Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
European Journal of Cancer 53 (2016) 42e50 Available online at www.sciencedirect.com ScienceDirect journal homepage: www.ejcancer.com Original Research Extra-gain of HER2-positive cases through HER2 reassessment in primary and metastatic sites in advanced gastric cancer with initially HER2-negative primary tumours: Results of GASTric cancer HER2 reassessment study 1 (GASTHER1) Sook Ryun Park a,1, Young Soo Park b,1, Min-Hee Ryu a,1, Baek-Yeol Ryoo a, Chang Gok Woo b, Hwoon-Yong Jung c, Jeong Hoon Lee c, Gin Hyug Lee c, Yoon-Koo Kang a,* a Department of Oncology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea Department of Pathology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea c Department of Gastroenterology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea b Received 5 May 2015; received in revised form 9 September 2015; accepted 19 September 2015 Available online 13 December 2015 KEYWORDS HER2; Gastric cancer; Intratumoural heterogeneity; Primary lesion; Metastatic lesion; Conversion; Reassessment Abstract Purpose: The intratumoural heterogeneity of human epidermal growth factor receptor 2 (HER2) expression in gastric cancer is a major challenge when identifying patients who might benefit from HER2-targeting therapy. We investigated the significance of reevaluation of HER2 status in primary sites and metastatic or recurrent sites in advanced gastric cancer patients whose primary tumours were initially HER2-negative. Patients and methods: In part I of this study, we evaluated the significance of repeat endoscopic biopsy in unresectable or metastatic gastric cancer patients whose tumours were initially HER2negative. In part II, we examined the HER2 positivity rate in metastatic or recurrent sites in patients whose primary tumours were HER2-negative in biopsy or surgical specimens. Results: In part I (n Z 183), we identified patients with HER2-positive tumours for a rescued HER2 positivity rate of 8.7% (95% confidence interval [CI], 4.6e12.8%) that was associated with tumour location (diffuse stomach versus other Z 0% versus 11.7%, P Z 0.013), Bormann type (IV versus others Z 0% versus 11.7%, P Z 0.013), and initial biopsy HER2 immunohistochemistry score (0 versus 1 versus 2 Z 6.7% versus 15.4% versus 25.0%, P Z 0.028). Part II (n Z 175) * Corresponding author: Department of Oncology, Asan Medical Center, University of Ulsan College of Medicine, 88, Olympic-ro 43-gil, Songpa-gu, Seoul, 138-736, Republic of Korea. Tel.: þ82 2 3010 3230; fax: þ82 2 3010 8772. E-mail address: [email protected] (Y.-K. Kang). 1 The first three authors contributed equally to this work. http://dx.doi.org/10.1016/j.ejca.2015.09.018 0959-8049/ª 2015 Elsevier Ltd. All rights reserved. S.R. Park et al. / European Journal of Cancer 53 (2016) 42e50 43 resulted in HER2 positivity of 5.7% (95% CI 2.3e9.1%) that was significantly associated with metastatic site (liver versus others Z 17.2% versus 3.4%, P Z 0.012). When compared with a historical control that showed HER2 positivity on initial assessment, patients who had rescued HER2 positivity had similar treatment benefits from trastuzumab-containing first-line chemotherapy. Conclusion: Repeat HER2 assessment in primary and metastatic or recurrent sites is recommended in patients with advanced gastric cancer whose primary tumour is initially HER2negative. ª 2015 Elsevier Ltd. All rights reserved. 1. Introduction Nowadays, new agents that target critical oncogenic pathways can improve treatment outcomes, and management based on the molecular characteristics of tumours is becoming routine. Human epidermal growth factor receptor 2 (HER2) is a transmembrane tyrosine kinase receptor and a well-established therapeutic target in breast and gastric cancer [1e3]. A randomised phase III study (Trastuzumab for GAstric Cancer [ToGA]) demonstrated the clinical benefit of combining trastuzumab, an anti-HER2 monoclonal antibody, with chemotherapy in patients with unresectable or metastatic HER2-positive gastric or gastroesophageal junction (GEJ) cancer [3]. The survival benefit compared with chemotherapy alone was more prominent in patients with immunohistochemistry (IHC) scores of 3þ or 2þ with gene amplification as evidence by fluorescence in situ hybridization (FISHþ) [3]. Based on these results, trastuzumab was integrated into the standard chemotherapy for HER2-positive gastric cancer [4]. Because of its therapeutic implications, the correct identification of HER2 positivity on tissue specimens is critical. Unlike breast cancers, gastric cancers show frequent incomplete membrane staining, and heterogeneity of HER2 status is common, which makes the diagnosis of HER2 overexpression difficult. Therefore, the ToGA study employed a new set of IHC scoring criteria that considered as positive basolateral or bilateral membranous reactivity as well, and used a 10% cutoff instead of the 30% used in breast cancers which was also modified to the 10% in the 2013 HER2 testing guideline for breast cancer to reduce false-negative results [5,6]. Nonetheless, intratumoural heterogeneity of HER2 positivity may still result in inaccurate assessment of HER2 status, which presents a major challenge in identifying patients who could benefit from HER2targeting therapy. Also important is that HER2 status is usually assessed in primary gastric tumours that is used to guide therapy in the metastatic or recurrent setting, but discordance between HER2 status in primary and metastatic tumours that develop synchronously or metachronously may be misleading. The purpose of this prospective study was to determine how many additional gastric cancer patients with HER2-positive tumours could be found through reassessment of HER2 in 1) primary tumour sites by repeat endoscopic biopsy and 2) metastatic or recurrent tumour sites in patients whose tumours were HER2negative in the initial evaluation for primary tumours, and to identify possible baseline factors associated with rescued HER2 positivity through reassessment. 2. Methods 2.1. Study design The study consisted of two parts, each with a separate patient population. In part I, we evaluated the significance of a repeat endoscopic biopsy in unresectable or metastatic gastric cancer patients whose tumours were HER2-negative in the initial endoscopic biopsy, and in part II we examined the HER2 positivity rate in metastatic or recurrent sites in patients whose primary tumours were HER2-negative in biopsy or surgical specimens. Patients were eligible for this study if they were over 18 years old with histologically proven unresectable, metastatic, or recurrent (recurrent disease was included only in part II) gastric or GEJ adenocarcinoma and no prior chemotherapy in the palliative setting. Initial HER2 assessment of the primary gastric tumour had to be negative in endoscopic biopsy in part I and in the endoscopic biopsy or surgical specimen in part II. HER2 positivity was defined as IHC 3þ or IHC 2þ plus FISHþ [5,7]. In part I, immediately after obtaining HER2-negative results in an initial biopsy, a repeat gastroendoscopic biopsy was performed at six different sites of the primary tumour lesions. To compare outcomes of trastuzumab-containing first-line chemotherapy between patients with HER2 positivity detected on the initial assessment in primary tumours and those with rescued HER2 positivity detected on the repeat assessment in primary or metastatic tumours, we used historical control data of unresectable, metastatic, or recurrent gastric cancer patients who had HER2 positivity on the initial endoscopic biopsy or gastrectomy specimen and were treated with trastuzumab-containing first-line chemotherapy between April 2009 and March 2013 at Asan Medical 44 S.R. Park et al. / European Journal of Cancer 53 (2016) 42e50 Center. The Center’s Institutional Review Board approved the study protocol, and all patients provided written informed consent. 2.2. Immunohistochemistry staining and fluorescence in situ hybridization HER2 IHC was performed on formalin fixed, paraffin embedded tissue using Pathway (monoclonal antibody, 4B5; Ventana Medical System, Tucson, AZ) with an automated staining device according to the manufacturer’s US Food and Drug Administration-approved procedures (see Appendix for more detail). HER2 protein expression was scored on a scale of 0 to 3 using the gastric cancer consensus panel recommendations [5,7]. FISH was performed only in cases with IHC 2þ using the Abbott PathVysion HER2 DNA Probe Kit protocol (Abbott Laboratories, Abbott Park, Des Plaines, IL) according to the manufacturer’s instructions (see Appendix for more detail). HER2 gene amplification status was evaluated by counting signals in 20 nonoverlapping tumour cells with the highest gene count. The interpretation followed the criteria of the American Society of Clinical Oncology (ASCO)/College of American Pathologists (CAP) guidelines [8]; negative for gene amplification if the HER2/chromosome enumeration probe (CEP)17 ratio was <1.8, positive if the HER2/CEP17 ratio was >2.2, and equivocal if the HER2/CEP17 ratio was 1.8e2.2 where we counted 20 additional tumour cells and considered the case positive if the HER2/CEP17 ratio was 2.0. 2.3. Statistical analysis The primary end-point was the rate of HER2 positivity that was converted through reassessment following initial HER2 negativity. Assuming that the rate of HER2 positivity meaningful in reassessment was 5% with a 95% confidence interval (CI) of 0.023e0.097, we needed a sample size of 179 patients in each study part. We used univariate logistic regression analysis to evaluate the effects of clinicopathologic factors on rescued HER2 positivity. For independent variables, we used the Wilcoxon-Mann-Whitney test for quantitative variables and the Pearson c2 test or the Fisher exact test for categorical variables. We used the CochranArmitage trend test to assess statistical significance of changes in the relationship between rescued HER2 positivity and HER2 IHC score on the initial biopsy. We assessed objective tumour response according to Response Evaluation Criteria in Solid Tumours v1.1 [9]. We defined progression-free survival as the time between the initiation of treatment and documented disease progression or death from any cause. We used the Kaplan-Meier method and the log-rank test to estimate and compare survival distribution. All tests were twosided, and we considered P 0.05 significant. Table 1 Baseline characteristics of patients at the time of initial HER2 assessment for primary tumour sites. Characteristic Part I Part II Number % Number % Eligible patients 183 100 175 100 Sex, male 121 66.1 96 54.9 Median age, years (range) 55 (24e78) 56 (30e79) Disease status Locally advanced unresectable 13 7.1 1 0.6 Initially metastatic 170 92.9 60 34.3 Recurrent 0 0 114 65.1 Primary tumour location GEJ/cardia/fundus 22 12.0 15 8.6 Body 47 25.7 92 52.6 Antrum 68 37.2 56 32.0 Diffuse stomach 46 25.1 12 6.9 Bormann type 1 3 1.6 3 1.7 2 34 18.6 19 10.9 3 98 53.6 105 60.0 4 46 25.1 25 14.3 EGC 2 1.1 23 13.1 Histology W/D adenocarcinoma 5 2.7 7 4.0 M/D adenocarcinoma 55 30.1 41 23.4 P/D adenocarcinoma 106 57.9 96 54.9 Signet ring cell carcinoma 12 6.6 25 14.3 Mucinous carcinoma 5 2.7 6 3.4 Lauren classification Intestinal 53 29.0 57 32.6 Diffuse 111 60.7 101 57.7 Mixed 19 10.4 17 9.7 HER2 IHC score 0 149 81.4 144 82.3 1 26 14.2 18 10.3 2 8 4.4 13 7.4 Tumour tissues for HER2 assessment in primary tumour Endoscopic biopsy 183 100 52 29.7 Surgical resection 0 0 123 70.3 Tumour tissues for HER2 assessment in metastatic or recurrent tumour Biopsy NA NA 119 68.0 Surgical resection NA NA 55 31.4 or wide excision Malignant ascites NA NA 1 0.6 Metastatic or recurrent sites where HER2 was assessed Peritoneum NA NA 37 21.2 Liver NA NA 29 16.6 Ovary NA NA 29 16.6 Lymph node NA NA 22 12.6 Anastomosis/remnant NA NA 24 13.7 stomach/afferent loop Others NA NA 34 19.4 Time relationship between HER2 assessments in primary and metastatic/recurrent sites Synchronous NA NA 61 34.9 Metachronous (recurrent disease) NA NA 114 65.1 Distant metastatic NA NA 91 52.0 Locoregional NA NA 23 13.1 GEJ, gastroesophageal junction; EGC, early gastric cancer; W/D, well differentiated; M/D, moderately differentiated; P/D, poorly differentiated; HER2, human epidermal growth factor receptor 2; IHC, immunohistochemistry; NA, not applicable. S.R. Park et al. / European Journal of Cancer 53 (2016) 42e50 3. Results 3.1. Study population 3.1.1. Part I From May 2011 to April 2013, 186 patients were enrolled in part I of the study; 183 met eligibility requirements. Table 1 shows their baseline characteristics. The median interval between the initial and repeat biopsies was 13 d (range, 5e147 d). 3.1.2. Part II From May 2011 to February 2014, 179 patients were enrolled in part II of the study; 175 met eligibility requirements (Table 1). The median time interval of HER2 assessment between primary and metastatic/ recurrent lesions was: 0.2 months (range, 0e4.0 months) in initially metastatic/locally advanced disease and 28.8 months (range, 3.7e158.2 months) in recurrent disease. 3.2. Rescued HER2 positivity through repeat biopsy in primary tumours and associated clinical factors in part I Table 2 shows the HER2 status of initial and repeat endoscopic biopsy of primary gastric tumours. The repeat biopsy identified 16 patients with HER2-positive tumour, resulting in a rescued HER2 positivity rate of 8.7% (16 of 183) (95% CI, 4.6e12.8%). Fig. 1A and B show representative photomicrographs of a case with rescued HER2 positivity. In univariate analysis, rescued HER2 positivity was significantly associated with primary tumour location (diffuse stomach versus others Z 0% versus 11.7%, P Z 0.013), Bormann type (IV versus others Z 0% versus 11.7%, P Z 0.013), and the initial biopsy HER2 IHC score (0 versus 1 versus 2 Z 6.7% versus 15.4% versus 25.0%, P Z 0.065 (trend test, P Z 0.028); 0 versus 1/2 Z 6.7% versus 18.2%, P Z 0.045) (Table 3). In logistic regression analysis, patients who had HER2 IHC 1 or 2þ primary tumours on initial biopsy were 3.1 times (95% CI, 1.04e9.28) more likely than those with HER2 IHC 0 to show HER2 positivity on repeat biopsy. 45 The median number of biopsy pieces per patient was five (range, 1e15) on the initial biopsy and 10 (range, 1e15) on the repeat biopsy (P < 0.0001). At least six pieces per patient were obtained in 64 patients (35.0%) on the initial biopsy and 177 patients (96.7%) on the repeat biopsy. There was no difference in the median ratio of number of cancer-containing biopsy pieces/total pieces; 0.86 (range, 0.13e1) on the initial biopsy versus 0.89 (range, 0.10e1) on the repeat biopsy (P Z 0.679). The median number of biopsy pieces per patient was five for the HER2-negative tumours and four for the HER2-positive tumours on the initial biopsy (P Z 0.420), and 10 for both on the repeat biopsy (P Z 0.083). The rescued HER2 positivity rate according to the number of biopsy pieces on the initial biopsy was 9.2% for <six pieces versus 7.8% for six (P Z 0.744) and 11.1% for <three versus 8.5% for three (P Z 0.661). 3.3. Rescued HER2 positivity through reassessment in recurrent or metastatic tumours and associated clinical factors in part II Reassessment of recurrent or metastatic sites identified 10 patients with HER2-positive tumour for a conversion rate of 5.7% (10 of 175) (95% CI, 2.3e9.1%) (Table 4). Fig. 1C and D show representative photomicrographs of such a case. In univariate analysis, this converted HER2 positivity was significantly associated only with metastatic sites where HER2 was reassessed (liver versus others Z 17.2% versus 3.4%, P Z 0.012) (Table 3). In logistic regression analysis, patients who had HER2 reassessment in liver metastasis had a 5.88 (95% CI, 1.58e21.84) times higher probability of HER2 positivity on repeat assessment than those who had HER2 reassessment in other recurrent or metastatic sites. Time relationship between primary and metastatic/recurrent sites (synchronous versus metachronous) was not associated with HER2 positivity at metastatic/recurrent sites in the entire population (3.3% versus 7.0%; P Z 0.497) or the subgroup with liver metastasis (21.1% [4/19] versus 10% [1/10]; P Z 0.633). 3.4. Outcomes of trastuzumab-treated patients with rescued HER2 positivity Table 2 Comparison of HER2 status between initial endoscopic biopsy and repeat biopsy in primary gastric cancer sites. HER2 IHC score in the initial biopsy HER2 IHC score in the repeat biopsy 0 1 2 3 0 1 2 98 8 3 32 12 1 11 (2) 3 (1) 2 8 3 2 HER2, human epidermal growth factor receptor 2; IHC, immunohistochemistry. The number indicates number of patients. Numbers in parenthesis indicate patients who had a HER2 IHC score of 2þ and FISHþ on the repeat biopsy. Twelve of the 16 patients who had rescued HER2 positivity by repeat endoscopic biopsy in primary tumours, and 7 of the 10 who had HER2 positivity at metastatic or recurrent sites, had received trastuzumab-containing first-line chemotherapy. We compared treatment outcomes between these 19 rescued HER2þ patients with 94 patients who had HER2 positivity on the initial assessment for primary tumours in biopsy or surgical specimens and also received trastuzumab-containing first-line chemotherapy as a historical control. In 46 S.R. Park et al. / European Journal of Cancer 53 (2016) 42e50 Fig. 1. Representative photomicrographs of converted human epidermal growth factor receptor 2 (HER2) positivity. Initial endoscopic biopsy in primary tumour shows HER2 negativity in immunohistochemistry (IHC) staining, score 0 (A, original magnification 200), but the tumour cells in repeat biopsy express strong HER2 protein, IHC score 3 (B, original magnification 200). While initial gastrectomy specimen shows HER2 negativity in IHC staining, score 0 (C, original magnification 40), the metastatic tumour cells express strong HER2 protein, IHC score 3 in liver biopsy (D, original magnification 200). patients with measurable lesion(s), the overall tumour response rate was 90.9% (95% CI, 73.9e100%) in patients with rescued HER2 positivity and 76.5% (95% CI, 67.5e85.5%) in patients with HER2 positivity in the initial assessment (P Z 0.446). With a median follow-up duration of 11.9 months (range, 1.5e25.4 months) and 14.2 months (range, 1.4e51.7 months), respectively, the median progression-free survival was 7.6 months (95% CI, 6.5e8.7 months) in patients with rescued HER2 positivity and 8.3 months (95% CI, 6.9e9.7 months) in patients with HER2 positivity in the initial assessment (P Z 0.444) (Fig. 2). 4. Discussion On the basis of therapeutic value of trastuzumab, the National Comprehensive Cancer Network (NCCN) Guidelines recommend that patients with inoperable locally advanced, recurrent, or metastatic adenocarcinoma of the stomach or GEJ be tested for HER2 status [10]. Although intratumoural heterogeneity of HER2 status reportedly occurs in breast cancer [11,12], it seems to occur more commonly in gastric cancer, causing sampling errors in determining patient eligibility for anti-HER2 therapy to be more of an issue [5,7,13e16]. In a study comparing whole tissue sections and corresponding multiple tissue micro-arrays (TMAs) of gastrectomy specimens that served as biopsies procedure, TMAs showed a false-negative HER2 status rate of 24%, providing strong evidence of the risk of biopsy sampling errors [17]. In the present study, we determined the clinical usefulness of the repeat endoscopic biopsy in patients with previous negative HER2 biopsies who would not usually be considered for anti-HER2 therapy for advanced gastric cancer. With repeat biopsy, 8.7% of patients were shown to have HER2-positive gastric cancer. In addition to primary tumour sites, metastatic and/or recurrent sites could be also considered for re-evaluation in cases of initially negative HER2 status. By doing that, 5.7% of patients turned out to have HER2-positive disease. Since the rate of HER2 positivity defined as HER2 IHC 3þ or IHC 2þ/FISHþ in advanced gastric cancer is already low (<20%) [3,18,19], the additional identification in this study of 8.7% (primary sites) and 5.7% (metastatic or recurrent sites) HER2-positive gastric cancers that were considered HER2-negative on initial assessment shows the need for repeat assessment so as not to deny patients the potential benefit of receiving anti-HER2 therapy for HER2-positive gastric cancer. According to our prospective database, HER2 status was evaluated in 875 patients before first-line chemotherapy during the same period as this study, and among them, 110 patients had HER2-positive tumours on initial endoscopic biopsy or gastrectomy assessment, S.R. Park et al. / European Journal of Cancer 53 (2016) 42e50 47 Table 3 Univariate analysis for HER2 positivity in reassessment according to clinicopathologic characteristics. Characteristic HER2 positivity Part I Sex Male Female Age <65 years 65 years Disease status Locally advanced unresectable Metastatic Recurrent Primary tumour location GEJ/cardia/fundus Body Antrum Diffuse stomach Bormann type 1 2 3 4 EGC Histology W/D adenocarcinoma M/D adenocarcinoma P/D adenocarcinoma Signet ring cell carcinoma Mucinous carcinoma Lauren classification Intestinal Diffuse Mixed HER2 IHC score at primary tumour 0 1 2 Tumour tissues in primary gastric HER2 assessment Endoscopic biopsy Surgical resection No. of biopsy pieces obtained at primary tumourd <6 6 Time relationship between HER2 assessment at primary and metastatic/recurrent sites Synchronous metastatic site Recurrent metastatic site Recurrent locoregional site Sites of metastasis/recurrence where HER2 was assessed Peritoneum Liver Ovary Lymph node Anastomosis/remnant stomach Others Chemotherapy between HER2 assessment at primary and metastatic/recurrent sites No chemotherapy Adjuvant chemotherapy Part II Number % 10/121 6/62 8.3 9.7 9/134 7/49 6.7 14.3 3/13 13/170 NA 23.1 7.6 NA 2/22 6/47 8/68 0/46 9.1 12.8 11.8 0 P Number % 0.749 0.756 5/96 5/79 5.2 6.3 6/132 4/43 4.5 9.3 0/1 2/60 8/114 0 3.3 7.0 2/15 7/92 1/56 0/12 13.3 7.6 1.8 0 0/3 1/19 5/105 1/25 3/23 0 5.3 4.8 4.0 13.0 0/7 5/41 5/96 0/25 0/6 0 12.2 5.2 0 0 4/57 3/101 3/17 7.0 3.0 17.6 0.138 0.263 0.091 0.526 0.013a 0.194 0.013b 0/3 3/34 13/98 0/46 0/2 0 8.8 13.3 0 0 0/5 4/55 10/106 2/12 0/5 0 7.3 9.4 16.7 0 4/53 9/111 3/19 7.5 8.1 15.8 P 0.513 0.799 0.352 0.530 0.055 0.045c 0.821 10/149 4/26 2/8 6.7 15.4 25.0 8/144 1/18 1/13 5.6 5.6 7.7 NA NA NA NA 2/52 8/123 3.8 6.5 11/119 5/64 9.2 7.8 1/28 1/24 3.6 4.2 0.725 0.744 1.000 0.485 NA NA NA NA NA NA 2/61 6/91 2/23 3.3 6.6 8.7 NA NA NA NA NA NA NA NA NA NA NA NA 0/37 5/29 2/29 1/22 2/24 0/34 0 17.2 6.9 4.5 8.3 0 0.019 0.749 NA NA NA NA 4/84 6/91 4.8 6.6 HER2, human epidermal growth factor receptor 2; GEJ, gastroesophageal junction; EGC, early gastric cancer; W/D, well differentiated; M/D, moderately differentiated; P/D, poorly differentiated; IHC, immunohistochemistry; NA, not applicable. a P value for diffuse stomach versus others. b P value for type 4 versus others. c P value for IHC 0 versus 1/2. d Patients only who had HER2 assessment for primary tumour in endoscopic biopsy. 48 S.R. Park et al. / European Journal of Cancer 53 (2016) 42e50 Table 4 HER2 status of primary tumour sites versus metastatic or recurrent tumour sites. HER2 IHC score at primary tumour sites HER2 IHC score at the metastatic or recurrent sites 0 1 2 3 0 1 2 107 9 8 26 7 1 5 (2) 2 (1) 4 (1) 6 0 0 The number indicates number of patients. Numbers in parenthesis indicate patients who had a HER2 IHC score of 2þ and FISHþ at metastatic or recurrent sites. HER2, human epidermal growth factor receptor 2; IHC, immunohistochemistry. resulting in a 12.6% HER2-positive rate. If the remaining 765 patients with HER2-negative primary tumour had been reassessed both at primary and metastatic tumour sites in cases with initially metastatic/locally advanced disease (N Z 413) and at recurrent tumour sites in cases with recurrent disease (N Z 352), 60 cases (413 8.7% þ 413 5.7%) and 20 cases (352 5.7%), respectively, would have had rescued HER2 positivity, and consequently, the rate of HER2 positivity would have been 21.7% (190/875). This corresponds to a 72.2% relative increase of HER2 positivity through repeat assessment. Although the cost and potential harm associated with repeat assessment requires further evaluation, our study demonstrates that this approach is Fig. 2. Kaplan-Meier curves of progression-free survival in patients with human epidermal growth factor receptor 2 (HER2) positivity in the initial assessment in endoscopic biopsy or gastrectomy specimen for primary tumour (initial HER2 positivity) and in patients with HER2 positivity in the repeat endoscopic biopsy after initial HER2-negative endoscopic biopsy or in the reassessment of metastatic or recurrent sites after initial HER2negative biopsy or gastrectomy specimen for primary tumour (converted HER2 positivity). All patients were treated with trastuzumab-containing first-line chemotherapy. feasible and not associated with significant complications while it has major therapeutic implications in that patients with rescued HER2 positivity can be offered the same benefits of trastuzumab-containing chemotherapy as those showing HER2 positivity on initial assessment (Fig. 2). Previous studies show that HER2 positivity in advanced gastric cancer is associated with clinicopathologic characteristics of Lauren’s intestinal type, the proximal primary tumour region encompassing gastric cardia/GEJ, and liver metastasis [19,20]. In the present study, the factors significantly associated with rescued HER2 positivity on the repeat endoscopic biopsy included the primary tumour location other than diffuse stomach, Bormann type other than type four, and the higher HER2 IHC score on the initial endoscopic biopsy. Of note, since patients who had HER2 IHC 1 or 2þ primary tumours on the initial biopsy were 3.1 times as likely as those who had HER2 IHC 0 tumours to show HER2 positivity on repeat biopsy, they, in particular, should be considered for re-biopsy. Regarding the number of endoscopic biopsy fragments, although there were significant differences between the initial and repeat biopsy (median, 5 versus 10, respectively; P < 0.0001), the number of biopsy fragments on the initial biopsy was not associated with rescued HER2 positivity on the repeat biopsy (9.2% for <six fragments versus 7.8% for six; P Z 0.744; 11.1% for <3 versus 8.5% for three, P Z 0.661). While the NCCN Guidelines recommend multiple (6e8) endoscopic biopsies to provide an adequate-sized material for histologic interpretation [10], there are controversial data about optimum numbers of viable biopsies of the cancer to account for intratumoural heterogeneity of HER2 expression. Tominaga and colleagues suggested at least five biopsy fragments based on 100% concordance between the biopsy and resection specimens in HER2 status in 24 HER2-positive gastric cancers (defined as HER2 IHC 2þ or 3þ) [21], whereas Warneke and colleagues showed the 24% false-negative HER2 status rate with five TMAs in each tumour when compared to corresponding whole tissue sections [17]. Huang and colleagues showed that there was no association between the number of biopsy fragments and HER2 discordance between biopsy and surgical specimens [22]. Further studies are needed to clarify the optimal number of endoscopic biopsy for accurate HER2 evaluation in gastric cancer. Previous studies reported variable concordance rates of HER2 status between primary gastric cancer sites and paired metastatic sites [15,23e25]. Bozzetti and colleagues [23] reported 94.9% concordance between HER2 status by IHC on 39 primary and corresponding metastatic sites, but they allowed equivocal (IHC 2þ) cases in one site to have either negative (IHC 0/1þ) or equivocal (IHC 2þ) status in the other site. Kim and colleagues [15] reported significant discordance (17.2%) between S.R. Park et al. / European Journal of Cancer 53 (2016) 42e50 250 paired primary and metastatic lesions by IHC. Kochi and colleagues [25] showed 90.2% concordance between HER2 status by IHC and FISH on 102 paired primary gastric cancer and related regional metastatic lymph nodes. Our study focused only on patients who had HER2-negative primary tumours on initial assessment and prospectively evaluated how many patients had positive conversion in recurrent or metastatic sites. Although the overall positive conversion rate of 5.7% was not so high, an interesting finding was that the positive conversion of HER2 status at metastatic sites occurred about six times more frequently in liver metastases than at other sites (17.2% versus 3.4%; P Z 0.012). This could be partially explained by preferential liver metastasis of HER2-positive cancer cells [19,26]. This relatively high conversion rate indicates that reassessment of HER2 status is particularly warranted in liver metastases even when the primary tumours were HER2-negative. In conclusion, on the basis of the recognition of intraand inter-tumoural heterogeneity, the rescued HER2 positivity found in our study, and its important therapeutic implications, we recommend that repeat assessment of HER2 status in primary and metastatic and/or recurrent tumours in patient with unresectable, metastatic, or recurrent gastric cancer even when initial assessment showed the primary tumour to be HER2negative. Conflict of interest statement Yoon-Koo Kang declares receiving research grants from Sanofi, Bayer, Roche, and Novartis and discloses a consultant role for Roche, Bayer, Sanofi, Taiho, Ono, Pfizer, and Novartis. Min-Hee Ryu declares receiving honoraria from Novartis, Roche, Sanofi, and Taiho, research grants from Novartis and having a consultant role with Novartis and Roche. All other authors declare no conflicts of interest. Acknowledgements This work was supported in part by Roche Korea. Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.ejca.2015.09.018 References [1] Slamon DJ, Godolphin W, Jones LA, Holt JA, Wong SG, Keith DE, et al. Studies of the HER-2/neu proto-oncogene in human breast and ovarian cancer. Science 1989;244:707e12. [2] Hudis CA. Trastuzumabemechanism of action and use in clinical practice. N Engl J Med 2007;357:39e51. 49 [3] Bang YJ, Van Cutsem E, Feyereislova A, Chung HC, Shen L, Sawaki A, et al. Trastuzumab in combination with chemotherapy versus chemotherapy alone for treatment of HER2-positive advanced gastric or gastro-oesophageal junction cancer (ToGA): a phase 3, open-label, randomised controlled trial. Lancet 2010; 376:687e97. [4] Yoon DH, Kang Y-K. Results and implications of the Trastuzumab for Gastric Cancer (ToGA) trial. Clin Invest 2011;1: 87e95. [5] Hofmann M, Stoss O, Shi D, Büttner R, van de Vijver M, Kim W, et al. Assessment of a HER2 scoring system for gastric cancer: results from a validation study. Histopathology 2008;52:797e805. [6] Wolff AC, Hammond ME, Hicks DG, Dowsett M, McShane LM, Allison KH, et al. Recommendations for human epidermal growth factor receptor 2 testing in breast cancer: American Society of Clinical Oncology/College of American Pathologists clinical practice guideline update. J Clin Oncol 2013;31: 3997e4013. [7] Ruschoff J, Dietel M, Baretton G, Arbogast S, Walch A, Monges G, et al. HER2 diagnostics in gastric cancer-guideline validation and development of standardized immunohistochemical testing. Virchows Arch 2010;457:299e307. [8] Wolff AC, Hammond ME, Schwartz JN, Hagerty KL, Allred DC, Cote RJ, et al. American Society of Clinical Oncology/College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer. J Clin Oncol 2007;25:118e45. [9] Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent D, Ford R, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009;45: 228e47. [10] National Comprehensive Cancer Network clinical practice guidelines in oncology (version I.2015): Gastric cancer. http:// www.nccn.org/professionals/physician_gls/pdf/gastric.pdf. [11] Seol H, Lee HJ, Choi Y, Lee HE, Kim YJ, Kim JH, et al. Intratumoral heterogeneity of HER2 gene amplification in breast cancer: its clinicopathological significance. Mod Pathol 2012;25: 938e48. [12] Bartlett AI, Starcyznski J, Robson T, Maclellan A, Campbell FM, van de Velde CJ, et al. Heterogeneous HER2 gene amplification: impact on patient outcome and a clinically relevant definition. Am J Clin Pathol 2011;136:266e74. [13] Lee HE, Park KU, Yoo SB, Nam SK, Park do J, Kim HH, et al. Clinical significance of intratumoral HER2 heterogeneity in gastric cancer. Eur J Cancer 2013;49:1448e57. [14] Ruschoff J, Nagelmeier I, Baretton G, Dietel M, Höfler H, Schildhaus HU, et al. Her2 testing in gastric cancer. What is different in comparison to breast cancer? Pathologe 2010;31: 208e17. [15] Kim MA, Lee HJ, Yang HK, Bang YJ, Kim WH. Heterogeneous amplification of ERBB2 in primary lesions is responsible for the discordant ERBB2 status of primary and metastatic lesions in gastric carcinoma. Histopathology 2011;59:822e31. [16] Lee S, de Boer WB, Fermoyle S, Platten M, Kumarasinghe MP. Human epidermal growth factor receptor 2 testing in gastric carcinoma: issues related to heterogeneity in biopsies and resections. Histopathology 2011;59:832e40. [17] Warneke VS, Behrens HM, Boger C, Becker T, Lordick F, Ebert MP, et al. Her2/neu testing in gastric cancer: evaluating the risk of sampling errors. Ann Oncol 2013;24:725e33. [18] Gomez-Martin C, Garralda E, Echarri MJ, Ballesteros A, Arcediano A, Rodrı́guez-Peralto JL, et al. HER2/neu testing for anti-HER2-based therapies in patients with unresectable and/or metastatic gastric cancer. J Clin Pathol 2012;65:751e7. [19] Janjigian YY, Werner D, Pauligk C, Steinmetz K, Kelsen DP, Jäger E, et al. Prognosis of metastatic gastric and gastroesophageal junction cancer by HER2 status: a European and USA international collaborative analysis. Ann Oncol 2012;23:2656e62. 50 S.R. Park et al. / European Journal of Cancer 53 (2016) 42e50 [20] Van Cutsem E, Bang YJ, Feng-Yi F, Xu JM, Lee KW, Jiao SC, et al. HER2 screening data from ToGA: targeting HER2 in gastric and gastroesophageal junction cancer. Gastric Cancer 2015;18:476e84. [21] Tominaga N, Gotoda T, Hara M, Hale MD, Tsuchiya T, Matsubayashi J, et al. Five biopsy specimens from the proximal part of the tumor reliably determine HER2 protein expression status in gastric cancer. Gastric Cancer 2015 [Epub ahead of print]. [22] Huang SC, Ng KF, Lee SE, Chen KH, Yeh TS, Chen TC, et al. HER2 testing in paired biopsy and excision specimens of gastric cancer: the reliability of the scoring system and the clinicopathological factors relevant to discordance. Gastric Cancer 2014 [Epub ahead of print]. [23] Bozzetti C, Negri FV, Lagrasta CA, Crafa P, Bassano C, Tamagnini I, et al. Comparison of HER2 status in primary and paired metastatic sites of gastric carcinoma. Br J Cancer 2011;104: 1372e6. [24] Shibata R, Nimura S, Hashimoto T, Miyake T, Takeno S, Hoshino S, et al. Expression of human epidermal growth factor receptor 2 in primary and paired parenchymal recurrent and/or metastatic sites of gastric cancer. Mol Clin Oncol 2014;2: 751e5. [25] Kochi M, Fujii M, Masuda S, Kanamori N, Mihara Y, Funada T, et al. Differing deregulation of HER2 in primary gastric cancer and synchronous related metastatic lymph nodes. Diagn Pathol 2013;8:191. [26] Shitara K, Yatabe Y, Matsuo K, Sugano M, Kondo C, Takahari D, et al. Prognosis of patients with advanced gastric cancer by HER2 status and trastuzumab treatment. Gastric Cancer 2013;16:261e7.