with cT1-2N0, HR+/HER2− tumors, we showed that pN2-3 nodal disease is uncommon, even when patients are staged as cN0 without the routine use of axillary ultrasound.

We found an overall very low rate of pN2-3 disease in all postmenopausal women (3.6%) and the subset of postmenopausal women with HR+/HER2− tumors (3.5%). As most of the cohort was treated prior to the implementation of the ACOSOG Z0011 results into practice, patients with positive SLNs underwent completion ALND, allowing us to evaluate the true nodal burden in this population. Our findings mirror the initial reports from the SOUND and INSEMA trials. The SOUND trial randomized patients with cT1N0 breast cancer (confirmed by preoperative axillary ultrasound) planned for breast-conservation surgery and whole-breast irradiation (WBI) to SLNB or no axillary surgery. Recently reported results of the first 176 patients enrolled showed that of the 94 patients in the SLNB arm, all patients had pN0-1 nodal disease.¹⁷ Similarly, the INSEMA trial is investigating omission of axillary surgery in cT1-2N0 patients selected with axillary ultrasound , with preliminary results showing that only 0.3% (12/4124) of patients randomized to SLNB had pN2-3 disease.¹⁸

In our study, several factors were associated with pN2-3 nodal disease, including age less than 65 years, tumor size larger than 2 cm, multifocality or multicentricity, and the presence of LVI. Many studies have investigated the prevalence of pN2-3 nodal disease or the rate of positive non-SLNs; however, these analyses include many variables, such as LVI, extracapsular extension, and size of nodal macrometastasis, which are not known until final pathological examination and which are therefore of limited clinical utility when considering surgical de-escalation.^19,20 Of the risk factors identified in our study, age, multifocality/centricity, and clinical tumor size can be determined preoperatively and may help direct axillary management. Several studies support tumor size and multifocality/centricity as factors associated with a heavier nodal burden. For example, in a study of 285 patients from 2 comprehensive cancer centers, Rivers at al. found that tumor size larger than 2 cm was associated with pN2-3 disease (OR 3.24, 95% CI 1.38-7.61).²¹ In a retrospective review, Lynch et al. found that multifocal and multicentric tumors were significantly associated with both pN2 and pN3 nodal burden.²² Our finding that 25% of the postmenopausal women in our study were less than 65 years of age and had multifocal/centric tumors larger than 2 cm suggests that for a significant minority of patients, SLNB should remain standard or that they should undergo multidisciplinary discussion prior to omission of axillary staging.

Such a multidisciplinary discussion is also important given the recently published results from the monarchE trial which demonstrated the benefit of abemaciclib, a CDK4/6 inhibitor, combined with endocrine therapy for patients with HR+/HER2− node-positive breast cancer at high risk for early recurrence.²³ The definition of high risk in this study included patients with 4 or more positive lymph nodes, or 1-3 positive lymph nodes and either tumor size ≥ 5 cm, histological grade 3, or central Ki67 ≥ 20%. Omission of axillary staging would not miss significant numbers of cN0 postmenopausal patients who would benefit from the addition of abemaciclib due to the presence of 4 or more positive lymph nodes, but may exclude women with 1-3 positive and other high-risk factors found only on final pathology from treatment with abemaciclib. This constellation of findings would be an indication to return to the operating room for axillary staging in a small subset of patients.

At the present time, the greatest barrier to elimination of axillary staging in the large population of postmenopausal women with HR+/HER2− breast cancer is the primacy of nodal status in decisions regarding the use of regional nodal irradiation (RNI). The MA.20 trial randomized patients with node-positive or high-risk node-negative breast cancer undergoing breast-conservation surgery and adjuvant systemic therapy to either WBI plus RNI or WBI alone. With a median follow-up of 9.5 years, there was no difference in overall survival; however, a significant decrease in locoregional and distant recurrences was seen with the addition of RNI.²⁴ Similarly, the European Organisation for Research and Treatment of Cancer (EORTC) 22922/10925 trial demonstrated improvements in disease-free survival, distant disease-free survival, and breast cancer mortality with RNI.²⁵ As a result, nodal staging remains critical for decisions regarding RNI, and the omission of axillary staging should not prompt the general use of RNI. Whether the indications for RNI can be refined based upon breast cancer subtype, genomic tests, and tumor features other than nodal status is the subject of ongoing clinical trials.

The limitations of this study include its retrospective nature and the small overall number of patients with pN2-3 nodal disease. The strengths of this study include its use of a prospectively maintained clinical database with a large cohort of consecutive patients for which granular clinical and pathological data were available.

Conclusions

In postmenopausal women with cN0 disease, pN2-3 nodal burden is uncommon. Omission of axillary staging would not miss a significant number of patients who might benefit from adjuvant chemotherapy based on the presence of metastases in 4 or more axillary nodes. Information available preoperatively indicating a higher risk of nodal disease, such as younger age and large, multifocal tumors, should be considered in the multidisciplinary management of the axilla.

Synopsis:

Here we evaluate the pN2-3 disease rate in postmenopausal women presenting with cT1-2N0 ER+ breast cancer; pN2-3 disease is uncommon. Risk factors include younger age, tumor size, lymphovascular invasion, and multifocality/centricity.