p95HER2 expression in HER2‐positive breast cancer with primary resistance

p95HER2 isoform is a truncated form of HER2 that retains the C terminal domain but lacks an N terminal trastuzumab binding site. From 2014 to 2016, we assessed the expression of p95HER2 expression in 59 HER2‐positive breast cancer patients from FUSCC. The median follow‐up was 54 months. In our study, 19 patients (32.2%) were p95HER2 positive. p95HER2‐positive expression rate is higher in premenopausal patients than in postmenopausal patients (68.4% vs. 31.6%, P = .026). p95HER2 positive was found more in premenopausal patients and was associated with worse DFS (hazard ratio, 2.21; 95% CI, 1.06–4.61; P = .034), indicating that p95HER2 expression tends to be a more aggressive isoform type of HER2‐positive breast cancer.

defined into several subtypes based on receptor and gene expression profiles.In clinical practice, the classification of breast cancer depends on estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor positive (HER2). 2 HER2 subtype accounted for nearly 20% of all breast cancer subtypes. 3R2 positive subtype is biologically aggressive with dismal outcomes before the introduction of HER2-targeted therapy.The HER2 receptors contain three parts, an extracellular ligand-binding domain, a transmembrane region, and an intracellular tyrosine domain.Ligand binding to HER2 receptors causes homodimerization or heterodimerization, leading to a cascade of intracellular signaling pathways, including the phosphoinositide-30kinase (PI3K)-AKT 4 and mitogen-activated protein kinases (MAPK).5 The humanized monoclonal antibody, trastuzumab binds to the extracellular domain (ECD) of HER2, suppresses the activation of intracellular HER2 signaling pathways, and inhibits cell proliferation.Trastuzumab has become the primary treatment of choice for patients with HER2+ breast cancer, in both early and metastatic stages.Despite the reported efficacy of trastuzumab, a proportion number of patients inevitably develop resistance, leading to disease progression.6 The HER2 receptor may exhibit a series of 90-100 kDa carboxy-terminal fragments, also known as p95HER2.6 Through HER2 shedding activity, there is one active form of p95HER2/CTF, which is likely to be a translation product of the HER2 gene, initiates an AUG codon starting at methionine 611 (611-CTF).In this study, the commonly defined p95HER2 is referred to as 611-HER2-CTF.7 The prognostic value of p95HER2 remains controversial as in previous studies, the positive and inverse association between p95HER2 and trastuzumab therapy expression have been described.9,20 Our study aimed to assess the expression of p95HER2 positive in HER2-positive breast cancer who received trastuzumab therapy and its clinical outcomes.

| Study population
From 2014 to 2016, we retrospectively assessed the expression of p95HER2 expression in 59 HER2-positive breast cancer patients from Fudan University Shanghai Cancer Centre (FUSCC).In this study, we included 29 patients with primary trastuzumab resistance and 30 non-relapse breast cancer patients.Primary trastuzumab resistance was defined as patients who had recurrence or metastasis during the treatment of trastuzumab or 12 months after completion of trastuzumab.All patients included in our study underwent mastectomy or breast-conserving surgery.Subjects who had bilateral and inflammatory breast cancer were excluded from our study.
Clinicopathological and follow-up data of patients were analyzed from a prospectively maintained institutional database.Written informed consent was obtained according to the institutional guidelines and patient information was fully anonymized before data analysis.

| Immunohistochemical analysis
Sections of 4 μm were sliced from FFPE blocks and stained with hematoxylin and eosin (H&E) for light microscopy examination.All of our patients were HER2 positive, with HER2 IHC 3+ or IHC 2+ and in situ hybridization [ISH]-positive.The p95HER2 antibody (87-1) used to analyze p95HER2 expression was provided and optimized by Simcere Pharma.IHC slides were baked at 60 C for an hour, followed by deparaffinization, rehydration, and antigen retrieval.IHC staining was carried out on Leica Biosystems BOND III with 87-1 antibody in 0.8 μg/mL and was incubated for 60 min.Amplification was first using Leica post-primary rabbit anti-mouse IgG and then incubated with anti-rabbit Poly-HRP-IgG.p95HER2 expression was assessed with a 10% threshold for positivity at membranous staining, cytoplasmic staining was excluded. 10HER2-positive staining was defined by either (IHC 3+ or IHC 2+/in situ hybridization +).HER2-positive immunohistochemical analyses were performed by Ventana autostainer (Roche).p95HER2 scoring was visually performed by two pathologists.Tumors with >1% of F I G U R E 1 Expression of p95HER2 in HE staining.Immunohistochemical staining of p95HER2 in HER2-positive breast cancer.(1A,1B) Representative IHC images of positive p95HER2 membranous staining (1A Â100, 1B Â200), (2A,2B) Representative IHC images of negative p95HER2 membranous staining (2A Â100, 2B Â200).
T A B L E 1 Clinicopathological characteristics of patients.

| Statistical analysis
Chi-square analysis or Fisher exact test was used to determine the relationship between p95HER2 expression and patients' characteristics.The main outcome measures of disease-free survival (DFS), distant disease-free survival (DDFS), and overall survival (OS) were measured using a log-rank test and Cox regression for hazard ratio.

| The relationship between p95HER2 expression and survival outcomes
Figure 3 shows the survival outcomes in p95HER2-negative and p95HER2-positive patients.Patients with p95HER2 positive had worse DFS than those with P95HRE2 negative.Tumor stage, nodal stage, and p95HER2 expression were factors related to DFS in univariate analysis (P < .1)and were included in multivariate analysis (Table 3A).Multivariate Cox hazard models showed that p95HER2 expression (P = .032)was the only variable that was found to be predictive of DFS (Figure 2).The 2-year-DFS for p95HER2-negative and p95HER2-positive patients was 62.5% versus 29.5%, respectively (HR = 2.21, log-rank P = .034)(Figure 3A).
Tumor stage, nodal stage, and p95HER2 expression were factors related to DDFS in univariate analysis (P < .1)and were included in multivariate analysis (Table 3B).Multivariate analysis showed that N stage (P = .029)significantly predicted DDFS while p95HER2 expression (P = .066)did not reach a statistical difference (Figure 2).Patients with p95HER2 positive tend to have worse DDFS than those with P95HRE2 negative with a marginal difference.The 2-year-DDFS for p95HER2-negative and p95HER2-positive patients was 74.8% versus 47.8%, respectively (HR = 2.10, log-rank P = .050)(Figure 3B).
A total of 11 patients died during follow-up.Univariate analysis showed that T-stage (P < .1)was associated with OS.The 2-year-OS for p95HER2-negative and p95HER2-positive patients was 97.4% versus 95.7%, respectively (HR = 1.06, log-rank P = .925)and did not reach a statistical difference.

| DISCUSSION
Subsequent clinical studies found that p95HER2 expression was related to poor prognostic markers, constituting an aggressive subtype of HER2-positive cancers with distinct biological and clinical features. 8,11,12In our study, 32.2% of patients are p95HER2 positive.
4][15] Results from our study found that p95HER2 expression was higher in premenopausal women.7][18] The role of p95HER2 in trastuzumab resistance remains inconclusive.In our study, we investigated the association between p95HER2 expression and primary trastuzumab resistance.Primary trastuzumab resistance was defined as patients who had recurrence or metastasis during the treatment of trastuzumab or within 12 months after the completion of trastuzumab.When comparing primary trastuzumab resistance and non-relapse patients, we observed that patients with positive p95HER2 expression had worse disease-free survival and distantdisease-free survival than those with negative expression.Based on the findings of our study, we support that p95HER2 expression was associated with primary resistance.Scaltriti et al. also suggested that p95HER2 is primarily resistant to trastuzumab but retains sensitivity to tyrosine kinase inhibitors. 19The authors found that only one out of nine (11.1%) patients with tumors expressing p95HER2 responded to trastuzumab, while 19 of 37 patients with tumors expressing fulllength HER2 responded to trastuzumab.However, Sperinde et al.
found that high expression of p95HER2 is correlated with increased benefit from trastuzumab. 20The differing conclusions between Sperinde et al. and our study may be attributed to the different time intervals over which patients underwent trastuzumab therapy.In their study, patients received 9 weeks of trastuzumab therapy, whereas, in our cohort, patients received 1 year of trastuzumab therapy with adjuvant chemotherapy, as recommended by the NCCN guidelines. 215HER2 expression was first identified by using Western blotting and immunoprecipitation by Christianson et al., 17 who reported that it was associated with positive lymph nodes.p95HER2 is highly oncogenic as it could form homodimers spontaneously, leading to cell proliferation.Results from experimental models have shown that p95HER2 extensively induces the expression of genes that drive the development of malignancy and metastasis, such as MMP1, IL11, and MET. 22r results show that p95HER2-positive patients have a higher recurrence or metastasis rate than negative patients in those who received trastuzumab therapy.The recurrence and metastasis rate in p95HER2 positive is significantly higher than in p95HER2-negative patients (P = .014).However, p95HER2 expression did not show significance in predicting OS in our cohort.This may be due to the short follow-up period as breast cancer patients have relatively long survival time.Most of the patients included in our study were pT1-2 and pN0-2, 23.5% of patients were pN3, and a longer follow-up time was Multivariate Cox regression analyses for (A) disease-free survival and (B) distant disease-free survival.
needed to evaluate the differences in DDFS and OS between patients.p95HER2, the truncated extracellular domain of HER2, has been implicated in causing resistance to agents binding to the extracellular domain of HER2 such as trastuzumab and pertuzumab.
Although p95HER2 lacks trastuzumab epitope, it remains functionally active in the expression of HER2.p95HER2 expression tumors would preserve tyrosine kinase activity and remain HER2-dependent for the activation of PI3K/AKT and MAPK/ERK signaling pathways.In contrast, tyrosine kinase inhibitors (TKIs) such as lapatinib and neratinib bind irreversibly to cysteine residues within the intracellular ATPbinding region of the receptor.Therefore, it was expected that p95HER2 expression tumors retain sensitivity to TKIs. 14NSABP FB-10 trial reported that two patients who were resistant to HER2-targeted therapy with high p95HER2 expression achieved complete response after receiving T-DM1 and neratinib. 23The HERLAP trial found that higher H2T/p95HER2 ratio patients have longer persistence to protocol period in HER2-positive metastatic breast cancer patients. 24This indicates that gene expression status and H2T/p95HER2 expression may identify the group of patients that benefit from trastuzumab or lapatinib for patients with metastasis. 25 p95HER2 is a tumor-specific antigen, researchers generated p95HER2 chimeric antigen receptors (CARs) T cells to target p95HER2-expressing tumors.Results showed that p95HER2 CAR T cells are effective against primary and metastases tumors in murine F I G U R E 3 Kaplan Meier survival curves for (A) disease-free survival, (B) distant disease-free survival, and (C) overall survival according to p95HER2 expression in HER2-positive patients.
T A B L E 3 A Univariable Cox regression analysis for disease-free survival.

P-value
Hazard ratio (95% CI) models. 26Ruiz et al. developed a p95HER2-T cell bispecific antibody (TCB), which binds to both T cell receptor and p95HER2 antigen and showed a potent antitumor effect on p95HER2 expressing breast primary and brain metastases. 27
These strategies showed efficacy at the experimental level, highlighting the potential of targeting p95HER2 in antitumor therapy.More research and preclinical trials are needed to Chih Wan Goh contributed to the collection of data, performed statistical analysis, took the lead in writing the manuscript; Yayun Chi was involved in the conceptualization and design of the study; Benlong Yang provided feedback and parcipated in manuscript revisions; Jiong Wu oversaw the investigation, provided guidance and revised the manuscript.Univariable Cox regression analysis for distant disease-free survival.