Outcomes for Patients With Micro‐Invasive or Small HER2+ Breast Cancer in Northern Sydney Local Health District

ABSTRACT

Aim

This study aimed to review the outcomes for patients with microinvasive and small HER2+ breast cancers, compare the clinicopathological features, and explore the patterns of care.

Methods

In this single‐center retrospective study, patients with HER2+, pT1mic‐bN0M0 disease who underwent surgery in Northern Sydney Local Health District between 2006 and 2022 were selected. The primary outcome was invasive disease‐free survival (iDFS), and secondary outcome was overall survival (OS).

Results

Of the 32 eligible patients, those with pT1b disease had higher histological grades and Ki‐67 indices, with a higher rate of trastuzumab administration. One recurrence and one death were observed in the pT1a group. The 5‐year iDFS and OS were 93% for the pT1mic/a group and 100% for the pT1b group (p = 0.29). Sixty‐six percent of patients received adjuvant trastuzumab. Trastuzumab prescription was associated with larger tumor sizes (p = 0.008) and higher histologic grades (p = 0.023).

Conclusions

Patients with microinvasive or small HER2+ node‐negative breast cancer experience favorable outcomes. The pT1b group had higher histological grades and proliferative indices. Patients with pT1b disease or higher tumor grades were more likely to receive trastuzumab. While trastuzumab is recommended for pT1b disease with strong evidence, its use in pT1a disease should be balanced against its unclear benefits and known toxicities.

This retrospective study of patients with small, node‐negative HER2+ breast cancer found favorable 5‐year invasive disease‐free survival rates—93% for pT1mic/a disease and 100% for pT1b disease. Trastuzumab was administered to 66% of patients and was associated with larger tumor size and higher histologic grade. While trastuzumab is strongly recommended for pT1b disease, its role in pT1a tumors remains uncertain, requiring careful consideration of benefits versus toxicity.

1. Introduction

Small breast carcinomas are increasingly seen by clinicians due to the widespread adoption of breast screening and improvements in screening techniques [1].

Previous studies have revealed a favorable long‐term prognosis for patients with small breast cancers [2, 3]. However, HER2‐positivity (HER2+) is associated with an increased risk of recurrence and reduced survival in patients with small node‐negative tumors [4, 5, 6, 7, 8, 9, 10].

Trastuzumab has been shown to be effective in improving outcomes for HER2+ breast cancer, based on evidence from several large clinical trials, including NSABP B31, NCCTG N9831, and HERA [11, 12, 13]. However, these trials excluded patients with tumors smaller than 10 mm without lymph node involvement. The NCCN guidelines [14] recommend treatment with trastuzumab for tumors larger than 6 mm, but for tumors smaller than 6 mm, there are no clear recommendations from National Guideline Committees, due to limited available evidence.

Emerging data on outcomes for patients with small HER2+ breast cancers revealed significant differences across different cohorts [15, 16, 17], as well as substantial variability in the effect of trastuzumab among these patients. Although some prospective trials included patients with small, node‐negative tumors, most enrolled patients had larger tumor sizes, more advanced nodal status, and high‐risk features [18, 19, 20, 21, 22]. Some meta‐analyses have concluded that trastuzumab significantly reduces the risk of recurrence in patients with early‐stage, HER2+ breast cancer [23, 24]. In contrast, some studies suggest that trastuzumab may only provide benefits in selected subgroups of patients, and that overtreatment of low‐risk patients could cause harm [25, 26, 27].

For HER2+ tumors smaller than 6 mm, emerging evidence suggests that the decision to prescribe trastuzumab should consider additional clinicopathological features [8, 17, 28]. In several studies, adverse clinicopathological features, such as higher histologic grades and hormone receptor (HR) negativity, were associated with trastuzumab prescription [24, 26, 29, 30, 31], but prescription patterns vary across centers.

Existing data have indicated that microinvasive and small HER2+ breast cancers have a less favorable prognosis than their HER2− counterparts. However, it remains unclear whether these patients warrant anti‐HER2 therapy. This study aimed to address these issues by reviewing the outcomes for patients with microinvasive and small (up to 10 mm), HER2+ node‐negative breast cancers, comparing clinicopathological factors across different tumor sizes, and evaluating the patterns of care for these patients to determine whether current clinical practice needs to be changed.

2. Materials and Methods

2.1. Study Design

This retrospective study included patients with small, HER2+ node‐negative breast cancers who underwent surgery in the Northern Sydney Local Health District (NSLHD) between January 2006 and December 2022.

2.2. Participants

The inclusion criteria were as follows: (1) patients with breast tumor sizes ≤ 10 mm (pT1mic‐b). (2) node‐negative (pN0) or with isolated tumor cells (pN0i+) with no distant metastasis (M0). (3) HER2+ status confirmed through immunohistochemistry and in situ hybridization in accordance with the American Society of Clinical Oncology (ASCO)/College of American Pathologists guidelines. Patients who received any neoadjuvant therapy or were simultaneously diagnosed with cancer in the contralateral breast were excluded from this study. Eligible patients were identified from the pathology database at the Royal North Shore Hospital (New South Wales Health Pathology). Clinical information was collected from eligible patients’ electronic medical records.

2.3. Data Analysis

The primary outcome was invasive disease‐free survival (iDFS), measured from the time of surgery to the occurrence of any of the following events: invasive local, regional, and distant recurrences; contralateral breast cancer; second non‐breast primary cancer; or death from any cause. The secondary outcome was overall survival (OS), defined as the time from surgery to death from any cause. Both outcome measures were defined according to the Standardized Definitions for Efficacy End Points (STEEP) version 2 [32].

Descriptive statistics were used to summarize baseline characteristics. Clinicopathological features between subgroups were compared using Pearson's chi‐squared test (χ2) or Fisher's exact test for categorical variables, and Student's t‐test for continuous variables. A univariate linear regression was used to compare pathological factors with the frequency of trastuzumab prescription. The Kaplan–Meier method with log‐rank tests was used to estimate iDFS and OS, and to compare outcomes between different tumor sizes and between different anti‐HER2 treatment status. Due to the small sample size and low number of events, univariate and multivariate Cox regression analyses were not performed in the outcome analysis, and confidence intervals for iDFS and OS were not included. p values ≤ 0.05 were considered statistically significant for all statistical tests. All analyses were performed using IBM SPSS Statistics 29.0.

3. Results

3.1. Patient Characteristics

Thirty‐two patients were included in this study with a median age at diagnosis of 60 years. The majority of patients were post‐menopausal (62%) and had other medical comorbidities (72%), including previous breast cancer, which was identified in two patients (6%). Two patients had pT1mic tumors, 13 patients had pT1atumors, and 17 patients had pT1b tumors. While all patients were node‐negative (pN0)—a requirement for inclusion in this study—four patients (12%) had isolated tumor cells in axillary lymph nodes. Twenty‐one patients (66%) had HR+ disease. Proliferative indices (Ki‐67 index) of the tumors ranged from 1% to 90%, with a median value of 33%. Lymphovascular invasion (LVI) was identified in four patients (12%). Twelve patients (37%) had multifocal disease. Compared to pT1mic/a tumors, pT1b tumors had significantly higher histologic grades (p = 0.017) and Ki67 (p = 0.030). Detailed patient characteristics are summarized in Table 1.

TABLE 1.

Baseline characteristics of the patients.

Characteristics	pT1mic + pT1a		pT1b		Overall
	N = 15	%	N = 17	%	N = 32	%	p value
Age at diagnosis (years) (Mean ± SD)	63	13	58	14	60	14	0.26
Menopausal status at diagnosis							0.23
Pre‐menopausal	4	27	8	47	12	38
Post‐menopausal	11	73	9	53	20	62
Number of comorbidities							0.62
0	3	20	6	35	9	28
1–3	5	33	5	30	10	31
3+	7	47	6	35	13	41
Previous breast cancer							0.99
No	13	86	15	88	28	88
Yes	1	7	1	6	2	6
Unknown	1	7	1	6	2	6
Nodal status							0.32
pN0(i−)	12	80	16	94	28	88
pN0(i+)	3	20	1	6	4	12
Histological type							0.39
Ductal	14	93	14	82	28	88
Lobular	1	7	1	6	2	6
Other	0	0	2	12	2	6
Grading							0.013
Grade 1	3	20	0	0	3	9
Grade 2	8	53	5	29	13	41
Grade 3	3	20	11	65	14	44
Unknown	1	7	1	6	2	6
IHC3 subtype							0.39
ER+ PR+	9	60	7	41	16	50
ER+ PR−	2	13	1	6	3	9
ER− PR+	0	0	2	12	2	6
ER− PR−	4	27	7	41	11	34
Lymphovascular invasion							0.60
No	14	93	14	82	28	88
Yes	1	7	3	18	4	12
Ki‐67 index							0.030
Low < 20%	6	40	2	12	8	25
High > 20%	5	33	15	88	20	63
Unknown	4	27	0	0	4	12
Number of foci							0.31
Single	8	53	12	71	20	63
Multiple	7	47	5	29	12	37

Note: Bold values are highlighting variables that have obtained statistical significance in the statistical analysis.

Abbreviations: ER, estrogen receptor; IHC, immunohistochemical subtype; pN0(i−), No regional lymph node involvement; pN0(i+), isolated tumor cells in regional lymph node; PR, progesterone receptor; SD, standard deviation.

3.2. Treatment Overview

Regarding treatment (Table 2), 15 patients (47%) underwent breast‐conserving surgery, while 17 patients (53%) had mastectomies. The majority (97%) underwent sentinel lymph node biopsy without axillary dissection. Thirteen patients (41%) were treated with radiation therapy. Endocrine therapy was administered to most patients with HR+ disease (86%). In these patients, endocrine therapy options included tamoxifen only (50%), tamoxifen with a GnRH inhibitor (6%), and aromatase inhibitor only (44%). Adjuvant chemotherapy was administered to 21 patients (66%), with details of the regimens shown in Table 2. Trastuzumab was administered to 21 patients (66%): one in pT1mic group, five in pT1a group, and 15 in pT1b group. The pT1b group had a significantly higher rate of trastuzumab administration compared to the pT1mic/a group (88% vs. 40%; p = 0.004). Notably, one patient in the pT1a group ceased trastuzumab prematurely due to associated cardiac side effects. On univariate regression analysis, larger tumor sizes and higher histological grades were the only predictors for trastuzumab prescription, as shown in Table 3. No statistically significant differences in other treatment modalities, recurrence, or death were found between different tumor size groups.

TABLE 2.

Overview on disease management and outcome.

Characteristics	pT1mic + pT1a		pT1b		Overall
	N = 15	%	N = 17	%	N = 32	%	p value
Type of surgery							0.49
Breast conserving surgery	8	53	7	41	15	47
Mastectomy	7	47	10	59	17	53
Lymph node procedure							0.47
Sentinel lymph node biopsy	14	93	17	100	31	97
Sentinel lymph node biopsy + axillary clearance	1	7	0	0	1	3
Radiation therapy							0.76
Yes	7	47	6	35	13	41
No	7	47	9	53	16	50
Unknown	1	6	2	12	3	9
Endocrine therapy							0.44
Tamoxifen only	6	40	3	18	9	28
Tamoxifen + GnRH inhibitor	0	0	1	6	1	3
Aromatase inhibitor only	3	20	5	29	8	25
No endocrine therapy	6	40	8	47	14	44
Chemotherapy							0.16
Adjuvant paclitaxel	3	20	7	41	10	31
Adjuvant carboplatin	1	7	2	12	3	9
Adjuvant docetaxel	0	0	1	6	1	3
Adjuvant docetaxel + carboplatin	1	7	3	18	4	13
Adjuvant doxorubicin + cyclophosphamide, followed by paclitaxel	1	7	1	6	2	6
No chemotherapy	9	59	2	12	11	34
Unknown scheme	0	0	1	6	1	3
Trastuzumab therapy							0.004
Yes	6	40	15	88	21	66
No	9	60	2	12	11	34
Recurrence							0.47
No evidence of recurrence	14	93	17	100	31	97
Local recurrence	1	7	0	0	1	3
Distant recurrence	0	0	0	0	0	0
Death							0.47
Yes	1	7	0	0	1	3
No	14	93	17	100	31	97
Cause of death
Breast cancer‐related	0	0	0	0	0	0
Other causes	1	100	0	0	1	100

Note: Bold values are highlighting variables that have obtained statistical significance in the statistical analysis.

Abbreviation: GnRH, gonadotropin‐releasing hormone.

TABLE 3.

Clinicopathological features of patients who received trastuzumab versus patients who did not receive trastuzumab.

Characteristics	Trastuzumab group		Non‐trastuzumab group
	N = 21	%	N = 11	%	p value
Age at diagnosis (years) (mean ± SD)	57	12	66	15	0.08
Menopausal status at diagnosis					0.39
Pre‐menopausal	9	43	3	27
Post‐menopausal	12	57	8	73
Previous breast disease					0.99
Yes	2	10	0	0
No	17	80	11	100
Unknown	2	10	0	0
Number of Comorbidities					0.46
0	7	33	2	18
1–3	6	29	4	36
3+	8	38	5	46
Cardiac Comorbidities					0.60
Yes	4	19	3	27
No	17	81	8	73
Tumor size					0.008
pT1mic/a: ≤ 5 mm	6	39	9	82
pT1b: > 5 to ≤ 10 mm	15	71	2	18
Nodal status [2]					0.99
pN0(i−)	17	81	11	100
pN0(i+)	4	19	0	0
Grading					0.023
Grade 1	0	0	3	27
Grade 2	9	43	4	36
Grade 3	12	57	2	18
Unknown	0	0	2	18
IHC3 subtype					0.34
HR+ HER2+	15	71	6	55
HR− HER2+	6	29	5	45
Lymphovascular invasion					0.99
No	17	81	11	100
Yes	4	19	0	0
Ki67					0.11
Low < 20%	3	14	5	46
High > 20%	17	81	3	27
Unknown	1	5	3	27
Number of foci					0.39
Single	12	57	8	73
Multiple	9	43	3	27
Type of surgery					0.53
Breast conserving surgery	9	43	6	55
Mastectomy	12	57	5	45
Radiation therapy					0.41
Yes	7	33	6	55
No	11	52	5	45
Unknown	3	14	0	0
Endocrine therapy					0.38
Yes	13	62	5	45
No	8	38	6	55

Note: Bold values are highlighting variables that have obtained statistical significance in the statistical analysis.

Abbreviations: HR, hormone receptor; IHC, immunohistochemical subtype; pN0(i−), No regional lymph node involvement; pN0(i+), isolated tumor cells in regional lymph node; SD, standard deviation.

3.3. Outcome Analysis

The median follow‐up time for the cohort was 62 months (IQR 37–89 months). One recurrence in the contralateral breast was observed in the pT1a group, where the patient had not received trastuzumab; the recurrence was detected 9.6 years after the initial diagnosis. Additionally, one death unrelated to breast cancer occurred in the pT1a group, 4 months after diagnosis. This patient was treated with adjuvant docetaxel, carboplatin, and trastuzumab for only 2 months without known cardiac side effects.

When analyzing by tumor size, both the 3‐ and 5‐year iDFS were 93% for the pT1mic/a group, and 100% for the pT1b group, with no statistically significant difference (p = 0.24). Similarly, the 3‐ and 5‐year OS were 93% for the pT1mic/a group and 100% for the pT1b group, with no significant difference (p = 0.29). Kaplan–Meier curves and confidence intervals are not included, as the small sample size and low number of events make interpretation difficult.

An exploratory analysis was performed to compare the outcomes between patients who received trastuzumab and those who did not. Both 3‐ and 5‐year iDFS were 95% for the trastuzumab group, and 100% for the non‐trastuzumab group, with no significant difference (p = 0.52). OS analysis was not performed, as the only death from non‐breast cancer cause occurred early after diagnosis, with the patient having received only 2 months of trastuzumab.

4. Discussion

This study found that patients with small (< 10 mm) HER2+ breast cancer had an excellent prognosis, regardless of tumor size. Patients with pT1b tumors had higher histologic grades and proliferative indices and were more frequently prescribed trastuzumab, compared to those with pT1mic/a disease. The decision to prescribe trastuzumab was predicted by larger tumor sizes and higher histologic grades.

Despite a higher risk of recurrence, emerging data have shown that patients with HER2+ T1N0 disease still have favorable outcomes [2, 15, 16, 17]. In our study, the 5‐year iDFS and OS in the pT1mic/a and pT1b subgroups were 93% and 100%, respectively. Similar favorable results have been reported in other retrospective studies, regardless of whether patients received systemic treatment. For patients with tumors smaller than 10 mm, the 5‐year DFS ranged from 84% to 100%, and OS ranged from 92% to 100% [2, 4, 15, 16, 33]. The variations could be attributed to non‐standardized definitions of outcome measures, as some studies excluded contralateral primaries and deaths from other causes [16]. A prospective study by Vaz‐Luis et al. [2] using data from the NCCN database showed a 5‐year iDFS of 68% in the HR− T1b subgroup. However, this difference may be due to chance, given the small sample size in this subgroup. The two patients in the pT1b group who did not receive trastuzumab were older (77 and 87 years) and had cardiovascular risk factors or a history of cardiovascular events. They were also HR+ and had endocrine therapy as an alternative option for systemic therapy.

The favorable outcomes of small HER2+ tumors raise questions about appropriate and justified treatment strategies. In our study, 21 patients (66%) with T1mic/a/b tumors received trastuzumab, similar to 61% in a single‐center study in China [29], but higher than the 50% rate observed in a Scottish multicenter study [30]. In the pT1b subgroup, 88% of our patients received trastuzumab, which is higher than a study using ASCO CancerLinQ database, where the rate was 58% [27], and 63% in a Canadian multicenter study [34]. The pT1b subgroup in our cohort had excellent outcomes, with no recurrences or deaths. Large retrospective studies support anti‐HER2‐containing adjuvant systemic therapy for pT1b tumors. In a propensity‐matched retrospective study by Parsons et al. using the National Cancer Database (NCDB), patients treated with systemic therapy containing anti‐HER2 agents had a higher 5‐year OS compared to untreated patients (97.1% vs. 92.3%; p = 0.0016) [35]. However, due to outdated terminology, it was unclear whether this benefit was driven by anti‐HER2 therapy or chemotherapy. In the ASCO CancerLinQ database study, the univariate hazard ratio for iDFS between the trastuzumab (with or without chemotherapy) group and the locoregional therapy alone group was 0.64 (95% CI 0.4–1.01; p = 0.05) [27]. Furthermore, the APT trial provides strong supportive evidence for the use of a paclitaxel and trastuzumab regimen in pT1b disease, demonstrating excellent long‐term outcomes with low recurrence rates [20, 21, 22].

While the benefits of trastuzumab in the pT1b cohort have been clearly demonstrated, its role in pT1mic/a tumors remains controversial, with wide variations in prescription patterns. In our study cohort, 40% of patients with pT1mic/a disease received trastuzumab, similar to the 49% in the ASCO CancerLinQ database study. In contrast, only 9% of patients were treated with trastuzumab in the Canadian multicenter study [34]. In another multicenter Italian study, 23% of patients received trastuzumab [17], which could be explained by the higher proportion of pT1mic disease (58% vs. 6% in our cohort). Both recurrence events in our study occurred in the pT1a subgroup, suggesting that anti‐HER2 therapy could be beneficial. Supporting this, the Italian study found that trastuzumab reduced relapse risk by 85% (p = 0.048) [17]. However, other studies have shown mixed results. Another NCDB study by Cao et al. found no significant OS improvement in the pT1a group with anti‐HER2‐containing adjuvant systemic therapy [26]. Similarly, our study showed no significant difference in iDFS with trastuzumab. Furthermore, the ASCO CancerLinQ database study reported a trend toward worse outcomes when comparing trastuzumab with or without chemotherapy to untreated patients in the pT1a subgroup (5‐year OS: 95.4% vs. 96.9%; p = 0.059) and the pT1mic subgroup (89.1% vs. 99.1%; p = 0.0006) [27].

Despite its favorable toxicity profile, trastuzumab is associated with cardiotoxicity, which is particularly relevant in this patient group, given their good long‐term outcomes and high cure rates [11, 12]. In the APT trial, 12% of participants discontinued treatment due to adverse effects; 3.2% experienced an asymptomatic decline in left ventricular ejection fraction, and 0.5% developed symptomatic heart failure [20]. The impact of cardiotoxicity was also observed in our study, where one patient with pT1a disease discontinued trastuzumab due to a drop in left ventricular ejection fraction. Therefore, the risks of cardiotoxicity must be balanced against the uncertain benefits of treatment [14]. Previous studies have explored less‐toxic therapy options. The ATEMPT trial investigated the use of trastuzumab emtansine (T‐DM1) in patients with T1N0 disease. In this trial, 17.5% of patients discontinued treatment due to unacceptable toxicities, compared to 6% in the paclitaxel plus trastuzumab group [36]. Future studies should explore regimens with lower toxicity and possible therapy de‐escalation for vulnerable patients to improve the risk/benefit ratio.

Although the benefit of trastuzumab in pT1mic/a disease remains unclear, patients with certain clinicopathological features may derive more benefits. HER2+ breast cancer is biologically heterogeneous, with varying intrinsic molecular subtypes, DNA alterations, and tumor microenvironment compositions [37], which may lead to different clinical outcomes and influence therapy choices. In our study, patients with higher tumor grades were more likely to receive trastuzumab, consistent with the findings from the NCDB study by Cao et al. [26], a French multicenter study by Houvenaeghel et al. [31], and a study by Pawloy et al. [30], which included T1c tumors. Other studies have identified predictors for trastuzumab prescription, such as younger age, the presence of LVI, ER‐negativity, a higher Ki‐67 index, and multifocality [9, 26, 28, 29, 30, 31]. Apart from tumor size and lymph node status, intrinsic factors such as HR‐negativity, the presence of LVI, multifocality, tumor‐infiltrating lymphocytes (TILs), CEP17 duplication, and TOP2A aberrations have been explored for prognostic or predicting therapeutic effects in HER2+ breast cancer [4, 17, 25, 38, 39, 40]. This study could not assess novel prognostic factors, such as TILs, as they were unavailable in most histopathological reports. More accurate tools are needed to stratify the risk in this patient population, likely based on a combination of prognostic factors, to identify subgroups that may benefit more from treatment.

A major limitation of our study is its retrospective design, leading to selection bias in trastuzumab prescription and unbalanced comparisons that may affect the observed clinical outcomes. Additionally, the small sample size and limited events make our conclusions more exploratory.

Our findings show that patients with microinvasive or small HER2+ breast cancer have a favorable prognosis. While trastuzumab is strongly recommended for pT1b disease, its use in pT1mic/a disease must be weighed against its uncertain benefits and potential side effects, in line with current international guidelines [14, 41]. Our study adds to the growing body of evidence regarding clinical outcomes and management in this unique patient group. While prospective studies provide the highest level of evidence, they require a large number of patients, long follow‐ups, and significant financial resources. Large‐scale retrospective studies are more feasible but are limited by inherent biases.

Funding

The authors have nothing to report.

Ethics Statement

This project was approved by the North Sydney Local Health District Human Research Ethics Committee, #2024/ETH00696.

Consent

A waiver of consent was granted by the North Sydney Local Health District Human Research Ethics Committee.

Conflicts of Interest

The authors declare no conflicts of interest.