|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 25, No 19 (July 1), 2007: pp. 2656-2663 © 2007 American Society of Clinical Oncology. DOI: 10.1200/JCO.2006.08.6850 Neuregulin Expression Modulates Clinical Response to Trastuzumab in Patients With Metastatic Breast Cancer
From the Centro de Investigación del Cáncer–Instituto de Biología Molecular y Celular del Cáncer, Consejo Superior de Investigaciones Científicas–Universidad de Salamanca, and Hospital Universitario de Salamanca, Salamanca, Spain Address reprint requests to Atanasio Pandiella, MD, Centro de Investigación del Cáncer, Campus Miguel de Unamuno, 37007-Salamanca, Spain; e-mail: atanasio{at}usal.es
Purpose Human epidermal growth factor receptor 2 (HER-2) overexpression has been associated with the genesis and progression of a subset of breast cancers. The function of HER-2 may be upregulated by overexpression or by the availability of neuregulins (NRGs), a group of transmembrane growth factors. Transmembrane NRGs strongly activated HER-2 and cell proliferation in breast cancer cells that did not overexpress HER-2, and treatment with trastuzumab prevented the proliferative action of transmembrane NRG. This raised the relevant clinical question of whether patients considered as HER-2 negative, but expressing transmembrane NRG, may benefit from treatment with trastuzumab.
Patients and Methods MCF7 cells expressing transmembrane NRG (MCF7-NRG
Results Trastuzumab inhibited tumor growth in mice injected with MCF7-NRG Conclusion We suggest that the spectrum of patients who may benefit from trastuzumab-based therapies may be widened to include patients with metastatic breast cancer without HER-2 amplification but who express transmembrane NRGs.
The human epidermal growth factor receptor (ErbB/HER) tyrosine kinases and their ligands are involved in numerous biologic and pathologic processes including cancer.1 The following four ErbB receptors have been described in mammals: ErbB1 (epidermal growth factor receptor or HER-1), ErbB2 (HER-2 or neu), ErbB3 (HER-3), and ErbB4 (HER-4).2,3 Activation of these receptors may occur by the following three different mechanisms4-6: specific HER ligands; overexpression; or molecular alterations such as point mutations or truncations. ErbB2/HER-2 overexpression has been implicated in the genesis and progression of a subset of breast and ovarian tumors.7,8 This led to the development of therapies to decrease the activation of this receptor. One of these therapeutic developments is trastuzumab (Herceptin; Genentech, South San Francisco, CA), a monoclonal antibody that has demonstrated clinical benefit.9,10 One important problem of trastuzumab therapy is the selection of responsive patients based on HER-2 overexpression, which is assessed using immunohistochemical techniques frequently combined with fluorescent in situ hybridization (FISH) to determine the level of amplification. Immunohistochemical detection of HER-2 was initially based on the use of different anti-HER-2 antibodies, which caused certain discrepancy in the evaluation of HER-2 positivity. More recently, availability of diagnostic kits based on well-established immunohistochemical staining and evaluation methods (ie, DAKO HercepTest; DAKO, Carpinteria, CA) has favored the standardization of HER-2 detection. Yet, these criteria for patient selection are insufficient because only 30% of patients respond to trastuzumab monotherapy.9 An alternative mode of HER-2 receptor activation is the presence of ligands, such as neuregulins (NRGs; also termed heregulins).11 The NRGs are produced as transmembrane ligands, known as proNRGs, that can be released as soluble factors by the action of cell surface proteases12,13 (Fig 1A). Increasing evidence indicates that NRGs may play a relevant role in breast cancer. Expression of NRG in the mammary gland induces adenocarcinomas in animal models14 and favors metastatic spread of breast cancer cells.15 Expression of transmembrane NRGs in breast cancer cells activates HER-2 and favors their proliferation in vitro.16-18 This activation occurs in the absence of HER-2 overexpression and is highly sensitive to trastuzumab.18 These findings raised the question of whether trastuzumab could be clinically useful in patients lacking HER-2 amplification but expressing transmembrane NRG.19 In this report, we have analyzed NRG expression in samples from patients with breast cancer and have retrospectively studied a potential correlation between transmembrane NRG expression and clinical response. Our results show that transmembrane NRG expression is frequent in breast cancer patients and, in the absence of HER-2 amplification, is associated with objective clinical response to trastuzumab. Therefore, the presence of transmembrane NRG in patients without HER-2 amplification may indicate trastuzumab sensitivity and, thus, could open a new therapeutic option for this subset of patients.
Patient Selection and Statistical Analyses One hundred fifty-one samples from 124 patients who were treated at the University Hospital of Salamanca from 2000 to 2005 were studied for NRG expression. Statistical analyses were performed in 63 patients for whom enough demographic characteristics or clinical follow-up data were available. Because some of these patients were evaluated for HER-2 expression before the HercepTest assay (DAKO), a retrospective review of their HER-2 status was performed. Data were tabulated into a Microsoft Excel worksheet (Microsoft, Redmond, WA) and exported to the SPSS 12.0 statistical suite (SPPS Inc, Chicago, IL). The Spearman's  2 test was used for correlation analyses, and a P < .05 was used as the cutoff for decisions of statistical significance. 2 and Fisher's exact tests were used to assess correlation of two dichotomous variables. In the case of one ordinal variable and one dichotomous variable, comparison was performed using the Mann-Whitney U test. Kaplan-Meier survival analyses were carried out for both overall survival (OS) and time to disease progression (TDP). Log-rank statistics were also calculated. The Student's t test (two sided) was used to compare tumor sizes in mice.
NRG Expression For the detection of NRG expression by Western blot, the tumors were minced, washed with phosphate-buffered saline, and homogenized with ice cold lysis buffer12 with a tight-fitting Dounce homogenizer. This homogenate was centrifuged at 10,000 x g for 20 minutes at 4°C, and the supernatants were transferred to new tubes. The samples (60 µg) were processed for Western blotting as previously described.20
HER-2 and Hormonal Status Assessment
In Vivo Assessment of Tumor Growth
Transmembrane NRG Confers Proliferation Advantage In Vivo Formerly, we reported that expression of transmembrane NRG in MCF7 breast cancer cells increased their in vitro proliferation.18 To analyze whether such a proliferation advantage could be reproduced in vivo, MCF7 cells expressing the transmembrane form of NRG  2c (MCF7-NRG2c), wild-type MCF7 cells, or MCF7 cells overexpressing HER-2 (MCF7-HER-2) were injected into the mammary fat pad of female nude mice, and the size of the masses were measured weekly. All of the mice overexpressing HER-2 developed tumors by 8 weeks after injection (Fig 1B; Table 1). However, only 31% of mice injected with wild-type MCF7 cells developed tumors. Expression of transmembrane NRG resulted in a substantial percentage of mice (87%) bearing tumor masses. Transmembrane NRG was expressed in the tumoral masses isolated from mice injected with MCF7-NRG2c cells (Fig 1D) and provoked tyrosine phosphorylation of HER-2 (Fig 1C). Tyrosine phosphorylation of HER-2 was high in tumors created by injection of MCF7-HER-2 cells. These data indicate that expression of transmembrane NRG also confers a proliferation advantage in vivo. Because we formerly reported that trastuzumab prevented in vitro proliferation of MCF7-NRG2c cells,18 we also explored the in vivo sensitivity to trastuzumab of tumors created by MCF7-NRG2c cells. As shown in Figure 1E, trastuzumab hampered tumor growth of MCF7-NRG2c–derived tumors.
NRG Expression in Breast Cancer Samples We studied the expression of transmembrane NRG in patients with breast cancer tumors by immunohistochemistry with the anti-proNRG antibody. Because we did not have previous experience of NRG expression assessment by immunohistochemistry using this antibody, a training set of 151 samples derived from 124 patients was analyzed. NRG expression was seen as a focal or diffuse staining of the tumor cell cytoplasms; some staining was seen in normal fibroblasts surrounding the infiltrative tumor nests (Fig 2A). Using a two-tier distribution (low/high) in which values greater than the median value of 5.5 were considered positive, high levels of NRG were observed in half of the patients. In the subset of patients (n = 32) with metastatic breast cancer for whom data of response to trastuzumab were available, 24 patients (75%) had high levels of NRG, whereas only eight patients (25%) showed low levels.
We also analyzed the presence of NRG forms in 11 of the breast tumor samples by Western blotting, using tissue derived from normal breast glandular tissue as a control. This approach is complementary to the immunohistochemical analysis because it allows for the distinction between different molecular forms of transmembrane NRGs. Western blotting using the anti-proNRG antiserum revealed the presence of several reactive bands of 150 to 200, 45, 35, 25, and 21 Kd in the breast tumor samples (Fig 2C, top panel). That these bands were specifically recognized by the anti-proNRG antiserum was demonstrated by preincubation of the anti-proNRG antibodies with an excess of the peptide against which the antibody had been raised (Fig 2C, second panel). The 45-Kd band was also detected in the normal breast tissue sample. In these samples, we also performed Western blotting analyses of HER-2, HER-3, and HER-4 receptors, together with an assessment of tyrosine phosphorylation. Most of the tumor samples contained higher levels of the HER receptors than the normal tissue. These samples also contained higher levels of tyrosine phosphorylated proteins in the 150- to 220-Kd region. To investigate whether HER-2 was tyrosine phosphorylated, we explored the level of HER-2 tyrosine phosphorylation using phosphospecific antibodies. As a representative sample, we used the BT149 tumor sample, whose level of HER-2 was low and contained high levels of the 45-Kd NRG form. Tyrosine phosphorylation of HER-2 in the BT149 sample was detected, especially with the antibodies that recognized phosphorylation at tyrosines 1248 and 1221/1222 (Fig 2D). As a control for the activation of HER-2 in this experiment, we used MCF7 cells, whose amount of HER-2 is considered low to normal,21,22 treated with soluble NRG.
NRG Expression and Breast Cancer Prognostic Factors
Correlation of NRG Expression With Response
Correlation of NRG Expression With Time to Progression Additional parameters to evaluate the efficacy of a given treatment in metastatic breast cancer included TDP and OS. We studied TDP in the 12 HER-2–negative patients treated with trastuzumab-based therapy and observed that patients with high NRG levels had a higher TDP compared with patients with low NRG levels (log-rank test, 5.73; P = .0167; Fig 3C). As a control group, we studied TDP in the subgroup of patients with HER-2 overexpression (Table 2). In this subgroup of patients, no differences were observed (log-rank test, 1.06; P = .303; Fig 3D). To further analyze whether the chemotherapy could influence these results, we studied the subgroup of 10 patients with HER-2-negative tumors (Table 2) who were treated only with taxanes and trastuzumab as first-line treatment. We observed a clear difference in TDP for patients with high levels of NRG (log-rank test, 5.54; P = .0186; Fig 4A). When selecting the subgroup of patients with HER-2 overexpression treated only with taxanes plus trastuzumab (seven patients), there was no difference in TDP depending on NRG levels (log-rank test, 1.74; P = .186; Fig 4B). However, the limited number of patients in this subgroup (only seven patients) makes the P value inconclusive.
Correlation of NRG Expression and Survival If NRG could predict the response to trastuzumab in patients without HER-2 overexpression, then patients with high levels of NRG treated with trastuzumab could have a longer OS. In the subgroup of 12 patients with HER-2–negative tumors who had been treated with a trastuzumab-based chemotherapy (Table 2), an increase in OS was observed in patients with high expression of NRG (log-rank test, 4.82; P = .0281; Fig 5A). In the subgroup of 10 HER-2–negative patients treated with taxanes and trastuzumab as first-line treatment, we also observed an improvement in OS in patients with high levels of NRG (log-rank test, 6.27; P = .0122; Fig 5B). However, patients with HER-2 overexpression and high NRG levels did not show any improvement in OS compared with patients with low levels (log-rank test, 1.08; P = .298; Fig 5C). Analogous analyses in the subgroup of patients with HER-2 overexpression who were only treated, as a first-line therapy, with taxanes and trastuzumab (seven patients) showed no differences in survival regardless of the NRG level (log-rank test, 0.64; P = .432). These results suggest that expression of transmembrane NRG can predict response to trastuzumab-based therapies in tumors with normal or low HER-2 expression.
This study was initiated with the purpose of evaluating whether expression of transmembrane NRG could facilitate response to trastuzumab in patients lacking overexpression of HER-2. This idea was supported by preclinical in vitro data that showed that transmembrane NRG was a potent inducer of cell proliferation in MCF7 breast cancer cells, which do not overexpress HER-2.18 Furthermore, this proliferation potential was inhibited by treatment with trastuzumab.18 The results we present here extend our former in vitro studies and agree with reports that indicated increased tumorogenic and metastatic potential of MCF7 cells expressing NRG.15,16 The design of the clinical part of the study merits some considerations. First, it was important to carry out a retrospective study because we already had access to the pathologic and clinical material, and more importantly for the objective of this study, we expected to find patients who received trastuzumab in the absence of HER-2 overexpression, as per current criteria. This was a critical aspect of the study because we wanted to be able to evaluate the potential effectiveness of trastuzumab in patients who expressed transmembrane NRG but who were HER-2 negative. Using a specific antibody, we detected transmembrane NRG in a substantial number of patient samples. Of 32 patients with metastatic breast cancer treated with trastuzumab and after re-evaluation of their HER-2 status by DAKO HercepTest and FISH, 12 patients were HER-2 negative. In this subset of patients, we analyzed a potential correlation between transmembrane NRG expression and response, TDP, and OS. A correlation between transmembrane NRG expression and TDP or OS was found in patients with low levels of HER-2 who were treated with trastuzumab, whereas no correlation was found in patients with HER-2 overexpression. When we focused on response rates, no difference was observed, probably because of the limited number of patients. However, FISH-negative patients with high NRG levels responded better than patients with low NRG levels. These results open new avenues for future research into the clinical use of receptor-targeted therapies. One of the consequences of our work is that the spectrum of patients who may benefit from these therapies may extend to those who express transmembrane growth factors but do not overexpress the specific receptors. Some scattered precedents on this have been reported. A study on the activity of trastuzumab in breast cancer patients indicated that 10% of HER-2-negative patients had clinical benefit from trastuzumab monotherapy.9 A recent report in colon cancer patients treated with cetuximab, a monoclonal antibody that interacts with the extracellular domain of HER-1, also demonstrated clinical effectiveness in patients who were HER-1 negative by immunohistochemistry and were treated with that antibody.23 Our work also indicates that proper molecular pathologic analysis of breast cancer patients should include the determination of HER-2 levels as well as other parameters such as ligand expression and phosphorylation status of the HER receptors. Because of the limited number of patients analyzed, our study must be considered exploratory. A larger clinicopathologic study considering these parameters may help in establishing the conditions for a better selection of patients susceptible to clinical benefit on trastuzumab treatment.
The author(s) indicated no potential conflicts of interest.
Conception and design: Enrique de Alava, Mar Abad, Alberto Ocaña, Juan Carlos Montero, Azucena Esparis-Ogando, Atanasio Pandiella Financial support: Enrique de Alava, Azucena Esparis-Ogando, Atanasio Pandiella Provision of study materials or patients: Alberto Ocaña, Mar Abad, Juan Carlos Montero, Azucena Esparis-Ogando, César A. Rodríguez, Juan J. Cruz, Atanasio Pandiella Collection and assembly of data: Enrique de Alava, Alberto Ocaña, Mar Abad, Juan Carlos Montero, Azucena Esparís-Ogando, César A. Rodríguez, Ana P. Otero, Teresa Hernández, Atanasio Pandiella Data analysis and interpretation: Enrique de Alava, Alberto Ocaña, Mar Abad, Juan Carlos Montero, Azucena Esparís-Ogando, César A. Rodríguez, Teresa Hernández, Ana P. Otero, Atanasio Pandiella Manuscript writing: Enrique de Alava, Mar Abad, Alberto Ocaña, Atanasio Pandiella Final approval of manuscript: Enrique de Alava, Alberto Ocaña, Mar Abad, Juan Carlos Montero, Azucena Esparís-Ogando, César A. Rodríguez, Ana P. Otero, Teresa Hernández, Juan J. Cruz, Atanasio Pandiella
We thank Maribel Ruiz, MD, Rafael Aparicio, MD, and Beatriz Torío, MD, Hospital ‘Río Carrión,’ Palencia, Spain, as well as Angeles Torres, MD, Hospital ‘Río Hortega,’ Valladolid, Spain, for contributing breast cancer samples belonging to the training data set; and María del Carmen Rodríguez, Hospital Universitario de Salamanca, Salamanca, Spain, and Rosa García-Centeno, Hospital Clínico de Valladolid, Valladolid, Spain, for tumor procurement and performance of tissue microarrays.
Supported by Grant No. BMC2003-01192 from the Ministry of Science and Technology of Spain and by grants from the Autonomous Government of Castilla y León SAN191/SA06/06 to our Center's Tumor Bank and to the Regional Tumor Bank Network of Castilla y León. Our Cancer Research Institute receives support from the European community through the Regional Development Funding Program and from Cancer Center Network Program of the Instituto de Salud Carlos III (ISCIII). J.C.M. and A.E.-O. were supported by Asociacion Española Contra el Cancer (AECC) and ISCIII contracts, respectively. Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Hynes NE, Lane HA: ERBB receptors and cancer: The complexity of targeted inhibitors. Nat Rev Cancer 5:341-354, 2005[CrossRef][Medline] 2. Yarden Y, Sliwkowski MX: Untangling the ErbB signalling network. Nat Rev Mol Cell Biol 2:127-137, 2001[CrossRef][Medline] 3. Holbro T, Civenni G, Hynes NE: The ErbB receptors and their role in cancer progression. Exp Cell Res 284:99-110, 2003[CrossRef][Medline] 4. Ullrich A, Schlessinger J: Signal transduction by receptors with tyrosine kinase activity. Cell 61:203-212, 1990[CrossRef][Medline] 5. Cruz JJ, Ocana A, Barco ED, et al: Targeting receptor tyrosine kinases and their signal transduction routes in head and neck cancer. Ann Oncol 18:421-430, 2007 6. Hynes NE, Schlange T: Targeting ADAMS and ERBBs in lung cancer. Cancer Cell 10:7-11, 2006[CrossRef][Medline] 7. Mendelsohn J, Baselga J: The EGF receptor family as targets for cancer therapy. Oncogene 19:6550-6565, 2000[CrossRef][Medline] 8. Hynes NE, Stern DF: The biology of erbB-2/neu/HER2 and its role in cancer. Biochim Biophys Acta 1198:165-184, 1994[Medline] 9. Vogel CL, Cobleigh MA, Tripathy D, et al: Efficacy and safety of trastuzumab as a single agent in first-line treatment of HER2-overexpressing metastatic breast cancer. J Clin Oncol 20:719-726, 2002 10. Slamon DJ, Leyland-Jones B, Shak S, et al: Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 344:783-792, 2001 11. Falls DL: Neuregulins: Functions, forms, and signaling strategies. Exp Cell Res 284:14-30, 2003[CrossRef][Medline] 12. Montero JC, Yuste L, Díaz-Rodríguez E, et al: Differential shedding of transmembrane neuregulin isoforms by the tumor necrosis factor-alpha converting enzyme. Mol Cell Neurosci 16:631-648, 2000[CrossRef][Medline] 13. Massagué J, Pandiella A: Membrane-anchored growth factors. Annu Rev Biochem 62:515-541, 1993[CrossRef][Medline] 14. Krane IM, Leder P: NDF/heregulin induces persistence of terminal end buds and adenocarcinomas in the mammary glands of transgenic mice. Oncogene 12:1781-1788, 1996[Medline] 15. Atlas E, Cardillo M, Mehmi I, et al: Heregulin is sufficient for the promotion of tumorigenicity and metastasis of breast cancer cells in vivo. Mol Cancer Res 1:165-175, 2003 16. Tsai MS, Shamon-Taylor LA, Mehmi I, et al: Blockage of heregulin expression inhibits tumorigenicity and metastasis of breast cancer. Oncogene 22:761-768, 2003[CrossRef][Medline] 17. Menendez JA, Mehmi I, Lupu R: Trastuzumab in combination with heregulin-activated Her-2 (erbB-2) triggers a receptor-enhanced chemosensitivity effect in the absence of Her-2 overexpression. J Clin Oncol 24:3735-3746, 2006 18. Yuste L, Montero JC, Esparis-Ogando A, et al: Activation of ErbB2 by overexpression or by transmembrane neuregulin results in differential signaling and sensitivity to Herceptin. Cancer Res 65:6801-6810, 2005 19. Arteaga CL: Can trastuzumab be effective against tumors with low HER2/Neu (ErbB2) receptors? J Clin Oncol 24:3722-3725, 2006 20. Cabrera N, Díaz-Rodríguez E, Becker E, et al: TrkA receptor ectodomain cleavage generates a tyrosine-phosphorylated cell-associated fragment. J Cell Biol 132:427-436, 1996 21. Montero JC, Rodriguez-Barrueco R, Yuste L, et al: The extracellular linker of pro-neuregulin-alpha 2c is required for efficient sorting and juxtacrine function. Mol Biol Cell 18:380-393, 2007 22. Agus D, Akita R, Fox W, et al: Targeting ligand-activated ErbB2 signaling inhibits breast and prostate tumor growth. Cancer Cell 2:127-137, 2002[CrossRef][Medline] 23. Chung KY, Shia J, Kemeny NE, et al: Cetuximab shows activity in colorectal cancer patients with tumors that do not express the epidermal growth factor receptor by immunohistochemistry. J Clin Oncol 23:1803-1810, 2005 Submitted August 13, 2006; accepted April 10, 2007.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2007 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
