Negative Survival Impact of Occult Lymph Node Involvement in Small HER2-Positive Early Breast Cancer Treated by Up-Front Surgery

Simple Summary Our objective was to investigate the impact of pN0(i+) or pN1mi in HER2-positive breast cancer patients undergoing up-front surgery on their outcomes. Survival was not adversely affected by pN0(i+) and pN1mi in 1771 HER2-positive patients. However, in the case of pT1a-b HER2-positive breast cancers, a negative impact on recurrence-free survival was observed specifically for patients with pN0(i+) and pN1mi diseases, particularly among those with pT1b tumors without adjuvant chemotherapy. Our findings highlight the importance of considering the pN0(i+) and pN1mi status in the decision-making process when discussing trastuzumab-based adjuvant chemotherapy for these patients. Abstract (1) Background: The independent negative prognostic value of isolated tumor cells or micro-metastases in axillary lymph nodes has been established in triple-negative breast cancers (BC). However, the prognostic significance of pN0(i+) or pN1mi in HER2-positive BCs treated by primary surgery remains unexplored. Therefore, our objective was to investigate the impact of pN0(i+) or pN1mi in HER2-positive BC patients undergoing up-front surgery on their outcomes. (2) Methods: We retrospectively analyzed 23,650 patients treated in 13 French cancer centers from 1991 to 2013. pN status was categorized as pN0, pN0(i+), pN1mi, and pNmacro. The effect of pN0(i+) or pN1mi on outcomes was investigated both in the entire cohort of patients and in pT1a-b tumors. (3) Results: Of 1771 HER2-positive BC patients included, pN status distributed as follows: 1047 pN0 (59.1%), 60 pN0(i+) (3.4%), 118 pN1mi (6.7%), and 546 pN1 macro-metastases (30.8%). pN status was significantly associated with sentinel lymph node biopsy, axillary lymph node dissection, age, ER status, tumor grade, and size, lymphovascular invasion, adjuvant systemic therapy (ACt), and radiation therapy. With 61 months median follow-up (mean 63.2; CI 95% 61.5–64.9), only pN1 with macro-metastases was independently associated with a negative impact on overall, disease-free, recurrence-free, and metastasis-free survivals in multivariate analysis. In the pT1a-b subgroup including 474 patients, RFS was significantly decreased in multivariate analysis for pT1b BC without ACt (HR 2.365, 1.04–5.36, p = 0.039) and for pN0(i+)/pN1mi patients (HR 2.518, 1.03–6.14, p = 0.042). (4) Conclusions: Survival outcomes were not adversely affected by pN0(i+) and pN1mi in patients with HER2-positive BC. However, in the case of pT1a-b HER2-positive BC, a negative impact on RFS was observed specifically for patients with pN0(i+) and pN1mi diseases, particularly among those with pT1b tumors without ACt. Our findings highlight the importance of considering the pN0(i+) and pN1mi status in the decision-making process when discussing trastuzumab-based ACt for these patients.


Introduction
The prognostic value of axillary lymph node invasion by isolated tumor cells (ITC) or micro-metastases has been the subject of numerous studies with divergent results but without any distinction between breast cancer (BC) molecular subtypes [1].Since the introduction of sentinel lymph node biopsy (SLNB), ITC and micro-metastases have been detected more often in patients with BC.This limited metastatic lymph node involvement is observed in 8-10% of patients with early BC and sentinel lymph node biopsy (SLNB), representing 10-28% of patients with involved sentinel node .Immuno-histo chemistry (IHC) analysis increased the SN involvement rate from 9% to 47% when compared with HES only [30].However, different rates of LN involvement according to tumor subtypes were reported with lower rates in triple-negative BC and higher rates in HER2-positive BC [31][32][33].The presence of ITC or micro-metastases in the axillary lymph nodes of triplenegative cancers has an independent negative prognostic value, particularly in association with the presence of LVI in patients treated by up-front surgery [34,35].Conversely, no independent negative prognostic value has been shown for the presence of ITC or micrometastases in the axillary lymph nodes of endocrine receptor (ER)-positive HER2-negative cancers [36].The prognostic value of ITC or micro-metastatic axillary lymph node involvement in HER2-positive BCs treated by primary surgery has not been specifically studied.HER2-positive BCs larger than 2 cm or with clinical and/or ultrasound involvement of the axillary lymph nodes are currently treated with neoadjuvant chemotherapy [37], except where contraindicated by age, comorbidities, and, in particular, physiological age.However, as the rate of complete pathological response does not correlate with the initial clinical size of the tumor, neoadjuvant chemotherapy is increasingly proposed for tumors with no clinical axillary lymph node involvement of more than 15 mm, or even more than 10 mm [38].This study aimed to determine, from a multicenter cohort, the prognostic value of axillary node invasion by ITC or micro-metastases in HER2-positive BCs treated by primary surgery for all patients, as well as for patients with pT1a-b cancer.

Study Design and Data Source
The medical records of 23,650 patients that were treated from January 1991 to December 2013 were retrieved from the clinical databases of 13 cancer centers in France for retrospective analysis.Of this initial cohort, all patients treated with primary surgery for HER2-positive BC, with or without adjuvant chemotherapy and trastuzumab, who had undergone breast conservative surgery (BCS) or mastectomy, were included.Data were collected on patient and tumor characteristics, treatments received, and clinical outcomes.

Pathological Assessment
The determination of ER and HER2 status followed national guidelines, where estrogen and/or progesterone receptor positivity was assessed using IHC with a 10% threshold for ER positivity.HER2 positivity was identified by either a 3+ IHC score or HER2 amplification detected through in situ hybridization.To determine lymphovascular invasion (LVI), trained pathologists examined HES slides and identified the presence of lymphovascular emboli, characterized as tumor cells within an endothelium-lined space in the peritumoral area [39].All sentinel lymph node biopsies were analyzed by serial sections with standard HES after fixation.No intraoperatively analysis were performed.If all the serial sections were negative, an additional IHC analysis was carried out.

Statistical Analysis
Analyses were performed separately for all patients and by ER status, on factors associated with pN status (categorized in four groups as pN0, pN0(i+), pN1mi, and pNmacro-defined as any pN+ greater than 2 mm) according to patient, disease, and clinical characteristics such as age, tumor size, Scarff-Bloom-Richardson (SBR) grade, IHC surrogate of molecular subtypes luminal B-like/HER2-positive and Her-positive/ERnegative, breast and axillary surgery, endocrine therapy (ET), adjuvant chemotherapy (ACt), and radiotherapy.Overall survival (OS) was defined as the time interval from the date of surgery to death or last follow-up; disease-free survival (DFS) was defined as the time interval from the date of surgery to any event (recurrence, metastasis, or death) or last follow-up; recurrence-free survival (RFS) was defined as the time interval from the date of surgery to local, regional, or distant recurrence whichever comes first or last follow-up; metastasis-free survival (MFS) was defined as the time interval from the date of surgery to distant recurrence or death as a first event or last follow-up.Patients lost to follow-up were considered as alive as of the date of last contact.The associations between categorical values were evaluated via χ 2 tests.Factors significantly associated with pN status were determined by binary logistic regression adjusted for all significant variables determined by univariate analysis.Kaplan-Meier method and log-rank tests were employed to analyze survival functions, and multivariate survival analyses were performed using the Cox proportional-hazard regression model adjusted for significant variables.Subsequent analyses focused on patients with pT1a-b pN0-pN0(i+) or pN1mi.A significance level of p ≤ 0.05 was set, and SPSS 16.0 (SPSS Inc., Chicago, IL, USA) was used for all analyses.All procedures performed in this study involving human participants were carried out by French ethical standards and with the 2008 Helsinki Declaration.As this was a retrospective non-interventional study, no formal personal consent was required.Authorization to use the database was obtained from the strategic orientation committee of the Paoli-Calmettes Institute (ClinicalTrials.govNCT02869607).
In logistic regression analysis, pN status was significantly associated with pT size ≥ 20 mm (p < 0.0001), ER status (p = 0.005), and LVI (p < 0.0001).Age groups and SBR grade were not statistically associated with pN status.
ACt was statistically significantly associated with SBR grade, ER status, age groups, LVI, and pN status (Supplementary Table S2).The type of surgery did not impact ACt use.In a binary logistic regression analysis, ACt was statistically significantly associated with SBR grade, ER status, age groups, LVI, and pT1.Statistical difference for pN status was not reached but a trend was observed (Table 4).
For patients with pT1a BC, no significant difference was observed between patients stratified according to ACt and pN status.Patients with pN0(i+) or pN1mi had a lower RFS but without a statistical difference (only nine patients in this later group) (Figure 4).

Months from surgery
Recurrence-free survival (%)   For patients with pT1a BC, no significant difference was observed between patients stratified according to ACt and pN status.Patients with pN0(i+) or pN1mi had a lower RFS but without a statistical difference (only nine patients in this later group) (Figure 4).For patients with pT1a BC, no significant difference was observed between patients stratified according to ACt and pN status.Patients with pN0(i+) or pN1mi had a lower RFS but without a statistical difference (only nine patients in this later group) (Figure 4).

Months from surgery
Recurrence-free survival (%)  For patients with pT1b BC, only patients with pN0 and ACt exhibit better RFS (Figure 5): HR 0.140, 0.04-0.48,p = 0.002 in comparison with patients with pN0 without ACt.

Discussion
The main result of this study is the absence of prognostic value of pN0(i+) or pN1mi in HER2-positive BC in the whole cohort, but a negative impact on RFS in pT1a-b patients,

Months from surgery
Recurrence-free survival (%)
In summary, there was no strong difference in ITC and micro-metastases rates and no strong difference in NSN rates, according to tumor subtypes.However, a strong downstaging on NSN involvement rate was reported when completion ALND was performed after ACt, whatever tumor subtype.

Indication of ACt and Trastuzumab in HER2-Positive BC
Axillary micro-metastases do not impact the indication of adjuvant treatment with chemotherapy ± trastuzumab in HER2-positive and triple-negative tumors with pathologic size >2 cm, and even > 10 mm [42,43].Neo-adjuvant chemotherapy is usually administered for patients with pT2 or pT1a-b-c N1 BC [37].ACt after up-front surgery is indicated for patients with ≥ pT1c or pN1 macro-metastases but should also be discussed for patients with pT1a-b pN0 or with ITC or micro-metastases.

Available Literature on HER2-Positive, pT1ab BC
Adjuvant trastuzumab-based chemotherapy clearly improved survival in randomized clinical trials [44][45][46][47], but there were few or no patients with pT1abN0 stages and this specific situation was only addressed in relatively rare retrospective cohort studies [48][49][50][51][52][53].In our study [42], ACt ± trastuzumab was associated with a significantly reduced risk of recurrence, including distant recurrence, in infra-centimetric node-negative HER2-positive BC and most of the benefit may have been driven by pT1b tumors.
In our recent study [39], LVI was present in 24% (4205/17,322) of patients and was most prevalent in patients with luminal B-like tumors, specifically high-grade HER2negative tumors (44%)-22% (3279/14,655) of ER-positive and 35% (926/2667) of ERnegative tumors-(Supplementary Table S4).In the present study, LVI was present in 33.2% (530/1594) of HER2-positive BC-35.2% (189/537) of ER-positive and 32.3% (341/1056) of ER-negative tumors In the subgroup of pT1a-b BC, LVI was present in 10.5% (36/344) of pN0 BC and 31.1% (14/45) of pN0(i+) and pN1mi BC.The presence of LVI was statistically significantly associated with worse OS, DFS, and MFS in all patients, both with and without ACt and both for ER-positive and ER-negative tumors [39].These findings are consistent with those of other recent studies that investigated the prognostic significance of LVI [46][47][48][49][50][51][52][53][54][55][56][57].Moreover, Gujam et al., in a systematic review concluded that LVI was a powerful prognostic factor of poorer survival, whose impact was mainly seen in patients with node-negative BC [58].The significance of LVI as an independent prognostic factor for patients with triple-negative tumors was also reported [34,35].In the present study, LVI was also significantly associated with a negative impact on OS, DFS, RFS, and MFS in multivariate analysis for all HER2-positive patients.Detecting ITCs and micro-metastases is challenging due to their small size and limited presence in tissue samples.Conventional hematoxylin and eosin (HE) staining may lack the necessary sensitivity to accurately identify these subpopulations.Considering the prognosis significance of ITC and micro-metastases in triple-negative BC, it becomes crucial to employ more sensitive techniques, such as serial sections and IHC examination of axillary lymph nodes, to improve their detection and characterization.This is especially important for small-sized tumors without other criteria warranting ACt administration.Considering the negative impact of ITCs and micro-metastases on prognosis in our cohort, our findings suggest that the need for serial sections and IHC examination of axillary lymph nodes may also apply to HER2-positive pT1a-b breast cancer to assist in the decision-making process regarding ACt.

Limitations
There are several limitations to our study, including its retrospective design, lack of differentiation between the types of chemotherapy used, the need for a larger cohort to analyze the impact of ER-status and LVI on pT1a-b BC, and the inability to evaluate RFS specifically for pN0(i+) and pN1mi due to the limited number of patients in this subgroup.Moreover, we were not able to report the presence of ductal carcinoma in situ (DCIS).The clinical and biological significance of HER2 overexpression in patients with DCIS remains poorly defined, and current practice guidelines do not recommend HER2 testing in DCIS patients.Nevertheless, evidence suggests that HER2-positive DCIS cases may be associated with adverse clinicopathological parameters and increased recurrence rates [59].

Conclusions
Survival outcomes were not adversely affected by pN0(i+) and pN1mi in patients with HER2-positive breast cancer.Adjuvant systemic therapy was commonly used for patients with BC ≥ pT1c.However, in the case of small HER2-positive breast cancer (pT1a-b), a negative impact on RFS was observed specifically for patients with pN0(i+) and pN1mi axillary nodes, particularly among those with pT1b tumors.These findings support the need to consider pN0(i+) and pN1mi status in the decision-making process when discussing trastuzumab-based ACt for patients with pT1b HER2-positive breast cancer.

Figure 2 .
Figure 2. Recurrence-free survival according to tumor size and nodal status in pT1a-b patients.

Figure 2 .
Figure 2. Recurrence-free survival according to tumor size and nodal status in pT1a-b patients.Cancers 2023, 15, x FOR PEER REVIEW 10 of 17

Figure 3 .
Figure 3. Recurrence-free survival adjusted on LVI, ER status, age groups, pN status, and pT size in (A) patients with pT1a-b pN0 or pN0(i+)/pN1mi tumors according to adjuvant chemotherapy, and (B) in patients with pT1a-b tumors according to nodal status.

Figure 3 .
Figure 3. Recurrence-free survival adjusted on LVI, ER status, age groups, pN status, and pT size in (A) patients with pT1a-b pN0 or pN0(i+)/pN1mi tumors according to adjuvant chemotherapy, and (B) in patients with pT1a-b tumors according to nodal status.

Figure 3 .
Figure 3. Recurrence-free survival adjusted on LVI, ER status, age groups, pN status, and pT size in (A) patients with pT1a-b pN0 or pN0(i+)/pN1mi tumors according to adjuvant chemotherapy, and (B) in patients with pT1a-b tumors according to nodal status.

Figure 4 .
Figure 4. Recurrence-free survival according to adjuvant chemotherapy and nodal status in pT1a patients.

Figure 4 .
Figure 4. Recurrence-free survival according to adjuvant chemotherapy and nodal status in pT1a patients.

Figure 5 .
Figure 5. Recurrence-free survival according to adjuvant chemotherapy and nodal status in pT1b patients.

Figure 5 .
Figure 5. Recurrence-free survival according to adjuvant chemotherapy and nodal status in pT1b patients.

4. 4 . 1 .
Value of IHC Testing for ITC and Micro-Metastases in Triple-Negative and HER2-Positive BC

Table 1 .
Characteristics of patients according to pathologic nodal status.

Table 2 .
Survival results in multivariate analyses.

Table 3 .
Characteristics of patients according to pN0 and pN0(i+) or pN1mi status.