- Email: [email protected]
Evaluation of serum HER2 extracellular domain in early breast cancer patients: correlation with clinicopathological parameters and survival
original article
Annals of Oncology 19: 883–890, 2008 doi:10.1093/annonc/mdm585 Published online 10 January 2008
Evaluation of serum HER2 extracellular domain in early breast cancer patients: correlation with clinicopathological parameters and survival V. Ludovini1 *, S. Gori1 , M. Colozza1, L. Pistola1, E. Rulli2, I. Floriani2, E. Pacifico3, F.R. Tofanetti1, A. Sidoni4, C. Basurto1, A. Rulli5 & L. Crino`1 1
Medical Oncology Division, Azienda Ospedaliera of Perugia, Perugia; 2Department of Oncology, Istituto di Ricerche Farmacologiche ‘Mario Negri’, Milan; Biochemistry and Clinical Chemistry Laboratory, Azienda Ospedaliera of Perugia, Perugia; 4Institute of Pathology; 5‘Breast Unit’, Department of Surgical, University of Perugia, Perugia, Italy 3
Received 2 October 2007; revised 1 December 2007; accepted 3 December 2007
extracellular domain (ECD) and tissue HER2 status, their relationship with clinicopathological parameters and their impact on disease-free survival (DFS) and overall survival in early breast cancer patients. Patients and methods: This prospective trial included patients with stage I–III breast cancer. Serum HER2 ECD levels were measured by two enzyme-linked immunosorbent assays before surgical treatment. Tissue HER2 status was analyzed by immunohistochemistry (IHC) in all tumors; FISH assay was utilized in HER2 2+ tumors by IHC. Results: From May 2000 to July 2005, 256 consecutive stage I–III breast cancer patients were included in this study. High serum HER2 ECD levels (‡15 ng/ml) were reported in 23 patients (9.0%) and HER2-positive status in tumor tissue was observed in 42 patients (16.4%) with a concordance of 87.1%. High HER2 ECD levels were significantly associated with high histological grade (P = 0.003), stage III (P = 0.008), lymph node involvment (P = 0.035) and negativity of both estrogen (P = 0.016) and progesterone (P = 0.007) receptors. At multivariate analysis, high serum HER2 ECD levels were a significant independent prognostic factor of worse DFS (P = 0.009). Conclusions: A statistically significant association was observed between high serum HER2 ECD levels and worse DFS in early breast cancer patients. Key words: early breast cancer, HER2 ECD, prognosis
introduction The human epidermal growth factor receptor-2 (HER2) proto-oncogene, also known as HER2/neu and c-erbB-2, encodes a growth factor receptor that has been found to play an important role in breast cancer [1]. In up to 30% of breast cancer patients, the HER2 gene is amplified and its associated receptor protein is overexpressed on the tumor cell surface, thus playing an important role in the malignant transformation and clinical aggressiveness of breast cancer. HER2 overexpression is an independent predictor of shorter overall survival (OS) and disease-free survival (DFS) in invasive primary breast cancer [2]. Retrospective evaluations of randomized clinical trials have revealed a correlation between HER2 overexpression and *Correspondence to: Dr V. Ludovini, Medical Oncology Division, Azienda Ospedaliera of Perugia, via Brunamonti 51, 06122 Perugia, Italy. Tel: +39-075-5783453; Fax: +39-075-5729946; E-mail: [email protected]
These authors contributed equally to this work.
resistance to tamoxifen [3–5] while an interaction between HER2 overexpression and enhanced response to doxorubicincontaining regimens has been reported [6–7]. A variety of methods are available to assess tissue HER2 status: immunohistochemistry (IHC), which detects protein overexpression and FISH, which detects HER2 gene amplification are the most widely applied techniques. Tissue HER2 status evaluation is very important because HER2 protein provides a target for therapeutic approaches. Trastuzumab, a mAb (Herceptin; Genentech Inc., South San Francisco, CA) alone or in combination with chemotherapy has been shown to be an effective treatment for HER2-positive breast cancer (IHC 3+ or FISH amplified) in metastatic, neo-adjuvant and adjuvant settings [8]. HER2 receptor is composed of a cytoplasmic domain with tyrosine kinase activity, a transmembrane domain and an extracellular domain (ECD). The HER2 ECD may be cleaved and shed from the surface of breast cancer cells and serum HER2 ECD levels can be detected by enzyme-linked immunosorbent assays (ELISAs) without any significant
ª The Author 2008. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected]
original article
Background: We explored the correlation between serum human epidermal growth factor receptor-2 (HER2)
original article cross-reactivity with other members of the HER receptor family [9]. The manual and automated ELISA testing methods of serum HER2 ECD are standardized methodologies and are not prone to subjective variable interpretation by a reviewer. Shed HER2 ECD has been detected in the serum from 20% to 40% of patients with metastatic disease [10–11] and high serum concentrations of HER2 ECD have been associated with HER2 overexpression [12], increased tumor burden [13], worse survival [14] and resistance to endocrine therapy and chemotherapy [15–17]. On the other hand, HER2 ECD seems to be associated with sensitivity to HER2-targeted agents, such as trastuzumab. A decrease in serum HER2 ECD levels in metastatic breast cancer patients treated with trastuzumab was predictive of response [18]. In early breast cancer, shed HER2 ECD has been detected up to 20% of patients [19], but the clinical significance of circulating HER2 ECD is uncertain and the relationship of baseline HER2 ECD levels with tissue HER2 overexpression in primary tumors has been little studied. Therefore, we carried out a prospective study in operable breast cancer patients to evaluate the correlation between HER2 ECD serum levels and HER2 tumor status measured by IHC and FISH, their relationship with clinical–pathological parameters and their impact on DFS and OS.
patients and methods This prospective trial was conducted in patients with a histological diagnosis of stage I–III breast cancer treated with conservative surgery or mastectomy. Tumor staging followed the tumor–node–metastasis (TNM)– American Joint Committee on Cancer classification [20] and the pTNM was obtained after classical pathological examination. Patients with metastatic disease and with other previous tumors were excluded from this study. Estrogen receptors (ERs), progesterone receptors (PgRs) and HER2 status were assessed at the time of surgery on formalin-fixed paraffinembedded tissue blocks of the primary tumor in the Pathology Department of the University of Perugia. Recorded clinical and pathological features for each patient include age, menopausal status, histology, grade, Ki67, ER and PgR status, stage, surgical treatment and medical adjuvant therapy. Follow-up, including clinical examination (every 3 months for the first 2 years, every 6 months for the next 3 years and yearly thereafter) and annual Rx-mammography, was carried out in all patients. Recurrence was defined as the first documented evidence of new disease manifestation in the locoregional area, in the contralateral breast, in distant sites, or in a combination of these. The study was reviewed and approved by the institution’s Ethics Committee and informed consent was obtained by all patients.
serum HER2 ECD assays Serum samples were prospectively collected from breast cancer patients before surgery. Five milliliters of peripheral blood were collected in a sterile test tube (without anticoagulants) and centrifuged at 3000 g for 10 min at room temperature. Serum was stocked in 0.5-ml aliquots in cryovials and stored at 280C until the time of HER2 ECD automated and manual analyses. Serum HER2 ECD levels were measured using both HER2 assays to compare the two ELISA methods. With the automated method, baseline serum HER2 ECD levels were determined with the ADVIA Centaur HER2/neu assay (Bayer Corporation, Tarrytown, NY), on the basis of two
884 | Ludovini et al.
Annals of Oncology
mAbs directed against the ECD of the HER2 antigen, using direct chemiluminescent technology. The measured chemiluminescence is directly proportional to the quantity of HER2/neu antigen in the sample. Quality control was ensured by assaying the two levels of control sera supplied with the kit in each series. Mean 6 standard deviation (SD) and coefficient of variation (CV) for the controls were 15.7 6 0.75 ng/ml (CV 4.5%) and 112.9 6 4.99 ng/ml (CV 4.4%), respectively. This automated assay for HER2 ECD was demonstrated to be accurate, precise, resistant to interferences and reliable for longitudinal monitoring [21]; the upper limit of normal was defined as 15 ng/ml [21]. With the manual method, the HER2 ECD levels were measured using a sandwich enzyme immunoassay (Human Neu Oncoprotein ELISA; Oncogene Science/Bayer Diagnostics, Cambridge, MA), following the manufacturer’s recommended protocol. The concentration of HER2 ECD in the samples was determined by interpolation of sample absorbance from the standard curve. The minimum detectable was 0.3 ng/ml and intra- and inter-assay coefficients of variation were <10%. The upper limit of normal was defined as 15 ng/ml, as previously reported [18].
tissue HER2 analyses by IHC and FISH IHC staining of specimens was carried out on formalin-fixed paraffinembedded breast cancer tissues using the mAb CB11 which targets the intracellular domain of HER2/neu protein (BioGenex, San Ramon, CA). According to the Hercep Test criteria, immunoreaction was scored as: 3+ if >10% of tumor cells showed strong and complete membrane staining, 2+ if membrane positivity was moderate and complete in >10% cells, 1+ if membrane positivity was weak and incomplete in >10% cells and 0 if membrane staining was absent or present in <10% cells. Tumors scored as 3+ were considered HER2 positive while tumors scored as 0/1+ were designated as HER2 negative. In 2+ tumors evaluated by IHC, FISH analysis was carried out, using the Abbott-Vysis Path Vysion HER2 DNA Probe Kit (Abbott Laboratories, Abbott Park, IL), following the manufacturer’s recommended protocol. The results were reported as the ratio between the average copy number of the HER2/neu gene and that of the chromosome 17 centromere, analyzing 60 neoplastic nuclei. Specimens with a signal ratio of <2.0 were considered as nonamplified and 2.0 or greater as amplified.
statistical analysis DFS was defined as the time from surgery to first appearance of disease or death for any cause. Patients known to be alive and without disease at the time of analysis were censored at their last follow-up date. OS was defined as the time from surgery to death for any cause. Survival curves were estimated using the Kaplan–Meier method. Cox proportional hazards model, as implemented in the PHREG program in SAS (SAS Institute Inc., Cary, NC) were used to evaluate the effect of serum HER2 ECD and HER2 tissue and clinicopathological variables on OS and DFS. A proportional hazards model was used for univariate and multivariate analyses. Variables that were found to be associated with OS and DFS in the univariate analysis (statistical significance was set at P < 0.05) were considered for the multivariate analysis. Results are expressed as hazard ratios (HRs) with 95% confidence xintervals (CIs). The v2 test was used to assess the association among clinical–pathological features and the expression of HER2 tissue or the levels of HER2 ECD. Our analyses were carried out utilizing the results obtained from automated ELISA assay and treating the levels of HER2 ECD as a dichotomous variable (HER2 high ‡ 15 ng/ml versus low < 15 ng/ml). All P values were derived from significance two-sided tests. Statistical analysis was carried out using SAS (Statistical Analysis System, SAS Institute Inc., version 9.1) software.
Volume 19 | No. 5 | May 2008
original article
Annals of Oncology
results patient characteristics From May 2000 to July 2005, a total of 256 consecutive patients with radically resected primary breast cancer, referred to the Breast Unit Surgical Department of the University of Perugia, Italy, were recruited. Histological diagnosis was made at the Institute of Pathology, University of Perugia. The main clinicopathological characteristics of the patients in our series are summarized in Table 1. Eastern Cooperative Oncology Group performance status was zero to one in 99.2% of patients. Ductal infiltrating carcinoma histology was reported in 216 patients (84.4%). All patients underwent surgery: conservative surgery was carried out in 214 patients (83.6%) and mastectomy in 42 patients (16.4%). Radiotherapy was delivered to 198 patients. Adjuvant chemotherapy was administered to 191 patients (74.6%): 90 patients (47.1%) received combination chemotherapy with cyclophosphamide, methotrexate and fluorouracil, 89 patients (46.6%) anthracycline-based therapy, 2 patients (1.1%) taxane-based therapy and 10 patients (5.2%) anthracycline- and taxane-based regimens. Endocrine therapy was administered to 181 of 256 patients (70%) (Table 1). baseline serum HER2 ECD and tissue HER2 status With a cut-off value of 15 ng/ml, 23 of the 256 patients (9.0%) had HER2 ECD levels ‡15 ng/ml (high levels) and 233 patients (91.0%) had HER2 ECD levels <15 ng/ml (low levels). Tissue HER2 status was positive (3+ by IHC and 2+ by IHC with FISH amplification) in 42 patients (16.4%) and negative in 214 patients (83.6%). correlation between automated and manual methods to assay serum HER2 ECD levels The correlation between the ADVIA Centaur and Oncogene ELISA methods was evaluated in all 256 breast cancer patients. The mean value of HER2 ECD measured by automated ADVIA Centaur was 10.8 ng/ml (SD = 3.3 ng/ml, median 10.2 ng/ml, range 5.6–33.5 ng/ml) while that measured by the Oncogene Science Manual Kit was 10.2 ng/ml (SD = 3.3 ng/ml, median 9.75 ng/ml, range 4.0–26.5 ng/ml). Pearson’s correlation test, applied to determine the linear regression between automated and manual HER2 ECD results, showed a good correlation between the two methods (r = 0.65, P < 0.0001) (Figure 1). association between tissue HER2 status (IHC/FISH) and baseline serum HER2 ECD levels The bivariate distributions of patients with tissue HER2 status and baseline serum HER2 ECD levels are shown in Table 2. We observed high serum HER2 ECD levels in 38.1% (16 of 42) of patients with tissue HER2-positive status versus only 3.3% (7 of 214) of patients with tissue HER2-negative status. The v2 P indicates an association between high HER2 ECD levels and tissue HER2 status (P < 0.0001). HER2 ECD levels were moderately concordant with tissue HER2 status (87.1%, K statistic = 0.42, 95% CI 0.26–0.58).
Volume 19 | No. 5 | May 2008
Table 1. Tumor and patient characteristics (n = 256) Characteristics Median age, years (range) Menopausal status Premenopausal Postmenopausal Stage I II III Tumor size, cm £2 >2 Undetermined Lymph nodes status Negative Positive Tumor grade G1 G2 G3 Unknown Receptor status (cut-off ‡ 10%) ER status ER negative ER positive Unknown PgR status PgR negative PgR positive Unknown Ki67 status (cut-off > 13%) Ki67 negative Ki67 positive Unknown Adjuvant systemic treatment Adjuvant CT Endocrine therapy Only endocrine therapy Endocrine therapy after CT Serum HER2 ECD ‡15 ng/ml <15 ng/ml Tissue HER2 (IHC/FISH) Positivea Negative
No. of patients 55.5
% (25.9–84.9)
95 161
(37.1) (62.9)
138 79 39
(53.9) (30.9) (15.2)
180 61 15
(70.3) (23.8) (5.8)
163 93
(63.7) (36.7)
24 146 55 31
(9.4) (57.0) (21.5) (12.1)
86 169 1
(33.6) (66.0) (0.4)
120 135 1
(46.9) (52.7) (0.4)
135 119 2
(52.7) (46.5) (0.8)
191 181 54/181 127/181
(74.6) (70.0) (29.8) (70.1)
23 233
(9.0) (91.0)
42 214
(16.4) (83.6)
a
IHC 3+, IHC 2+ and FISH amplified. ER, estrogen receptor; PgR, progesterone receptor; CT, chemotherapy; HER2, human epidermal growth factor receptor-2; ECD, extracellular domain; IHC, immunohistochemistry; FISH, fluorescence in situ hybridization.
relationship between baseline serum HER2 ECD levels and clinicopathological variables High HER2 ECD levels were significantly associated with stage III (P = 0.008), lymph node involvement (P = 0.035), high histological grade (P = 0.003), negativity of ER (P = 0.016) and negativity of PgR (P = 0.007) (Table 3). No statistical relationship was found between HER2 ECD levels and the other
doi:10.1093/annonc/mdm585 | 885
original article
Annals of Oncology
Table 3. Relationship between serum HER2 ECD levels and clinicopathological variables Characteristics
Figure 1. Comparison of serum human epidermal growth factor receptor2 extracellular domain (HER2 ECD) levels in 256 breast cancer patients evaluated by automated ADVIA Centaur and manual Oncogene Science enzyme-linked immunosorbent assay methods. Table 2. Association between tissue HER2 status and serum HER2 ECD levels Tissue HER2 statusa Positive Negative Serum HER2 ECD High ‡15 ng/ml Low <15 ng/ml Total
16 26 42
7 207 214 v2 test, P < 0.0001
Total 23 233
a
Positive tissue HER2 status: IHC 3+, IHC 2+ and FISH amplified; observed concordance 87.1%, K statistic = 0.42, CI 0.26–0.58. HER2, human epidermal growth factor receptor-2; ECD, extracellular domain; IHC, immunohistochemistry; CI, confidence interval; FISH, fluorescence in situ hybridization.
variables, such as age, menopausal status, histological tumor type and tumor size.
relationship between HER2 tissue status and clinicopathological variables Statistically significant associations were found among tissue HER2-positive status, high histological grade (P < 0.0001) and negativity of ER (P < 0.0001) and PgR (P = 0.015) (Table 4). No statistically significant relationship was found between HER2 tissue status and the other variables (menopausal status, histological tumor type and tumor size, stage and lymph node involvement). survival analysis At a median follow-up of 3.14 years (2.85–3.38 years), 24 patients (9.4%) had a recurrence. Locoregional recurrence was observed in 8 patients (33.4%) and metastatic disease in 16 patients (66.6%); dominant site was visceral in 12 of 16 patients
886 | Ludovini et al.
HER2 ECD <15 ng/ml N %
Menopausal status Premenopausal 89 (93.7) Postmenopausal 144 (89.4) Stage I 130 (94.2) II 72 (91.2) III 31 (79.5) Tumor size, cm £2 166 (92.2) >2 55 (90.2) Lymph nodes status Negative 153 (93.9) Positive 80 (86.0) Grade G1 24 (100) G2 136 (93.2) G3 45 (81.8) Receptor status (cut-off ‡ 10%) ER status ER negative 73 (84.9) ER positive 159 (94.2) PgR status PgR negative 103 (85.8) PgR positive 129 (95.6)
HER2 ECD ‡15 ng/ml N %
P value
6 17
(6.3) (10.6)
0.25
8 7 8
(5.8) (8.9) (20.5)
0.008*
14 6
(7.8) (9.8)
0.61
10 13
(6.1) (14.0)
0.035*
0 10 10
(0) (6.8) (18.2)
0.003*
13 10
(15.2) (5.9)
0.016*
17 6
(14.2) (4.4)
0.007*
HER2, human epidermal growth factor receptor-2; ECD, extracellular domain; ER, estrogen receptor; PgR, progesterone receptor. *P value <0.05, statistically significant.
(75%). Two patients died without recurrence of breast cancer; so far 26 events occurred in our series for the analysis of DFS. We observed 9 events in 23 patients (39.1%) with high HER2 ECD levels and only 17 events in 233 patients (7.3%) with low HER2 ECD levels. Dominant site of relapse was visceral in 6 of 9 patients with high HER2 ECD levels (66.6%) versus 6 of 15 patients with low HER2 ECD levels (40%). DFS was significantly shorter in patients with high HER2 ECD levels at diagnosis when compared with patients with low HER2 ECD levels (HR = 5.76, 95% CI 2.50–13.27, P < 0.0001) (Figure 2). DFS was also worse in patients with tissue HER2-positive status in comparison to those with tissue HER2 negative, but this difference was not statistically significant (HR = 2.14, 95% CI 0.91–5.04, P = 0.082) (Figure 3). To date, 12 patients died and no difference in OS has been observed either according to serum HER2 ECD levels (HR = 2.40, 95% CI 0.51–11.35, P = 0.269) or to HER2 tissue expression (HR = 2.18, 95% CI 0.59–8.09, P = 0.245). To assess the effect of each variable on DFS and OS, a Cox model was carried out first using univariate analysis. High HER2 ECD levels (P < 0.0001), stage III (P < 0.0001), lymph node positivity (P = 0.001) and negative PgR status (P = 0.026) were significantly associated with worse DFS (Table 5). Stage III status (HR = 4.53, 95% CI 1.83–11.23, P = 0.001) and lymph node positivity (HR = 6.43, 95% CI 1.37–30.10, P = 0.018) were significantly associated with OS (data not shown).
Volume 19 | No. 5 | May 2008
original article
Annals of Oncology
Table 4. Relationship between tissue HER2 status and clinicopathological variables Characteristics
HER2 negative N %
Menopausal status Premenopausal 80 (84.2) Postmenopausal 13 (83.2) Stage I 11 (83.3) II 68 (86.1) III 31 (79.5) Tumor size, cm £2 15 (86.7) >2 49 (80.3) Lymph nodes status Negative 13 (84.1) Positive 77 (82.8) Grade G1 24 (100) G2 12 (88.4) G3 33 (60.0) Receptor status (cut-off ‡ 10%) ER status ER negative 60 (69.7) ER positive 15 (90.5) PgR status PgR negative 93 (77.5) PgR positive 12 (88.9)
HER2 positive N %
P value
15 27
(15.8) (16.8)
0.83
23 11 8
(16.7) (13.9) (20.5)
0.77
24 12
(13.3) (19.7)
0.23
26 16
(15.9) (17.2)
0.79
0 17 22
(0) (11.6) (40.0)
<0.0001*
26 16
(30.3) (9.5)
<0.0001*
27 15
(22.5) (11.1)
0.015*
A multivariate Cox regression model for DFS and OS was built using the variables that were found significant at the univariate analysis. High serum HER2 ECD levels (HR = 3.25, 95% CI 1.34–7.88, P = 0.009) and positive lymph nodes (HR = 2.90, 95% CI 1.01–8.29, P = 0.048) were significantly associated with worse DFS (Table 5). Stage (III versus I and II) was an independent prognostic factor for OS (HR = 4.41, 95% CI 1.76–11.06, P = 0.002) (data not shown). We also explored the impact of systemic adjuvant therapy on DFS and OS, but no association was observed. Adjuvant trastuzumab was not administered to any patient. After recurrence, seven of nine patients with tissue HER2-positive disease received trastuzumab-based therapy.
discussion
HER2, human epidermal growth factor receptor-2; ER, estrogen receptor; PgR, progesterone receptor. *P value <0.05, statistically significant.
The role of serum HER2 ECD in patients with early breast cancer as prognostic factor is not well defined. The percentage of elevated serum levels of HER2 ECD are extremely variable in early breast cancer patients at the time of diagnosis. In our study, high HER2 ECD levels were observed in 9.0% of early breast cancer patients, similar to the results reported by other authors [22–24]. To measure the levels of HER2 ECD, we used both the automated and manual ELISA methods which are standardized, sensitive, reproducible and appropriate for routine analysis. Both assays utilize the same two antibodies (NB-3 and TA-1) that are directed against different epitopes of the HER2 ECD and both use a value of 15 ng/ml as cut-off. We showed that there was a strong correlation between these two ELISA methods, as reported by others [25, 26]. Both methods have been cleared by the Food and Drug Administration for use in
Figure 2. Kaplan–Meier estimates for disease-free survival according to serum human epidermal growth factor receptor-2 extracellular domain (HER2 ECD) levels.
Volume 19 | No. 5 | May 2008
doi:10.1093/annonc/mdm585 | 887
original article
Annals of Oncology
Figure 3. Kaplan–Meier estimates for disease-free survival according to tissue human epidermal growth factor receptor-2 (HER2) expression
the management and monitoring of women with metastatic breast cancer besides the automated Immuno-1 HER2 test. High HER2 EDC levels were associated with shorter DFS in our series of early breast cancer patients and according to Cox regression analysis, patients with elevated HER2 ECD showed a relative risk of 3.25 for reduced DFS. Elevated HER2 ECD levels were also correlated with the risk of visceral recurrence (66.6% versus 40%), in accordance with observations already reported, analyzing tissue HER2 overexpression in early breast cancer patients [27]. Few trials have evaluated the prognostic role of serum HER2 ECD in early breast cancer. The results of our prospective study are in agreement with two retrospective studies [28, 24] that showed a shorter DFS in patients with elevated serum HER2 ECD levels. The adverse prognostic value of elevated HER2 ECD levels in early breast cancer could reflect the presence of micrometastases or high HER2 cleavage and shedding with the production of truncated cell-associated fragments that contain the signaling kinase domain activated in absence of the ECD [29]. Therefore, these tumors with a deregulated growth-promoting pathway could have a more aggressive behavior. Moreover, in our study, elevated levels of HER2 ECD were associated with several factors related to tumor aggressiveness, such as stage III, lymph node involvement, poor histological differentiation, ER negativity and PgR negativity (Table 3). The correlation of tissue HER2 status (detected by IHC and/or FISH) with serum HER2 ECD levels was also analyzed. High serum HER2 ECD levels were observed in 38.1% (16 of 42) of patients with tissue HER2 positivity and in only 3.3% (7 of 214) of the patients with tissue HER2 negativity, with a moderate concordance (87.1%) as reported by others [23]. The targeted molecular domains and protein function are different: the IHC method recognizes
888 | Ludovini et al.
Table 5. Univariate and multivariate analyses for variables considered for DFS (Cox proportional hazards regression model) DFS Hazard ratio
Univariate analysis Stage (III versus I and II) Lymph nodes status (positive versus negative) PgR (cut-off ‡ 10%) (negative versus positive) Serum HER2 ECDa (high versus low) HER2 tissueb (positive versus negative) Multivariate analysis Serum HER2 ECDa (high versus low) Lymph nodes status (positive versus negative)
95% confidence interval
P value
5.02 4.66
2.32–10.90 1.96–11.10
<0.0001* 0.001*
0.39
0.17–0.89
0.026*
6.46
2.75–15.17
<0.0001*
2.14
0.91–5.04
0.082*
3.25
1.34–7.88
0.009*
2.9
1.01–8.29
0.048*
a
HER2 ECD: high (‡15 ng/ml) versus low (<15 ng/ml). Positive tissue HER2 status: IHC 3+, IHC 2+ and FISH amplified. DFS, disease-free survival; PgR, progesterone receptor; HER2, human epidermal growth factor receptor-2; ECD, extracellular domain; FISH, fluorescence in situ hybridization; IHC, immunohistochemistry. *P value <0.05, statistically significant. b
intracytoplasmic domains of the entire HER2 receptor; the serum HER2 assay measures the ECD of the HER2 receptor by using two mAbs recognizing two independent epitopes [30].
Volume 19 | No. 5 | May 2008
Annals of Oncology
In conclusion, the ELISA method for measuring the circulating HER2 ECD is a useful tool to obtain the real-time status of HER2 and both manual and automated methods may be used. Our findings indicate that the determination of serum HER2 ECD might aid to assess the HER2 status, in addition to the conventional assays carried out on tumor specimens. Due to the correlation between HER2 ECD levels and HER2 expression in tumoral tissue, patients could be considered for trastuzumab therapy only if tissue is not available, on the basis of high HER2 ECD levels. Elevated HER2 ECD levels may also represent a subgroup of HER2-positive tumors with a higher level of HER2 cleavage and shedding. In vitro studies have shown that high amounts of soluble HER2 ECD may neutralize the biological activity of anti-HER2 antibodies by forming immune complexes and blocking the accessibility of the antibodies to the tumor [31]. This subgroup of tumors may have a more aggressive clinical course and the availability of a noninvasive, repeatable and reproducible technique to measure circulating HER2 ECD appears to be a useful tool for identifying high-risk breast cancer patients. In conclusion, to our knowledge, this is the only prospective trial that has evaluated the prognostic role HER2 ECD in patients with operable breast cancer. A significant association with other poor prognostic factors has been found as well as with a worse DFS; however, due to limited percentage of patients with elevated serum levels of HER2 ECD in this series, these results must be interpreted cautiously and need to be confirmed in large prospective trials with serial measurements. It could also be worthwhile to evaluate in patients with high serum HER2 ECD levels the efficacy of different treatments targeting HER2 (receptor tyrosine kinase inhibitors) in comparison with trastuzumab.
funding Bayer Diagnostics; Umbria Association Against Cancer; ‘Conoscere per Vincere’ charities. This work was supported in part (reagents for HER2 ECD analysis) by Bayer Diagnostics (Cambridge, MA 02142-1168, USA), by the Umbria Association Against Cancer (AUCC) and by ‘‘Conoscere per Vincere’’ charities.
acknowledgements We thank the patients who participated in this study.
references 1. Aaronson SA. Growth factors and cancer. Science 1991; 254: 1146–1153. 2. Slamon DJ, Clark GM, Wong SG et al. Human breast cancer: correlation of relapse and survival with amplification of the HER2/neu oncogene. Science 1987; 235: 177–182. 3. Carlomagno C, Perrone F, Gallo C et al. C-erbB-2 overexpression decreases the benefit of adjuvant tamoxifen in early stage breast cancer without axillary lymph node metastases. J Clin Oncol 1996; 14: 2702–2708.
Volume 19 | No. 5 | May 2008
original article 4. Gusterson BA, Gelber RD, Goldhirsch A et al. Prognostic importance of c-erbB-2 expression in breast cancer. J Clin Oncol 1992; 10: 1049–1056. 5. Stal O, Sullivan S, Wingren S et al. C-erbB-2 expression and benefit from adjuvant chemotherapy and radiotherapy of breast cancer. Eur J Cancer 1995; 31: 2185–2190. 6. Muss HB, Thor AD, Berry DA et al. C-erbB-2 expression and response to adjuvant therapy in women with node-positive early breast cancer. N Engl J Med 1994; 330: 1260–1266. 7. Paik S, Bryant J, Park C et al. ErbB-2 and response to doxorubicin in patients with axillary lymph node-postitive, hormone receptor-negative breast cancer. J Natl Cancer Inst 1998; 90: 1361–1370. 8. Hudis CA. Trastuzumab-mechanism of action and use in clinical practice. N Engl J Med 2007; 357: 39–51. 9. Muller V, Thomssen C, Karakas C et al. Quantitative assessment of HER2/neu protein concentration in breast cancer by enzyme-linked immunosorbent assay. Int J Biol Markers 2003; 18: 13–20. 10. Krainer M, Brodowicz T, Zeillinger R et al. Tissue expression and serum levels of HER2/neu in patients with breast cancer. Oncology 1997; 54: 475–481. 11. Molina R, Jo J, Filella X et al. C-erbB-2 CEA and Ca 15.3 serum levels in the early diagnosis recurrence of breast cancer patients. Anticancer Res 1999; 19(4A): 2551–2555. 12. Kong SY, Nam BH, Lee KS et al. Predicting tissue HER2 status using serum HER2 levels in patients with metastatic breast cancer. Clin Chem 2006; 52: 1510–1515. 13. Beltran M, Colomer R, Lipton A et al. Does HER2 status predict only a decreased response to hormone therapy in advanced breast cancer, or does it also predict the extent of metastatic disease? J Clin Oncol 2002; 20: 4605–4606. 14. Fehm T, Gebauer G, Jager W. Clinical utility of serial serum c-erbB-2 determinations in the follow-up of breast cancer patients. Breast Cancer Res Treat 2002; 75: 97–106. 15. Lipton A, Ali SM, Leitzel K et al. Elevated serum HER2/neu level predicts decreased response to hormone therapy in metastatic breast cancer. J Clin Oncol 2002; 20: 1467–1472. 16. Perez EA, Geeraerts L, Suman VJ et al. A randomized phase II study of sequential docetaxel and doxorubicin/cyclophosphamide in patients with metastatic breast cancer. Ann Oncol 2002; 13: 1225–1235. 17. Colomer R, Llombart-Cussac A, Lluch A et al. Biweekly paclitaxel plus gemcitabine in advanced breast cancer: phase II trial and predictive value of HER2 extracellular domain. Ann Oncol 2004; 15: 201–206. 18. Fornier MN, Seidman AD, Schwartz MK et al. Serum HER2 extracellular domain in metastatic breast cancer patients treated with weekly trastuzumab and paclitaxel: association with HER2 status by immunohistochemistry and fluorescence in situ hybridization and with response rate. Ann Oncol 2005; 16: 234–239. 19. Yuan P, Xu BH, Chu DT. Correlation between serum HER-2 oncoprotein and patients with breast cancer. Chin Med Sci J 2004; 19(5): 883–890. 20. Singletary SE, Allred C, Ashley P et al. Revision of the American Joint Committee on Cancer staging system for breast cancer. J Clin Oncol 2002; 20: 3628–3636. 21. Schwartz MK, Smith C, Schwartz DC et al. Monitoring therapy by serum HER2/neu. Int J Biol Markers 2000; 15: 324–329. 22. Isola JJ, Holli K, Oksa H et al. Elevated erbB-2 oncoprotein levels in preoperative and follow-up serum samples define an aggressive disease course in patients with breast cancer. Cancer 1994; 73: 652–658. 23. Molina R, Filella X, Zanon G et al. Prospective evaluation of tumor markers (c-erbB-2 oncoprotein, CEA and CA 15.3) in patients with locoregional breast cancer. Anticancer Res 2003; 23: 1043–1050. 24. Saghatchian M, Guepratte S, Hacene K et al. Serum HER-2 extracellular domain: relationship with clinicobiological presentation and prognostic value before and after primary treatment in 701 breast cancer patients. Int J Biol Markers 2004; 19: 14–22. 25. Cook GB, Neaman IE, Goldblatt JL et al. Clinical utility of serum HER2/neu testing on the Bayer Immuno 1 automated system in breast cancer. Anticancer Res 2001; 21: 1465–1470.
doi:10.1093/annonc/mdm585 | 889
original article 26. Dittadi R, Zancan M, Perasole A, Gion M. Evaluation of HER-2/neu in serum and tissue of primary and metastatic breast cancer patients using an automated enzyme immunoassay. Int J Biol Markers 2001; 16: 255–261. 27. Kallioniemi OP, Holli K, Visakorpi T et al. Association of c-erbB-2 protein overexpression with high rate of cell proliferation, increased risk of visceral metastases and poor long-term survival in breast cancer. Int J Cancer 1991; 49: 650–655. 28. Mehta RR, McDermott JH, Heiken TJ et al. Plasma c-erbB-2 levels in breast cancer patients: prognostic significance in predicting response to chemotherapy. J Clin Oncol 1998; 16: 2409–2416.
890 | Ludovini et al.
Annals of Oncology
29. Molina MA, Codony-Servat J, Albanell J et al. Trastuzumab (Herceptin), a humanized antiHER2 receptor monoclonal antibody, inhibits basal and activated HER2 ectodomain cleavage in breast cancer cells. Cancer Res 2001; 61: 4744–4749. 30. Carney WO, Hamer PJ, Petit D et al. Detection and quantitation of the human neu oncoprotein. J Tumour Mark Oncol 1991; 6: 53–72. 31. Brodowicz T, Wiltschke C, Budinsky AC et al. Soluble HER2/neu neutralizes biologic effects of anti-HER2/neu antibody on breast cancer cells in vitro. Int J Cancer 1997; 73: 875–879.
Volume 19 | No. 5 | May 2008
Annals of Oncology 19: 883–890, 2008 doi:10.1093/annonc/mdm585 Published online 10 January 2008
Evaluation of serum HER2 extracellular domain in early breast cancer patients: correlation with clinicopathological parameters and survival V. Ludovini1 *, S. Gori1 , M. Colozza1, L. Pistola1, E. Rulli2, I. Floriani2, E. Pacifico3, F.R. Tofanetti1, A. Sidoni4, C. Basurto1, A. Rulli5 & L. Crino`1 1
Medical Oncology Division, Azienda Ospedaliera of Perugia, Perugia; 2Department of Oncology, Istituto di Ricerche Farmacologiche ‘Mario Negri’, Milan; Biochemistry and Clinical Chemistry Laboratory, Azienda Ospedaliera of Perugia, Perugia; 4Institute of Pathology; 5‘Breast Unit’, Department of Surgical, University of Perugia, Perugia, Italy 3
Received 2 October 2007; revised 1 December 2007; accepted 3 December 2007
extracellular domain (ECD) and tissue HER2 status, their relationship with clinicopathological parameters and their impact on disease-free survival (DFS) and overall survival in early breast cancer patients. Patients and methods: This prospective trial included patients with stage I–III breast cancer. Serum HER2 ECD levels were measured by two enzyme-linked immunosorbent assays before surgical treatment. Tissue HER2 status was analyzed by immunohistochemistry (IHC) in all tumors; FISH assay was utilized in HER2 2+ tumors by IHC. Results: From May 2000 to July 2005, 256 consecutive stage I–III breast cancer patients were included in this study. High serum HER2 ECD levels (‡15 ng/ml) were reported in 23 patients (9.0%) and HER2-positive status in tumor tissue was observed in 42 patients (16.4%) with a concordance of 87.1%. High HER2 ECD levels were significantly associated with high histological grade (P = 0.003), stage III (P = 0.008), lymph node involvment (P = 0.035) and negativity of both estrogen (P = 0.016) and progesterone (P = 0.007) receptors. At multivariate analysis, high serum HER2 ECD levels were a significant independent prognostic factor of worse DFS (P = 0.009). Conclusions: A statistically significant association was observed between high serum HER2 ECD levels and worse DFS in early breast cancer patients. Key words: early breast cancer, HER2 ECD, prognosis
introduction The human epidermal growth factor receptor-2 (HER2) proto-oncogene, also known as HER2/neu and c-erbB-2, encodes a growth factor receptor that has been found to play an important role in breast cancer [1]. In up to 30% of breast cancer patients, the HER2 gene is amplified and its associated receptor protein is overexpressed on the tumor cell surface, thus playing an important role in the malignant transformation and clinical aggressiveness of breast cancer. HER2 overexpression is an independent predictor of shorter overall survival (OS) and disease-free survival (DFS) in invasive primary breast cancer [2]. Retrospective evaluations of randomized clinical trials have revealed a correlation between HER2 overexpression and *Correspondence to: Dr V. Ludovini, Medical Oncology Division, Azienda Ospedaliera of Perugia, via Brunamonti 51, 06122 Perugia, Italy. Tel: +39-075-5783453; Fax: +39-075-5729946; E-mail: [email protected]
These authors contributed equally to this work.
resistance to tamoxifen [3–5] while an interaction between HER2 overexpression and enhanced response to doxorubicincontaining regimens has been reported [6–7]. A variety of methods are available to assess tissue HER2 status: immunohistochemistry (IHC), which detects protein overexpression and FISH, which detects HER2 gene amplification are the most widely applied techniques. Tissue HER2 status evaluation is very important because HER2 protein provides a target for therapeutic approaches. Trastuzumab, a mAb (Herceptin; Genentech Inc., South San Francisco, CA) alone or in combination with chemotherapy has been shown to be an effective treatment for HER2-positive breast cancer (IHC 3+ or FISH amplified) in metastatic, neo-adjuvant and adjuvant settings [8]. HER2 receptor is composed of a cytoplasmic domain with tyrosine kinase activity, a transmembrane domain and an extracellular domain (ECD). The HER2 ECD may be cleaved and shed from the surface of breast cancer cells and serum HER2 ECD levels can be detected by enzyme-linked immunosorbent assays (ELISAs) without any significant
ª The Author 2008. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected]
original article
Background: We explored the correlation between serum human epidermal growth factor receptor-2 (HER2)
original article cross-reactivity with other members of the HER receptor family [9]. The manual and automated ELISA testing methods of serum HER2 ECD are standardized methodologies and are not prone to subjective variable interpretation by a reviewer. Shed HER2 ECD has been detected in the serum from 20% to 40% of patients with metastatic disease [10–11] and high serum concentrations of HER2 ECD have been associated with HER2 overexpression [12], increased tumor burden [13], worse survival [14] and resistance to endocrine therapy and chemotherapy [15–17]. On the other hand, HER2 ECD seems to be associated with sensitivity to HER2-targeted agents, such as trastuzumab. A decrease in serum HER2 ECD levels in metastatic breast cancer patients treated with trastuzumab was predictive of response [18]. In early breast cancer, shed HER2 ECD has been detected up to 20% of patients [19], but the clinical significance of circulating HER2 ECD is uncertain and the relationship of baseline HER2 ECD levels with tissue HER2 overexpression in primary tumors has been little studied. Therefore, we carried out a prospective study in operable breast cancer patients to evaluate the correlation between HER2 ECD serum levels and HER2 tumor status measured by IHC and FISH, their relationship with clinical–pathological parameters and their impact on DFS and OS.
patients and methods This prospective trial was conducted in patients with a histological diagnosis of stage I–III breast cancer treated with conservative surgery or mastectomy. Tumor staging followed the tumor–node–metastasis (TNM)– American Joint Committee on Cancer classification [20] and the pTNM was obtained after classical pathological examination. Patients with metastatic disease and with other previous tumors were excluded from this study. Estrogen receptors (ERs), progesterone receptors (PgRs) and HER2 status were assessed at the time of surgery on formalin-fixed paraffinembedded tissue blocks of the primary tumor in the Pathology Department of the University of Perugia. Recorded clinical and pathological features for each patient include age, menopausal status, histology, grade, Ki67, ER and PgR status, stage, surgical treatment and medical adjuvant therapy. Follow-up, including clinical examination (every 3 months for the first 2 years, every 6 months for the next 3 years and yearly thereafter) and annual Rx-mammography, was carried out in all patients. Recurrence was defined as the first documented evidence of new disease manifestation in the locoregional area, in the contralateral breast, in distant sites, or in a combination of these. The study was reviewed and approved by the institution’s Ethics Committee and informed consent was obtained by all patients.
serum HER2 ECD assays Serum samples were prospectively collected from breast cancer patients before surgery. Five milliliters of peripheral blood were collected in a sterile test tube (without anticoagulants) and centrifuged at 3000 g for 10 min at room temperature. Serum was stocked in 0.5-ml aliquots in cryovials and stored at 280C until the time of HER2 ECD automated and manual analyses. Serum HER2 ECD levels were measured using both HER2 assays to compare the two ELISA methods. With the automated method, baseline serum HER2 ECD levels were determined with the ADVIA Centaur HER2/neu assay (Bayer Corporation, Tarrytown, NY), on the basis of two
884 | Ludovini et al.
Annals of Oncology
mAbs directed against the ECD of the HER2 antigen, using direct chemiluminescent technology. The measured chemiluminescence is directly proportional to the quantity of HER2/neu antigen in the sample. Quality control was ensured by assaying the two levels of control sera supplied with the kit in each series. Mean 6 standard deviation (SD) and coefficient of variation (CV) for the controls were 15.7 6 0.75 ng/ml (CV 4.5%) and 112.9 6 4.99 ng/ml (CV 4.4%), respectively. This automated assay for HER2 ECD was demonstrated to be accurate, precise, resistant to interferences and reliable for longitudinal monitoring [21]; the upper limit of normal was defined as 15 ng/ml [21]. With the manual method, the HER2 ECD levels were measured using a sandwich enzyme immunoassay (Human Neu Oncoprotein ELISA; Oncogene Science/Bayer Diagnostics, Cambridge, MA), following the manufacturer’s recommended protocol. The concentration of HER2 ECD in the samples was determined by interpolation of sample absorbance from the standard curve. The minimum detectable was 0.3 ng/ml and intra- and inter-assay coefficients of variation were <10%. The upper limit of normal was defined as 15 ng/ml, as previously reported [18].
tissue HER2 analyses by IHC and FISH IHC staining of specimens was carried out on formalin-fixed paraffinembedded breast cancer tissues using the mAb CB11 which targets the intracellular domain of HER2/neu protein (BioGenex, San Ramon, CA). According to the Hercep Test criteria, immunoreaction was scored as: 3+ if >10% of tumor cells showed strong and complete membrane staining, 2+ if membrane positivity was moderate and complete in >10% cells, 1+ if membrane positivity was weak and incomplete in >10% cells and 0 if membrane staining was absent or present in <10% cells. Tumors scored as 3+ were considered HER2 positive while tumors scored as 0/1+ were designated as HER2 negative. In 2+ tumors evaluated by IHC, FISH analysis was carried out, using the Abbott-Vysis Path Vysion HER2 DNA Probe Kit (Abbott Laboratories, Abbott Park, IL), following the manufacturer’s recommended protocol. The results were reported as the ratio between the average copy number of the HER2/neu gene and that of the chromosome 17 centromere, analyzing 60 neoplastic nuclei. Specimens with a signal ratio of <2.0 were considered as nonamplified and 2.0 or greater as amplified.
statistical analysis DFS was defined as the time from surgery to first appearance of disease or death for any cause. Patients known to be alive and without disease at the time of analysis were censored at their last follow-up date. OS was defined as the time from surgery to death for any cause. Survival curves were estimated using the Kaplan–Meier method. Cox proportional hazards model, as implemented in the PHREG program in SAS (SAS Institute Inc., Cary, NC) were used to evaluate the effect of serum HER2 ECD and HER2 tissue and clinicopathological variables on OS and DFS. A proportional hazards model was used for univariate and multivariate analyses. Variables that were found to be associated with OS and DFS in the univariate analysis (statistical significance was set at P < 0.05) were considered for the multivariate analysis. Results are expressed as hazard ratios (HRs) with 95% confidence xintervals (CIs). The v2 test was used to assess the association among clinical–pathological features and the expression of HER2 tissue or the levels of HER2 ECD. Our analyses were carried out utilizing the results obtained from automated ELISA assay and treating the levels of HER2 ECD as a dichotomous variable (HER2 high ‡ 15 ng/ml versus low < 15 ng/ml). All P values were derived from significance two-sided tests. Statistical analysis was carried out using SAS (Statistical Analysis System, SAS Institute Inc., version 9.1) software.
Volume 19 | No. 5 | May 2008
original article
Annals of Oncology
results patient characteristics From May 2000 to July 2005, a total of 256 consecutive patients with radically resected primary breast cancer, referred to the Breast Unit Surgical Department of the University of Perugia, Italy, were recruited. Histological diagnosis was made at the Institute of Pathology, University of Perugia. The main clinicopathological characteristics of the patients in our series are summarized in Table 1. Eastern Cooperative Oncology Group performance status was zero to one in 99.2% of patients. Ductal infiltrating carcinoma histology was reported in 216 patients (84.4%). All patients underwent surgery: conservative surgery was carried out in 214 patients (83.6%) and mastectomy in 42 patients (16.4%). Radiotherapy was delivered to 198 patients. Adjuvant chemotherapy was administered to 191 patients (74.6%): 90 patients (47.1%) received combination chemotherapy with cyclophosphamide, methotrexate and fluorouracil, 89 patients (46.6%) anthracycline-based therapy, 2 patients (1.1%) taxane-based therapy and 10 patients (5.2%) anthracycline- and taxane-based regimens. Endocrine therapy was administered to 181 of 256 patients (70%) (Table 1). baseline serum HER2 ECD and tissue HER2 status With a cut-off value of 15 ng/ml, 23 of the 256 patients (9.0%) had HER2 ECD levels ‡15 ng/ml (high levels) and 233 patients (91.0%) had HER2 ECD levels <15 ng/ml (low levels). Tissue HER2 status was positive (3+ by IHC and 2+ by IHC with FISH amplification) in 42 patients (16.4%) and negative in 214 patients (83.6%). correlation between automated and manual methods to assay serum HER2 ECD levels The correlation between the ADVIA Centaur and Oncogene ELISA methods was evaluated in all 256 breast cancer patients. The mean value of HER2 ECD measured by automated ADVIA Centaur was 10.8 ng/ml (SD = 3.3 ng/ml, median 10.2 ng/ml, range 5.6–33.5 ng/ml) while that measured by the Oncogene Science Manual Kit was 10.2 ng/ml (SD = 3.3 ng/ml, median 9.75 ng/ml, range 4.0–26.5 ng/ml). Pearson’s correlation test, applied to determine the linear regression between automated and manual HER2 ECD results, showed a good correlation between the two methods (r = 0.65, P < 0.0001) (Figure 1). association between tissue HER2 status (IHC/FISH) and baseline serum HER2 ECD levels The bivariate distributions of patients with tissue HER2 status and baseline serum HER2 ECD levels are shown in Table 2. We observed high serum HER2 ECD levels in 38.1% (16 of 42) of patients with tissue HER2-positive status versus only 3.3% (7 of 214) of patients with tissue HER2-negative status. The v2 P indicates an association between high HER2 ECD levels and tissue HER2 status (P < 0.0001). HER2 ECD levels were moderately concordant with tissue HER2 status (87.1%, K statistic = 0.42, 95% CI 0.26–0.58).
Volume 19 | No. 5 | May 2008
Table 1. Tumor and patient characteristics (n = 256) Characteristics Median age, years (range) Menopausal status Premenopausal Postmenopausal Stage I II III Tumor size, cm £2 >2 Undetermined Lymph nodes status Negative Positive Tumor grade G1 G2 G3 Unknown Receptor status (cut-off ‡ 10%) ER status ER negative ER positive Unknown PgR status PgR negative PgR positive Unknown Ki67 status (cut-off > 13%) Ki67 negative Ki67 positive Unknown Adjuvant systemic treatment Adjuvant CT Endocrine therapy Only endocrine therapy Endocrine therapy after CT Serum HER2 ECD ‡15 ng/ml <15 ng/ml Tissue HER2 (IHC/FISH) Positivea Negative
No. of patients 55.5
% (25.9–84.9)
95 161
(37.1) (62.9)
138 79 39
(53.9) (30.9) (15.2)
180 61 15
(70.3) (23.8) (5.8)
163 93
(63.7) (36.7)
24 146 55 31
(9.4) (57.0) (21.5) (12.1)
86 169 1
(33.6) (66.0) (0.4)
120 135 1
(46.9) (52.7) (0.4)
135 119 2
(52.7) (46.5) (0.8)
191 181 54/181 127/181
(74.6) (70.0) (29.8) (70.1)
23 233
(9.0) (91.0)
42 214
(16.4) (83.6)
a
IHC 3+, IHC 2+ and FISH amplified. ER, estrogen receptor; PgR, progesterone receptor; CT, chemotherapy; HER2, human epidermal growth factor receptor-2; ECD, extracellular domain; IHC, immunohistochemistry; FISH, fluorescence in situ hybridization.
relationship between baseline serum HER2 ECD levels and clinicopathological variables High HER2 ECD levels were significantly associated with stage III (P = 0.008), lymph node involvement (P = 0.035), high histological grade (P = 0.003), negativity of ER (P = 0.016) and negativity of PgR (P = 0.007) (Table 3). No statistical relationship was found between HER2 ECD levels and the other
doi:10.1093/annonc/mdm585 | 885
original article
Annals of Oncology
Table 3. Relationship between serum HER2 ECD levels and clinicopathological variables Characteristics
Figure 1. Comparison of serum human epidermal growth factor receptor2 extracellular domain (HER2 ECD) levels in 256 breast cancer patients evaluated by automated ADVIA Centaur and manual Oncogene Science enzyme-linked immunosorbent assay methods. Table 2. Association between tissue HER2 status and serum HER2 ECD levels Tissue HER2 statusa Positive Negative Serum HER2 ECD High ‡15 ng/ml Low <15 ng/ml Total
16 26 42
7 207 214 v2 test, P < 0.0001
Total 23 233
a
Positive tissue HER2 status: IHC 3+, IHC 2+ and FISH amplified; observed concordance 87.1%, K statistic = 0.42, CI 0.26–0.58. HER2, human epidermal growth factor receptor-2; ECD, extracellular domain; IHC, immunohistochemistry; CI, confidence interval; FISH, fluorescence in situ hybridization.
variables, such as age, menopausal status, histological tumor type and tumor size.
relationship between HER2 tissue status and clinicopathological variables Statistically significant associations were found among tissue HER2-positive status, high histological grade (P < 0.0001) and negativity of ER (P < 0.0001) and PgR (P = 0.015) (Table 4). No statistically significant relationship was found between HER2 tissue status and the other variables (menopausal status, histological tumor type and tumor size, stage and lymph node involvement). survival analysis At a median follow-up of 3.14 years (2.85–3.38 years), 24 patients (9.4%) had a recurrence. Locoregional recurrence was observed in 8 patients (33.4%) and metastatic disease in 16 patients (66.6%); dominant site was visceral in 12 of 16 patients
886 | Ludovini et al.
HER2 ECD <15 ng/ml N %
Menopausal status Premenopausal 89 (93.7) Postmenopausal 144 (89.4) Stage I 130 (94.2) II 72 (91.2) III 31 (79.5) Tumor size, cm £2 166 (92.2) >2 55 (90.2) Lymph nodes status Negative 153 (93.9) Positive 80 (86.0) Grade G1 24 (100) G2 136 (93.2) G3 45 (81.8) Receptor status (cut-off ‡ 10%) ER status ER negative 73 (84.9) ER positive 159 (94.2) PgR status PgR negative 103 (85.8) PgR positive 129 (95.6)
HER2 ECD ‡15 ng/ml N %
P value
6 17
(6.3) (10.6)
0.25
8 7 8
(5.8) (8.9) (20.5)
0.008*
14 6
(7.8) (9.8)
0.61
10 13
(6.1) (14.0)
0.035*
0 10 10
(0) (6.8) (18.2)
0.003*
13 10
(15.2) (5.9)
0.016*
17 6
(14.2) (4.4)
0.007*
HER2, human epidermal growth factor receptor-2; ECD, extracellular domain; ER, estrogen receptor; PgR, progesterone receptor. *P value <0.05, statistically significant.
(75%). Two patients died without recurrence of breast cancer; so far 26 events occurred in our series for the analysis of DFS. We observed 9 events in 23 patients (39.1%) with high HER2 ECD levels and only 17 events in 233 patients (7.3%) with low HER2 ECD levels. Dominant site of relapse was visceral in 6 of 9 patients with high HER2 ECD levels (66.6%) versus 6 of 15 patients with low HER2 ECD levels (40%). DFS was significantly shorter in patients with high HER2 ECD levels at diagnosis when compared with patients with low HER2 ECD levels (HR = 5.76, 95% CI 2.50–13.27, P < 0.0001) (Figure 2). DFS was also worse in patients with tissue HER2-positive status in comparison to those with tissue HER2 negative, but this difference was not statistically significant (HR = 2.14, 95% CI 0.91–5.04, P = 0.082) (Figure 3). To date, 12 patients died and no difference in OS has been observed either according to serum HER2 ECD levels (HR = 2.40, 95% CI 0.51–11.35, P = 0.269) or to HER2 tissue expression (HR = 2.18, 95% CI 0.59–8.09, P = 0.245). To assess the effect of each variable on DFS and OS, a Cox model was carried out first using univariate analysis. High HER2 ECD levels (P < 0.0001), stage III (P < 0.0001), lymph node positivity (P = 0.001) and negative PgR status (P = 0.026) were significantly associated with worse DFS (Table 5). Stage III status (HR = 4.53, 95% CI 1.83–11.23, P = 0.001) and lymph node positivity (HR = 6.43, 95% CI 1.37–30.10, P = 0.018) were significantly associated with OS (data not shown).
Volume 19 | No. 5 | May 2008
original article
Annals of Oncology
Table 4. Relationship between tissue HER2 status and clinicopathological variables Characteristics
HER2 negative N %
Menopausal status Premenopausal 80 (84.2) Postmenopausal 13 (83.2) Stage I 11 (83.3) II 68 (86.1) III 31 (79.5) Tumor size, cm £2 15 (86.7) >2 49 (80.3) Lymph nodes status Negative 13 (84.1) Positive 77 (82.8) Grade G1 24 (100) G2 12 (88.4) G3 33 (60.0) Receptor status (cut-off ‡ 10%) ER status ER negative 60 (69.7) ER positive 15 (90.5) PgR status PgR negative 93 (77.5) PgR positive 12 (88.9)
HER2 positive N %
P value
15 27
(15.8) (16.8)
0.83
23 11 8
(16.7) (13.9) (20.5)
0.77
24 12
(13.3) (19.7)
0.23
26 16
(15.9) (17.2)
0.79
0 17 22
(0) (11.6) (40.0)
<0.0001*
26 16
(30.3) (9.5)
<0.0001*
27 15
(22.5) (11.1)
0.015*
A multivariate Cox regression model for DFS and OS was built using the variables that were found significant at the univariate analysis. High serum HER2 ECD levels (HR = 3.25, 95% CI 1.34–7.88, P = 0.009) and positive lymph nodes (HR = 2.90, 95% CI 1.01–8.29, P = 0.048) were significantly associated with worse DFS (Table 5). Stage (III versus I and II) was an independent prognostic factor for OS (HR = 4.41, 95% CI 1.76–11.06, P = 0.002) (data not shown). We also explored the impact of systemic adjuvant therapy on DFS and OS, but no association was observed. Adjuvant trastuzumab was not administered to any patient. After recurrence, seven of nine patients with tissue HER2-positive disease received trastuzumab-based therapy.
discussion
HER2, human epidermal growth factor receptor-2; ER, estrogen receptor; PgR, progesterone receptor. *P value <0.05, statistically significant.
The role of serum HER2 ECD in patients with early breast cancer as prognostic factor is not well defined. The percentage of elevated serum levels of HER2 ECD are extremely variable in early breast cancer patients at the time of diagnosis. In our study, high HER2 ECD levels were observed in 9.0% of early breast cancer patients, similar to the results reported by other authors [22–24]. To measure the levels of HER2 ECD, we used both the automated and manual ELISA methods which are standardized, sensitive, reproducible and appropriate for routine analysis. Both assays utilize the same two antibodies (NB-3 and TA-1) that are directed against different epitopes of the HER2 ECD and both use a value of 15 ng/ml as cut-off. We showed that there was a strong correlation between these two ELISA methods, as reported by others [25, 26]. Both methods have been cleared by the Food and Drug Administration for use in
Figure 2. Kaplan–Meier estimates for disease-free survival according to serum human epidermal growth factor receptor-2 extracellular domain (HER2 ECD) levels.
Volume 19 | No. 5 | May 2008
doi:10.1093/annonc/mdm585 | 887
original article
Annals of Oncology
Figure 3. Kaplan–Meier estimates for disease-free survival according to tissue human epidermal growth factor receptor-2 (HER2) expression
the management and monitoring of women with metastatic breast cancer besides the automated Immuno-1 HER2 test. High HER2 EDC levels were associated with shorter DFS in our series of early breast cancer patients and according to Cox regression analysis, patients with elevated HER2 ECD showed a relative risk of 3.25 for reduced DFS. Elevated HER2 ECD levels were also correlated with the risk of visceral recurrence (66.6% versus 40%), in accordance with observations already reported, analyzing tissue HER2 overexpression in early breast cancer patients [27]. Few trials have evaluated the prognostic role of serum HER2 ECD in early breast cancer. The results of our prospective study are in agreement with two retrospective studies [28, 24] that showed a shorter DFS in patients with elevated serum HER2 ECD levels. The adverse prognostic value of elevated HER2 ECD levels in early breast cancer could reflect the presence of micrometastases or high HER2 cleavage and shedding with the production of truncated cell-associated fragments that contain the signaling kinase domain activated in absence of the ECD [29]. Therefore, these tumors with a deregulated growth-promoting pathway could have a more aggressive behavior. Moreover, in our study, elevated levels of HER2 ECD were associated with several factors related to tumor aggressiveness, such as stage III, lymph node involvement, poor histological differentiation, ER negativity and PgR negativity (Table 3). The correlation of tissue HER2 status (detected by IHC and/or FISH) with serum HER2 ECD levels was also analyzed. High serum HER2 ECD levels were observed in 38.1% (16 of 42) of patients with tissue HER2 positivity and in only 3.3% (7 of 214) of the patients with tissue HER2 negativity, with a moderate concordance (87.1%) as reported by others [23]. The targeted molecular domains and protein function are different: the IHC method recognizes
888 | Ludovini et al.
Table 5. Univariate and multivariate analyses for variables considered for DFS (Cox proportional hazards regression model) DFS Hazard ratio
Univariate analysis Stage (III versus I and II) Lymph nodes status (positive versus negative) PgR (cut-off ‡ 10%) (negative versus positive) Serum HER2 ECDa (high versus low) HER2 tissueb (positive versus negative) Multivariate analysis Serum HER2 ECDa (high versus low) Lymph nodes status (positive versus negative)
95% confidence interval
P value
5.02 4.66
2.32–10.90 1.96–11.10
<0.0001* 0.001*
0.39
0.17–0.89
0.026*
6.46
2.75–15.17
<0.0001*
2.14
0.91–5.04
0.082*
3.25
1.34–7.88
0.009*
2.9
1.01–8.29
0.048*
a
HER2 ECD: high (‡15 ng/ml) versus low (<15 ng/ml). Positive tissue HER2 status: IHC 3+, IHC 2+ and FISH amplified. DFS, disease-free survival; PgR, progesterone receptor; HER2, human epidermal growth factor receptor-2; ECD, extracellular domain; FISH, fluorescence in situ hybridization; IHC, immunohistochemistry. *P value <0.05, statistically significant. b
intracytoplasmic domains of the entire HER2 receptor; the serum HER2 assay measures the ECD of the HER2 receptor by using two mAbs recognizing two independent epitopes [30].
Volume 19 | No. 5 | May 2008
Annals of Oncology
In conclusion, the ELISA method for measuring the circulating HER2 ECD is a useful tool to obtain the real-time status of HER2 and both manual and automated methods may be used. Our findings indicate that the determination of serum HER2 ECD might aid to assess the HER2 status, in addition to the conventional assays carried out on tumor specimens. Due to the correlation between HER2 ECD levels and HER2 expression in tumoral tissue, patients could be considered for trastuzumab therapy only if tissue is not available, on the basis of high HER2 ECD levels. Elevated HER2 ECD levels may also represent a subgroup of HER2-positive tumors with a higher level of HER2 cleavage and shedding. In vitro studies have shown that high amounts of soluble HER2 ECD may neutralize the biological activity of anti-HER2 antibodies by forming immune complexes and blocking the accessibility of the antibodies to the tumor [31]. This subgroup of tumors may have a more aggressive clinical course and the availability of a noninvasive, repeatable and reproducible technique to measure circulating HER2 ECD appears to be a useful tool for identifying high-risk breast cancer patients. In conclusion, to our knowledge, this is the only prospective trial that has evaluated the prognostic role HER2 ECD in patients with operable breast cancer. A significant association with other poor prognostic factors has been found as well as with a worse DFS; however, due to limited percentage of patients with elevated serum levels of HER2 ECD in this series, these results must be interpreted cautiously and need to be confirmed in large prospective trials with serial measurements. It could also be worthwhile to evaluate in patients with high serum HER2 ECD levels the efficacy of different treatments targeting HER2 (receptor tyrosine kinase inhibitors) in comparison with trastuzumab.
funding Bayer Diagnostics; Umbria Association Against Cancer; ‘Conoscere per Vincere’ charities. This work was supported in part (reagents for HER2 ECD analysis) by Bayer Diagnostics (Cambridge, MA 02142-1168, USA), by the Umbria Association Against Cancer (AUCC) and by ‘‘Conoscere per Vincere’’ charities.
acknowledgements We thank the patients who participated in this study.
references 1. Aaronson SA. Growth factors and cancer. Science 1991; 254: 1146–1153. 2. Slamon DJ, Clark GM, Wong SG et al. Human breast cancer: correlation of relapse and survival with amplification of the HER2/neu oncogene. Science 1987; 235: 177–182. 3. Carlomagno C, Perrone F, Gallo C et al. C-erbB-2 overexpression decreases the benefit of adjuvant tamoxifen in early stage breast cancer without axillary lymph node metastases. J Clin Oncol 1996; 14: 2702–2708.
Volume 19 | No. 5 | May 2008
original article 4. Gusterson BA, Gelber RD, Goldhirsch A et al. Prognostic importance of c-erbB-2 expression in breast cancer. J Clin Oncol 1992; 10: 1049–1056. 5. Stal O, Sullivan S, Wingren S et al. C-erbB-2 expression and benefit from adjuvant chemotherapy and radiotherapy of breast cancer. Eur J Cancer 1995; 31: 2185–2190. 6. Muss HB, Thor AD, Berry DA et al. C-erbB-2 expression and response to adjuvant therapy in women with node-positive early breast cancer. N Engl J Med 1994; 330: 1260–1266. 7. Paik S, Bryant J, Park C et al. ErbB-2 and response to doxorubicin in patients with axillary lymph node-postitive, hormone receptor-negative breast cancer. J Natl Cancer Inst 1998; 90: 1361–1370. 8. Hudis CA. Trastuzumab-mechanism of action and use in clinical practice. N Engl J Med 2007; 357: 39–51. 9. Muller V, Thomssen C, Karakas C et al. Quantitative assessment of HER2/neu protein concentration in breast cancer by enzyme-linked immunosorbent assay. Int J Biol Markers 2003; 18: 13–20. 10. Krainer M, Brodowicz T, Zeillinger R et al. Tissue expression and serum levels of HER2/neu in patients with breast cancer. Oncology 1997; 54: 475–481. 11. Molina R, Jo J, Filella X et al. C-erbB-2 CEA and Ca 15.3 serum levels in the early diagnosis recurrence of breast cancer patients. Anticancer Res 1999; 19(4A): 2551–2555. 12. Kong SY, Nam BH, Lee KS et al. Predicting tissue HER2 status using serum HER2 levels in patients with metastatic breast cancer. Clin Chem 2006; 52: 1510–1515. 13. Beltran M, Colomer R, Lipton A et al. Does HER2 status predict only a decreased response to hormone therapy in advanced breast cancer, or does it also predict the extent of metastatic disease? J Clin Oncol 2002; 20: 4605–4606. 14. Fehm T, Gebauer G, Jager W. Clinical utility of serial serum c-erbB-2 determinations in the follow-up of breast cancer patients. Breast Cancer Res Treat 2002; 75: 97–106. 15. Lipton A, Ali SM, Leitzel K et al. Elevated serum HER2/neu level predicts decreased response to hormone therapy in metastatic breast cancer. J Clin Oncol 2002; 20: 1467–1472. 16. Perez EA, Geeraerts L, Suman VJ et al. A randomized phase II study of sequential docetaxel and doxorubicin/cyclophosphamide in patients with metastatic breast cancer. Ann Oncol 2002; 13: 1225–1235. 17. Colomer R, Llombart-Cussac A, Lluch A et al. Biweekly paclitaxel plus gemcitabine in advanced breast cancer: phase II trial and predictive value of HER2 extracellular domain. Ann Oncol 2004; 15: 201–206. 18. Fornier MN, Seidman AD, Schwartz MK et al. Serum HER2 extracellular domain in metastatic breast cancer patients treated with weekly trastuzumab and paclitaxel: association with HER2 status by immunohistochemistry and fluorescence in situ hybridization and with response rate. Ann Oncol 2005; 16: 234–239. 19. Yuan P, Xu BH, Chu DT. Correlation between serum HER-2 oncoprotein and patients with breast cancer. Chin Med Sci J 2004; 19(5): 883–890. 20. Singletary SE, Allred C, Ashley P et al. Revision of the American Joint Committee on Cancer staging system for breast cancer. J Clin Oncol 2002; 20: 3628–3636. 21. Schwartz MK, Smith C, Schwartz DC et al. Monitoring therapy by serum HER2/neu. Int J Biol Markers 2000; 15: 324–329. 22. Isola JJ, Holli K, Oksa H et al. Elevated erbB-2 oncoprotein levels in preoperative and follow-up serum samples define an aggressive disease course in patients with breast cancer. Cancer 1994; 73: 652–658. 23. Molina R, Filella X, Zanon G et al. Prospective evaluation of tumor markers (c-erbB-2 oncoprotein, CEA and CA 15.3) in patients with locoregional breast cancer. Anticancer Res 2003; 23: 1043–1050. 24. Saghatchian M, Guepratte S, Hacene K et al. Serum HER-2 extracellular domain: relationship with clinicobiological presentation and prognostic value before and after primary treatment in 701 breast cancer patients. Int J Biol Markers 2004; 19: 14–22. 25. Cook GB, Neaman IE, Goldblatt JL et al. Clinical utility of serum HER2/neu testing on the Bayer Immuno 1 automated system in breast cancer. Anticancer Res 2001; 21: 1465–1470.
doi:10.1093/annonc/mdm585 | 889
original article 26. Dittadi R, Zancan M, Perasole A, Gion M. Evaluation of HER-2/neu in serum and tissue of primary and metastatic breast cancer patients using an automated enzyme immunoassay. Int J Biol Markers 2001; 16: 255–261. 27. Kallioniemi OP, Holli K, Visakorpi T et al. Association of c-erbB-2 protein overexpression with high rate of cell proliferation, increased risk of visceral metastases and poor long-term survival in breast cancer. Int J Cancer 1991; 49: 650–655. 28. Mehta RR, McDermott JH, Heiken TJ et al. Plasma c-erbB-2 levels in breast cancer patients: prognostic significance in predicting response to chemotherapy. J Clin Oncol 1998; 16: 2409–2416.
890 | Ludovini et al.
Annals of Oncology
29. Molina MA, Codony-Servat J, Albanell J et al. Trastuzumab (Herceptin), a humanized antiHER2 receptor monoclonal antibody, inhibits basal and activated HER2 ectodomain cleavage in breast cancer cells. Cancer Res 2001; 61: 4744–4749. 30. Carney WO, Hamer PJ, Petit D et al. Detection and quantitation of the human neu oncoprotein. J Tumour Mark Oncol 1991; 6: 53–72. 31. Brodowicz T, Wiltschke C, Budinsky AC et al. Soluble HER2/neu neutralizes biologic effects of anti-HER2/neu antibody on breast cancer cells in vitro. Int J Cancer 1997; 73: 875–879.
Volume 19 | No. 5 | May 2008