bevacizumab in combination with chemotherapy as first-line therapy in advanced gastric cancer: a biomarker evaluation from the avagast randomized phase iii trial Bevacizumab in Combination With Chemotherapy As First-Line Therapy in Advanced Gastric Cancer: A Biomarker Evaluation From the AVAGAST Randomized Phase III Trial

Bevacizumab in Combination With Chemotherapy As First-Line Therapy in Advanced Gastric Cancer: A Biomarker Evaluation From the AVAGAST Randomized Phase III Trial

  1. Manish A. Shah⇓
  1. Eric Van Cutsem, University Hospital Gasthuisberg, Leuven, Belgium; Sanne de Haas and Paul Delmar, F. Hoffmann-La Roche, Basel, Switzerland; Yoon-Koo Kang, Asan Medical Center, University of Ulsan College of Medicine, Seoul, South Korea; Atsushi Ohtsu, National Cancer Center Hospital East, Kashiwa, Chiba, Japan; Niall C. Tebbutt, Austin Health, Heidelberg, Victoria, Australia; Jian Ming Xu, PLA 307 Hospital, Fengtai District, Beijing, China; Wei Peng Yong, National University Hospital, Singapore; Bernd Langer and Stefan J. Scherer, Genentech, South San Francisco, CA; and Manish A. Shah, Memorial Sloan-Kettering Cancer Center, Weill Cornell Medical College, New York, NY.
  1. Corresponding author: Manish A. Shah, MD, Weill Cornell Medical College, New York-Presbyterian Hospital, 1305 York Ave, 12th Floor, Box 189, New York, NY 10021; e-mail: mas9313{at}med.cornell.edu.

  1. Presented in part at the 35th European Society for Medical Oncology Congress, Milan, Italy, October 8-12, 2010, and the European Multidisciplinary Cancer Congress, Stockholm, Sweden, September 23-27, 2011.

  1. E.V.C., S.D.H., and Y.-K.K. contributed equally to this work.

 
Next Section

Abstract

Purpose The AVAGAST study showed that adding bevacizumab to chemotherapy in patients with advanced gastric cancer improves progression-free survival and tumor response rate but not overall survival. To examine the hypothesis that angiogenic markers may have predictive value for bevacizumab efficacy in gastric cancer, AVAGAST included a prospective, mandatory biomarker program.

Patients and Methods Patients with previously untreated, locally advanced or metastatic gastric cancer were randomly assigned to bevacizumab (n = 387) or placebo (n = 387) in combination with chemotherapy. Blood and tumor tissue samples were collected at baseline. Prespecified biomarkers included plasma vascular endothelial growth factor-A (VEGF-A), protein expression of neuropilin-1, and VEGF receptors-1 and -2 (VEGFR-1 and VEGFR-2). Correlations between biomarkers and clinical outcomes were assessed by using a Cox proportional hazards model.

Results Plasma was available from 712 patients (92%), and tumor samples were available from 727 patients (94%). Baseline plasma VEGF-A levels and tumor neuropilin-1 expression were identified as potential predictors of bevacizumab efficacy. Patients with high baseline plasma VEGF-A levels showed a trend toward improved overall survival (hazard ratio [HR], 0.72; 95% CI, 0.57 to 0.93) versus patients with low VEGF-A levels (HR, 1.01; 95% CI, 0.77 to 1.31; interaction P = .07). Patients with low baseline expression of neuropilin-1 also showed a trend toward improved overall survival (HR, 0.75; 95% CI, 0.59 to 0.97) versus patients with high neuropilin-1 expression (HR, 1.07; 95% CI, 0.81 to 1.40; interaction P = .06). For both biomarkers, subgroup analyses demonstrated significance only in patients from non-Asian regions.

Conclusion Plasma VEGF-A and tumor neuropilin-1 are strong biomarker candidates for predicting clinical outcome in patients with advanced gastric cancer treated with bevacizumab.

Previous SectionNext Section

INTRODUCTION

Antiangiogenic therapy through inhibition of vascular endothelial growth factor-A (VEGF-A) with bevacizumab shows broad clinical activity.1–5 However, the magnitude of benefit from bevacizumab varies among patients. Identifying tumors most sensitive to anti-VEGF therapy would improve our therapeutic approach and provide important insights into the biologic mechanisms of resistance, thereby providing opportunities to further refine antiangiogenic therapy. This strategy has proven successful when targeting human epithelial growth factor receptor-2 (HER2) in breast and gastric cancers.6,7 Herein, we report on the evaluation of putative antiangiogenic biomarkers in metastatic gastric cancer, the largest prospective biomarker evaluation for this class of cancer therapeutics to date.

AVAGAST (BO20904/AVF4200g) was a global, randomized, double-blind, phase III study designed to compare the efficacy of bevacizumab plus chemotherapy versus placebo plus chemotherapy as first-line treatment for patients with advanced gastric cancer.8 Although bevacizumab was associated with a significantly longer progression-free survival versus placebo (median, 6.7 v 5.3 months; hazard ratio [HR], 0.80; P = .0037) and higher overall response rate (46.0% v 37.4%; P = .0315), the difference in overall survival, the primary study end point, did not reach statistical significance (12.1 v 10.1 months; HR, 0.87; P = .1002).8 A differential effect was also noted in Asian compared with non-Asian patients.8

Our a priori hypothesis was that a comprehensive, prospective biomarker analysis program would improve our understanding of the complexity of tumor angiogenesis. AVAGAST contained a mandatory biomarker program in which blood samples and tissue were gathered to examine the hypotheses that plasma and tumor tissue markers involved in the VEGF pathway may have predictive value for the efficacy of bevacizumab in gastric cancer. The markers evaluated were circulating VEGF-A and tumor expression of VEGF-A, VEGF receptors-1 and -2 (VEGFR-1 and VEGFR-2), and neuropilin-1. In addition, given the positive results of the ToGA study (A Study of Herceptin [Trastuzumab] in Combination With Chemotherapy Compared With Chemotherapy Alone in Patients With HER2-Positive Advanced Gastric Cancer),6 we examined the prognostic value of epidermal growth factor receptor-1 (EGFR-1) and EGFR-2 (or HER2) in this study in which patients did not receive trastuzumab therapy.

Previous SectionNext Section

PATIENTS AND METHODS

Study Design

Design details of the AVAGAST study have been described previously.8 In brief, patients with previously untreated, histologically confirmed, unresectable, locally advanced or metastatic adenocarcinoma of the stomach or gastroesophageal junction were assigned to bevacizumab (n = 387) or placebo (n = 387) in combination with cisplatin for six cycles plus capecitabine until disease progression or intolerable adverse effects. Fluorouracil was substituted for capecitabine in patients unable to take oral medications. The protocol was approved at each participating site by an independent ethics committee or institutional review board. The trial was carried out in accordance with the Declaration of Helsinki. All patients provided written informed consent before study entry.

Sample Collection

The biomarker analysis was predefined in the study protocol. Baseline blood and tumor tissue (primary or recurrent) samples were collected from all randomly assigned patients. Blood samples (5 mL collected in EDTA) for plasma isolation were taken between random assignment and start of treatment. Tumor samples were collected at baseline as either paraffin-embedded tissue blocks or four to 20 slides of unstained, freshly cut tissue from the formalin-fixed, paraffin-embedded tumor.

Biomarker Assessments and Methodology

Plasma samples were analyzed centrally at Roche Diagnostics (Penzberg, Germany) by using IMPACT (immunologic multiparametric chip technique) technology,9 a Roche proprietary technology that is not available commercially, and a multiplex enzyme-linked immunosorbent assay platform for several angiogenic factors, including VEGF-A. This technology uses a three-step sandwich immunoassay principle, with immobilized target-specific capture antibodies, a digoxigenin-labeled detection antibody, and antidigoxin antibody conjugated with fluorescent latex. In the VEGF-A assay, all VEGF-A isoforms are detected, and the assay shows a preference for short (VEGF110, VEGF121) over longer (VEGF165, VEGF189) isoforms.

Tissue biomarkers were analyzed centrally at Targos Molecular Pathology in Kassel, Germany. Immunohistochemistry (IHC) was performed on 5-μm sections of paraffin-embedded tissue by using the techniques described in Appendix Table A1 (online only). After deparaffinization and rehydration, antigen retrieval was performed in 1 mmol/L EDTA pH 8.0 at 95°C to 99°C for 20 minutes or as described in the kit instructions (HER2 and EGFR). Sections were stained by using an Autostainer or Autostainer Plus (Dako A/S, Glostrup, Denmark). Primary antibodies were incubated for 30 minutes (HER2 and EGFR), 120 minutes (VEGF-A, VEGFR-2, and neuropilin-1), or overnight (VEGFR-1). All sections were counterstained with Mayer's hematoxylin.

To assess tissue biomarker expression, an H-score was calculated for each sample. H-score was defined as the percentage of cells with weak stain intensity plus two times the percentage of cells with moderate stain intensity plus three times the percentage of cells with strong stain intensity.10 HER2 status was determined by using a previously reported scoring system for gastric cancer,11 which involved both IHC (HercepTest; Dako) and fluorescent in situ hybridization (pharmDx, Dako). HER2-positive tumor samples were IHC 3+ or were fluorescent in situ hybridization positive (ie, HER2:CEP ratio > 2.0). Scoring was performed blind to treatment and outcome.

Statistical Analysis

The cutoff date for the primary efficacy analysis was November 30, 2009, when the preplanned number of 509 deaths had occurred. There were no preplanned sample size or power calculations for the biomarker analyses.

Correlations between biomarkers and the following outcome measures were assessed: overall survival, defined as time between random assignment and death irrespective of cause; progression-free survival, defined as time between random assignment and first documented disease progression or death as evaluated by the investigator; and overall response rate, which was evaluated by using Response Evaluation Criteria in Solid Tumors (RECIST), version 1.0.12 Time-to-event end points were analyzed by using Kaplan-Meier methods.

The prognostic and predictive value of each biomarker were assessed by using a Cox proportional hazards model for time-to-event end points and logistic regression for tumor response rates. Cox proportional hazards models were used to determine HRs and 95% CIs with terms of treatment group, biomarker, and interaction between treatment and biomarker. Additional models that included stratification factors (ie, geographic region, fluoropyrimidine, disease status) as covariates were also run. The interaction between treatment and each biomarker (potential predictive strength) was assessed by using the Wald test.

For the purposes of analysis, a logarithmic transformation of plasma VEGF-A levels was used. Biomarker H-scores were dichotomized according to the sample median (ie, greater than the median denoted high expression levels, and below or equal to the median denoted low expression levels). For HER2, patients were categorized as HER2-positive or HER2-negative. To identify trends that may not have been apparent when using median cutoffs, biomarker H-scores were further categorized and analyzed by quartile.

Previous SectionNext Section

RESULTS

Patients

From September 2007 to December 2008, 774 patients were enrolled and randomly assigned (CONSORT diagram; Fig 1). Baseline plasma samples were available from 712 patients (92%) and tumor samples from 727 (94%). All tumor biomarkers, except for VEGFR-1 (81%) and neuropilin-1 (88%), were analyzed in at least 90% of the overall study population. Patient characteristics of the biomarker populations were similar to those of the overall population (data not shown).

Fig 1.
View larger version:
Fig 1.

CONSORT diagram. ECOG, Eastern Cooperative Oncology Group; GI, gastrointestinal; PS, performance status.

Most tissue samples (approximately 60%) were provided as slides, although there were differences in how samples were supplied between regions. In Asian centers, 7% of samples were provided as tumor blocks and 93% as slides, whereas in non-Asian centers, 83% of samples were supplied as blocks and 17% as slides.

Biomarker Levels at Baseline

Biomarker levels at baseline are shown in Appendix Table A2 (online only). There were no noteworthy differences between treatment groups in the baseline levels of each of the biomarkers (data not shown).

Median plasma levels of VEGF-A in the biomarker population were 111 ng/L (range, 20 to 1,868 ng/L). All six tissue biomarkers were expressed at baseline. Median H-scores for VEGF-A, VEGFR-1, VEGFR-2, and neuropilin-1 ranged from 90 to 130. However, there was variability in the expression of some biomarkers between tumor blocks and slides. Neuropilin-1 median H-score was significantly lower in samples provided as slides (n = 365; median H-score, 80; range, 0 to 250) versus block samples (n = 314; median H-score, 100; range, 0 to 215; P < .001). Differences were also noted in the tumor expression of VEGF-A (n= 320 in blocks; median H-score, 103; n = 381 in slides; median H-score, 70) and VEGFR-1 (n = 304 in blocks; median H-score, 110; and n = 325 in slides; median H-score, 140), although these differences may have been because of regional differences in sample collection. Tumor expression of both EGFR and HER2 was low. The median H-score for EGFR was 10, and 35% of patients (253 of 731) had a value of zero; 14% of patients (98 of 712) had HER2-positive tumors.

Biomarker Association With Baseline Characteristics

Demographic and baseline characteristics were generally balanced for most biomarkers (Appendix Table A3, online only). Baseline plasma VEGF-A levels were higher in patients enrolled from non-Asian regions than in patients from the Asia-Pacific region (median, 147 v 94 ng/L; P < .001; Mann-Whitney U test) and in patients with a poorer Eastern Cooperative Oncology Group (ECOG) performance status (P < .001; Mann-Whitney U test).

Neuropilin-1 expression levels were higher in patients in the Asia-Pacific region versus other regions (P < .001; Fisher's exact test). Because only 52 samples from metastatic tumors were collected, the regional differences could not be explained by differences among samples from primary versus metastatic tumors. Neuropilin-1 expression also correlated with gastric tumor histologic subtype (P = .0012; Fisher's exact test).

Biomarker Association With Outcomes

Forest plots of the HRs for overall survival and progression-free survival for each biomarker are presented in Figure 2. Most biomarkers did not show either prognostic or predictive effects, except for plasma VEGF-A and tissue neuropilin-1.

Fig 2.
View larger version:
Fig 2.

Forest plots of hazard ratios (bevacizumab plus chemotherapy v placebo plus chemotherapy) for (A) overall survival and (B) progression-free survival by biomarker (dichotomized by median value or human epithelial growth factor receptor 2 [HER2] status). HER2-positive samples had an immunohistochemistry score of 3+ or were fluorescent in situ hybridization positive. BL, baseline; EGFR, epidermal growth factor receptor; NRP1, neuropilin-1; VEGFA, vascular endothelial growth factor-A; VEGFR1, VEGF receptor-1; VEGFR2, VEGF receptor-2.

Prognostic Biomarkers

Both plasma VEGF-A and tumor neuropilin-1 showed potential prognostic effects. For plasma VEGF-A, patients in the placebo group with high baseline plasma VEGF-A levels had a shorter median overall survival (8.3 months) than patients with low levels (12.9 months; Figure 3A). For tumor neuropilin-1, patients in the placebo group with low expression levels had a shorter overall survival (9.8 months) than those with high expression levels (11.1 months; Figure 4A). The observed prognostic effect was still present in models adjusting for stratification variables for both biomarkers.

Fig 3.
View larger version:
Fig 3.

Kaplan-Meier estimates of (A) overall survival and (B) progression-free survival by baseline plasma vascular endothelial growth factor-A (pVEGF-A) levels (dichotomized by median value) and by treatment group (biomarker population). Chemo, chemotherapy.

Fig 4.
View larger version:
Fig 4.

Kaplan-Meier estimates of (A) overall survival and (B) progression- free survival by baseline tissue neuropilin-1 expression levels (dichotomized by median value) and by treatment group (biomarker population). Chemo, chemotherapy.

HER2 status was notably not prognostic. In the placebo group, patients with HER2-positive tumors had a median overall survival (9.8 months) similar to that of patients with HER2-negative tumors (10.5 months).

Predictive Biomarkers

Patients with high baseline plasma levels of VEGF-A were more sensitive to bevacizumab treatment as measured by overall survival (HR, 0.72; 95% CI, 0.57 to 0.93) than those with low VEGF-A levels (HR, 1.01; 95% CI, 0.77 to 1.31; interaction P = .07; Table 1). Progression-free survival was also increased in patients treated with bevacizumab who had high baseline plasma VEGF-A levels compared with patients in the placebo group (HR, 0.66 [95% CI, 0.52 to 0.85] v HR, 0.86 [95% CI, 0.67 to 1.10] with low VEGF-A levels; interaction P = .11; Table 1). These results were similar when models that adjusted for stratification variables were applied. No correlation between plasma VEGF-A levels and tumor VEGF-A expression was noted. In addition, only baseline VEGF-A levels were analyzed; changes in VEGF-A during treatment are unreliable because of interference by bevacizumab. When analyzed by quartile, the patient populations with plasma VEGF-A levels in the highest two quartiles (> 50%) experienced most benefit from bevacizumab in terms of both overall survival and progression-free survival (Fig 5).

View this table:
Table 1.

Correlation Between Efficacy Outcomes and Baseline Plasma VEGF-A and Tissue Neuropilin-1 Expression Levels (biomarker populations)

Fig 5.
View larger version:
Fig 5.

Forest plots of hazard ratios (bevacizumab plus chemotherapy v placebo plus chemotherapy) (A) overall survival and (B) progression-free survival by biomarker (by quartile; biomarker population). BL, baseline; EGFR, epidermal growth factor receptor; NRP1, neuropilin-1; VEGF, vascular endothelial growth factor; VEGFA, vascular endothelial growth factor-A; VEGFR1, VEGF receptor-1; VEGFR2, VEGF receptor-2.

The trends observed for overall survival and progression-free survival were mainly observed in patients enrolled in non-Asian regions (ie, pan-America and Europe; Table 2). In non-Asian patients, those with high plasma VEGF-A levels appeared to benefit from bevacizumab therapy as measured by both overall survival (HR, 0.59 [95% CI, 0.43 to 0.82] v HR, 1.01 [95% CI, 0.68 to 1.51] with low VEGF-A levels; interaction P = .04) and progression-free survival (HR, 0.54 [95% CI, 0.39 to 0.76] v HR, 0.85 [95% CI, 0.57 to 1.26] with low VEGF-A levels; interaction P = .06). This effect was not evident in patients enrolled in the Asia-Pacific region (Table 2).

View this table:
Table 2.

Correlation Between Efficacy Outcomes and Baseline Plasma VEGF-A and Tissue Neuropilin-1 Expression Levels (dichotomized by median value) by Geographic Region

Patients with low neuropilin-1 expression at baseline appeared to experience a larger bevacizumab treatment effect in terms of overall survival (HR, 0.75; 95% CI, 0.59 to 0.97) than patients with high expression levels (HR, 1.07; 95% CI, 0.81 to 1.40; interaction P = .06; Tables 1 and 2). There was also a numerical difference in the HRs for progression-free survival (HR, 0.68 [95% CI, 0.53 to 0.87] for low neuropilin-1 expression levels v HR, 0.80 [95% CI, 0.62 to 1.05] for high neuropilin-1 expression levels; interaction P = .37) and odds ratios (ORs) for overall response rate (OR, 1.97 [95% CI, 1.19 to 3.28] for low neuropilin-1 expression levels v OR, 1.08 [95% CI, 0.67 to 1.76] for high neuropilin-1 expression levels; interaction P = .09). These observations were supported in models adjusted for stratification variables. When analyzed by quartile, there was a clear inverse relationship between bevacizumab treatment effect on overall survival and neuropilin-1 expression (Fig 5), with increased benefit from bevacizumab treatment observed in patients with lower levels of neuropilin-1 expression.

Previous SectionNext Section

DISCUSSION

To the best of our knowledge, this is the first and largest randomized phase III study evaluating the efficacy of bevacizumab with a comprehensive prospective biomarker analysis. We demonstrate that baseline plasma levels of VEGF-A and baseline tumor expression of neuropilin-1 are candidate biomarkers of bevacizumab efficacy in patients with advanced gastric cancer.

Pretreatment levels of circulating VEGF-A have been previously studied as a biomarker for VEGF-targeted therapies,13 but have been shown to be primarily prognostic14–17 rather than predictive. In AVAGAST, we demonstrate that high plasma VEGF-A levels are associated with worse prognosis in metastatic gastric cancer. In addition, we demonstrate that high plasma levels of VEGF-A at baseline are associated with a trend toward a larger bevacizumab effect in terms of overall survival, progression-free survival, and overall response rate. These benefits were not evident in patients with low baseline VEGF-A levels. The potential predictive role of plasma VEGF-A identified in AVAGAST is supported by two other recent independent biomarker analyses performed in patients with metastatic breast cancer (AVADO study)18 and pancreatic cancer (AViTA study).19 In these studies, the HRs for progression-free survival in patients with high plasma VEGF-A levels ranged from 0.49 to 0.52 (0.66 in AVAGAST), and in patients with low plasma VEGF-A levels, they ranged from 0.77 to 0.96 (0.86 in AVAGAST).18,19 All three analyses used the same novel enzyme-linked immunosorbent-based assay for the detection of VEGF-A in plasma. Different findings were reported in other studies, but they may have been confounded by sample conditions.20 A potential predictive role for plasma VEGF-A in these indications cannot be ruled out at this stage.

We noted that plasma VEGF-A levels appeared to be predictive of efficacy in non-Asian patients. Patients enrolled from the Asia-Pacific region tended to have lower baseline VEGF-A levels, and those with high VEGF-A levels did not show a significant bevacizumab treatment effect. Similarly, bevacizumab appeared to demonstrate improved efficacy in patients in non-Asian countries compared with patients in the Asia-Pacific region.8 Together, these data strengthen the postulate of plasma VEGF-A (high) as a candidate biomarker to predict bevacizumab efficacy in advanced gastric cancer.

Baseline tumor neuropilin-1 expression was also identified as a potential predictive candidate for bevacizumab efficacy in patients with gastric cancer. In AVAGAST, low neuropilin-1 expression was associated with worse overall survival, suggesting neuropilin-1 is a prognostic variable in advanced gastric cancer. In addition, patients with low levels of neuropilin-1 expression appeared to derive more benefit from bevacizumab therapy than patients with higher levels of expression in terms of overall survival, progression-free survival, and overall response rate. These observations are consistent with two recent retrospective, exploratory biomarker analyses in colorectal cancer (NO16966 study)21 and breast cancer (AVF2119g study),22 although different antibodies and methodologies were used in all studies.

Neuropilins are transmembrane glycoproteins, but the molecular mechanism responsible for neuropilin signaling remains unknown. They are, however, coreceptors for the VEGF family and are known to play a major role in vasculogenesis, angiogenesis, and tumor growth.23 A possible explanation for its predictive potential in this study is that low neuropilin-1 expression may reflect greater dependence on VEGF-A ligand binding without providing an alternative pathway by semaphorin binding for VEGFR activation, making the tumor more susceptible to bevacizumab therapy. We did observe differences in neuropilin-1 expression according to geographic region that may have been confounded by the tissue acquisition from each region. Specifically, the Asia-Pacific region tended to supply precut slides (versus tumor blocks), and slides demonstrated a significantly lower neuropilin score than blocks. Asia-Pacific patients also tended to have a lower neuropilin score and demonstrated the least benefit with bevacizumab. Further investigation will be required to better interpret the apparent differences between Asia-Pacific and non-Asia-Pacific regions with regard to tumor neuropilin-1 expression.

Finally, the prognostic significance of HER2 overexpression or amplification in advanced gastric cancer is controversial. Although HER2 positivity is associated with a poor prognosis in breast cancer,24 studies on the prognostic value of HER2 status in gastric cancer have been contradictory.25–29 AVAGAST is the largest prospective study to examine the prognostic value of HER2 overexpression or amplification in gastric cancer and demonstrates that HER2 status is not prognostic in this disease. This is supported by the recent analysis by Shah et al,30 which also failed to establish HER2 as an independent prognostic factor for advanced gastric cancer. It should be noted that the low rate of HER2-positive tumors in this study (13%) was influenced by the concurrent ToGA trial6 and is unlikely to be an accurate estimate of the HER2-positive rate in the overall population of patients with gastric cancer.

The AVAGAST biomarker evaluation is part of an extensive ongoing biomarker program for bevacizumab and is notable by its prospective design, large sample size, lack of sampling bias, and high rate of biospecimen collection. Limitations include the differential results obtained from tumor samples provided as slides and blocks, which confounded the interpretation of some of our data. Loss of immunoreactivity is known to occur in paraffin-embedded tumor sections stored on slides31 and may explain this observation. In addition, the limitations of using IHC as a quantitative measure (eg, influenced by preanalytic tissue processing and subjective scoring) need to be acknowledged, especially with the use of nonstandardized tests. Other biomarker analyses from AVAGAST will include other plasma markers, RNA, and single-nucleotide polymorphisms and will be reported at a later date. Several other candidate markers have been reported for bevacizumab across other tumor types, including candidate single-nucleotide polymorphisms,32,33 tumor markers such as VEGF-D,34 and other circulating markers,14,35,36 and their value in AVAGAST has not been assessed.

In conclusion, plasma VEGF-A and tumor neuropilin-1 are strong biomarker candidates with the potential to predict clinical outcome in patients with advanced gastric cancer treated with bevacizumab. Prospective studies are required to further characterize these markers. Collection of data for plasma VEGF-A and tumor neuropilin-1 should be integrated prospectively into new gastric cancer studies of bevacizumab to validate the findings from AVAGAST.

Previous SectionNext Section

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Although all authors completed the disclosure declaration, the following author(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a “U” are those for which no compensation was received; those relationships marked with a “C” were compensated. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.

Employment or Leadership Position: Sanne de Haas, F. Hoffmann-La Roche (C); Bernd Langer, F. Hoffmann-La Roche (C); Paul Delmar, F. Hoffmann-La Roche (C); Stefan J. Scherer, Genentech (C) Consultant or Advisory Role: Yoon-Koo Kang, F. Hoffmann-La Roche (C); Niall C. Tebbutt, F. Hoffmann-La Roche (U); Manish A. Shah, Genentech (C) Stock Ownership: None Honoraria: Yoon-Koo Kang, F. Hoffmann-La Roche; Atsushi Ohtsu, Taiho Pharmaceutical, Chugai-Roche, Novartis Research Funding: Eric Van Cutsem, F. Hoffmann-La Roche; Niall C. Tebbutt, F. Hoffmann-La Roche Expert Testimony: None Other Remuneration: None

Previous SectionNext Section

AUTHOR CONTRIBUTIONS

Conception and design: Eric Van Cutsem, Yoon-Koo Kang, Atsushi Ohtsu, Bernd Langer, Stefan J. Scherer, Manish A. Shah

Provision of study materials or patients: Eric Van Cutsem, Yoon-Koo Kang, Atsushi Ohtsu, Niall C. Tebbutt, Jian Ming Xu, Wei Peng Yong, Manish A. Shah

Collection and assembly of data: Eric Van Cutsem, Sanne de Haas, Atsushi Ohtsu, Niall C. Tebbutt, Jian Ming Xu, Wei Peng Yong, Bernd Langer, Stefan J. Scherer, Manish A. Shah

Data analysis and interpretation: Eric Van Cutsem, Sanne de Haas, Yoon-Koo Kang, Atsushi Ohtsu, Niall C. Tebbutt, Bernd Langer, Paul Delmar, Stefan J. Scherer, Manish A. Shah

Manuscript writing: All authors

Final approval of manuscript: All authors

Previous SectionNext Section

Acknowledgment

In addition to the investigators in the author list, we would like to acknowledge the following investigators who also participated in this trial: Australia: P. Cooray, G. Marx, T. Price; Brazil: E.H. Costa e Silva, G. Delgado, M.M. De Queiróz, O. Gampel, P. Hoff, G. Ismael, D. Jendiroba, S. Jobim De Azevedo, A. Kruschewsky Coutinho, J. Nunes, L. Olivatto, C.S. Svend Chicata Sutmoller; China: J. Wang; Costa Rica: D. Otero Reyes; France: J. Bennouna Louridi, O. Bouche, F. Husseini, J.-P. Metges; Germany: D. Arnold, P.R. Galle, R. Hofheinz, F. Mayer, A. Vogel, G. von Wichert; Hong Kong: K.M. Chu; Italy: G. Aprile, C. Barone, A. Buonadonna, S. Cascinu, S. Chiara, M. di Bartolomeo, F. di Costanzo; Japan: K. Chin, T. Doi, Y. Hamamoto, W. Koizumi, Y. Komatsu, N. Boku, T. Nishina, Y. Omuro, S. Saji, T. Satoh, A. Sawaki, H. Takiuchi, Y. Yamada, K. Yamaguchi; Korea: Y.-J. Bang, Y.H. Kim, H.Y. Lim, S.R. Park, S.Y. Rha; Peru: L. Casanova, H. Gomez-Moreno, P. Pimentel, A-B. Renzo, F. Salas; Russian Federation: A. Garin, V. Gorbunova, E. Gotovkin, R. Khasanov, M.V. Kopp, E. Kulikov, O. Lipatov, G.M. Manikhas, L. Roman, D. Viushkov, L. Vladimirova; Singapore: S.P. Choo; Spain: J. Bellmunt, M. Martin, F. Rivera, J. Sastre, J. Tabernero; Taiwan: C.-P. Li, K.-H. Yeh; United Kingdom: S. Falk, S. Gollins; United States: S. Dakhil, J.D. Hainsworth, H.-J. Lenz, C. Lobo, R. Mena, T. Reid.

Previous SectionNext Section

Appendix

⇓⇓⇓

View this table:
Table A1.

Immunohistochemistry Antibodies Used to Test Biomarkers in Tumor Tissue Samples

View this table:
Table A2.

Biomarker Expression Levels at Baseline (biomarker populations)

View this table:
Table A3.

Correlation Between Baseline Characteristics and Baseline Biomarker Expression Levels (biomarker populations)

Previous SectionNext Section

Footnotes

  • Supported by Genentech, F. Hoffmann-La Roche, and Chugai Pharmaceutical. Third-party writing assistance for this article was supported by F. Hoffmann-La Roche.

  • Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.

  • Clinical trial information can be found for the following: NCT00548548.

  • Received October 12, 2011.
  • Accepted February 1, 2012.
Previous Section
 

REFERENCES

  1. ↵
    1. Hurwitz H,
    2. Fehrenbacher L,
    3. Novotny W,
    4. et al.
    (2004) Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 350:2335–2342.
    CrossRefMedline
    1. Escudier B,
    2. Pluzanska A,
    3. Koralewski P,
    4. et al.
    (2007) Bevacizumab plus interferon alfa-2a for treatment of metastatic renal cell carcinoma: A randomised, double-blind phase III trial. Lancet 370:2103–2111.
    CrossRefMedline
    1. Friedman HS,
    2. Prados MD,
    3. Wen PY,
    4. et al.
    (2009) Bevacizumab alone and in combination with irinotecan in recurrent glioblastoma. J Clin Oncol 27:4733–4740.
    Abstract/FREE Full Text
    1. Miller K,
    2. Wang M,
    3. Gralow J,
    4. et al.
    (2007) Paclitaxel plus bevacizumab versus paclitaxel alone for metastatic breast cancer. N Engl J Med 357:2666–2676.
    CrossRefMedline
  5. ↵
    1. Reck M,
    2. von Pawel J,
    3. Zatloukal P,
    4. et al.
    (2009) Phase III trial of cisplatin plus gemcitabine with either placebo or bevacizumab as first-line therapy for nonsquamous non-small-cell lung cancer: AVAil. J Clin Oncol 27:1227–1234.
    Abstract/FREE Full Text
  6. ↵
    1. Bang YJ,
    2. Van Cutsem E,
    3. Feyereislova A,
    4. et al.
    (2010) 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 376:687–697.
    CrossRefMedline
  7. ↵
    1. Piccart-Gebhart MJ,
    2. Procter M,
    3. Leyland-Jones B,
    4. et al.
    (2005) Trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer. N Engl J Med 353:1659–1672.
    CrossRefMedline
  8. ↵
    1. Ohtsu A,
    2. Shah MA,
    3. Van Cutsem E,
    4. et al.
    (2011) Bevacizumab in combination with chemotherapy as first-line therapy in advanced gastric cancer: A randomized double-blind placebo-controlled phase III study. J Clin Oncol 29:3968–3976.
    Abstract/FREE Full Text
  9. ↵
    1. Claudon A,
    2. Vergnaud P,
    3. Valverde C,
    4. et al.
    (2008) New automated multiplex assay for bone turnover markers in osteoporosis. Clin Chem 54:1554–1563.
    Abstract/FREE Full Text
  10. ↵
    1. Ishibashi H,
    2. Suzuki T,
    3. Suzuki S,
    4. et al.
    (2003) Sex steroid hormone receptors in human thymoma. J Clin Endocrinol Metab 88:2309–2317.
    Abstract/FREE Full Text
  11. ↵
    1. Hofmann M,
    2. Stoss O,
    3. Shi D,
    4. et al.
    (2008) Assessment of a HER2 scoring system for gastric cancer: Results from a validation study. Histopathology 52:797–805.
    CrossRefMedline
  12. ↵
    1. Therasse P,
    2. Arbuck SG,
    3. Eisenhauer EA,
    4. et al.
    (2000) New guidelines to evaluate the response to treatment in solid tumors: European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 92:205–216.
    Abstract/FREE Full Text
  13. ↵
    1. Murukesh N,
    2. Dive C,
    3. Jayson GC
    (2010) Biomarkers of angiogenesis and their role in the development of VEGF inhibitors. Br J Cancer 102:8–18.
    CrossRefMedline
  14. ↵
    1. Bernaards C,
    2. Hegde P,
    3. Chen D,
    4. et al.
    (2010) Circulating vascular endothelial growth factor (VEGF) as a biomarker for bevacizumab-based therapy in metastatic colorectal, non-small cell lung, and renal cell cancers: Analysis of phase III studies. J Clin Oncol 28(suppl; abstr 10519):727s.
    Search Google Scholar
    1. Leighl N,
    2. Reck M,
    3. de Haas S,
    4. et al.
    (2009) Analysis of biomarkers (BMs) in the AVAiL phase III randomised study of first-line bevacizumab (Bv) with cisplatin-gemcitabine (CG) in patients (pts) with non-small cell lung cancer (NSCLC) Eur J Cancer Suppl 7:558, abstr P-9172.
    Search Google Scholar
  17. ↵
    1. Holden SN,
    2. Ryan E,
    3. Kearns A,
    4. et al.
    (2005) Benefit from bevacizumab (BV) is independent of pretreatment plasma vascular endothelial growth factor-A (pl-VEGF) in patients (pts) with metastatic colorectal cancer (mCRC) J Clin Oncol 23(suppl; abstr 3555):166s.
    Search Google Scholar
  18. ↵
    1. Miles DW,
    2. de Haas SL,
    3. Dirix L,
    4. et al.
    33rd Annual San Antonio Breast Cancer Symposium (December 10, 2010, San Antonio, TX), Plasma biomarker analyses in the AVADO phase III randomized study of first-line bevacizumab + docetaxel in patients with human epidermal growth factor receptor (HER) 2-negative metastatic breast cancer. abstr P2-16-04.
    Search Google Scholar
  19. ↵
    1. Van Cutsem E,
    2. Jayson G,
    3. Dive C,
    4. et al.
    (2011) Analysis of blood plasma factors in the AVITA phase III randomized study of bevacizumab (bev) with gemcitabine-erlotinib (GE) in patients (pts) with metastatic pancreatic cancer (mPC) Eur J Cancer 47:S95–S96, abstr 803.
    Search Google Scholar
  20. ↵
    1. Jayson GC,
    2. de Haas S,
    3. Delmar P,
    4. et al.
    Proceedings of the 2011 European Multidisciplinary Cancer Congress (September 23-27, 2011, Stockholm, Sweden), Evaluation of plasma VEGF-A as a potential predictive pan-tumour biomarker for bevacizumab. abstr 804.
    Search Google Scholar
  21. ↵
    1. Foernzler D,
    2. Delmar P,
    3. Kockx M,
    4. et al.
    Presented at 2010 ASCO Gastrointestinal Cancers Symposium (January 22-24, 2010, Orlando, FL), Tumor tissue based biomarker analysis in NO16966: A randomized phase III study of first-line bevacizumab in combination with oxaliplatin-based chemotherapy in patients with mCRC. abstr 374.
    Search Google Scholar
  22. ↵
    1. Jubb AM,
    2. Miller KD,
    3. Rugo HS,
    4. et al.
    (2011) Impact of exploratory biomarkers on the treatment effect of bevacizumab in metastatic breast cancer. Clin Cancer Res 17:372–381.
    Abstract/FREE Full Text
  23. ↵
    1. Gaur P,
    2. Bielenberg DR,
    3. Samuel S,
    4. et al.
    (2009) Role of class 3 semaphorins and their receptors in tumor growth and angiogenesis. Clin Cancer Res 15:6763–6770.
    Abstract/FREE Full Text
  24. ↵
    1. Kaptain S,
    2. Tan LK,
    3. Chen B
    (2001) Her-2/neu and breast cancer. Diagn Mol Pathol 10:139–152.
    CrossRefMedline
  25. ↵
    1. Uchino S,
    2. Tsuda H,
    3. Maruyama K,
    4. et al.
    (1993) Overexpression of c-erbB-2 protein in gastric cancer: Its correlation with long-term survival of patients. Cancer 72:3179–3184.
    CrossRefMedline
    1. Park DI,
    2. Yun JW,
    3. Park JH,
    4. et al.
    (2006) HER-2/neu amplification is an independent prognostic factor in gastric cancer. Dig Dis Sci 51:1371–1379.
    CrossRefMedline
    1. Begnami MD,
    2. Fukuda E,
    3. Fregnani JH,
    4. et al.
    (2011) Prognostic implications of altered human epidermal growth factor receptors (HERs) in gastric carcinomas: HER2 and HER3 are predictors of poor outcome. J Clin Oncol 29:3030–3036.
    Abstract/FREE Full Text
    1. Tateishi M,
    2. Toda T,
    3. Minamisono Y,
    4. et al.
    (1992) Clinicopathological significance of c-erbB-2 protein expression in human gastric carcinoma. J Surg Oncol 49:209–212.
    CrossRefMedline
  29. ↵
    1. Sasano H,
    2. Date F,
    3. Imatani A,
    4. et al.
    (1993) Double immunostaining for c-erbB-2 and p53 in human stomach cancer cells. Hum Pathol 24:584–589.
    CrossRefMedline
  30. ↵
    1. Shah MA,
    2. Janjigian YY,
    3. Pauligk C,
    4. et al.
    (2011) Prognostic significance of human epidermal growth factor-2 (HER2) in advanced gastric cancer: A U.S. and European international collaborative analysis. J Clin Oncol 29(suppl; abstr 4014):259s.
    Search Google Scholar
  31. ↵
    1. Jacobs TW,
    2. Prioleau JE,
    3. Stillman IE,
    4. et al.
    (1996) Loss of tumor marker-immunostaining intensity on stored paraffin slides of breast cancer. J Natl Cancer Inst 88:1054–1059.
    Abstract/FREE Full Text
  32. ↵
    1. Schneider BP,
    2. Wang M,
    3. Radovich M,
    4. et al.
    (2008) Association of vascular endothelial growth factor and vascular endothelial growth factor receptor-2 genetic polymorphisms with outcome in a trial of paclitaxel compared with paclitaxel plus bevacizumab in advanced breast cancer: ECOG 2100. J Clin Oncol 26:4672–4678.
    Abstract/FREE Full Text
  33. ↵
    1. Schultheis AM,
    2. Lurje G,
    3. Rhodes KE,
    4. et al.
    (2008) Polymorphisms and clinical outcome in recurrent ovarian cancer treated with cyclophosphamide and bevacizumab. Clin Cancer Res 14:7554–7563.
    Abstract/FREE Full Text
  34. ↵
    1. Weickhardt AJ,
    2. Williams D,
    3. Lee C,
    4. et al.
    (2011) Vascular endothelial growth factors (VEGF) and VEGF receptor expression as predictive biomarkers for benefit with bevacizumab in metastatic colorectal cancer (mCRC): Analysis of the phase III MAX study. J Clin Oncol 29(suppl; abstr 3531):228s.
    CrossRef
  35. ↵
    1. Nixon AB,
    2. Pang H,
    3. Starr M,
    4. et al.
    (2011) Prognostic and predictive blood-based biomarkers in patients with advanced pancreatic cancer: Results from CALGB 80303. J Clin Oncol 29(suppl; abstr 10508):632s.
    CrossRef
  36. ↵
    1. Liu Y,
    2. Tran HT,
    3. Lin Y,
    4. et al.
    (2011) Circulating baseline plasma cytokines and angiogenic factors (CAF) as markers of tumor burden and therapeutic response in a phase III study of pazopanib for metastatic renal cell carcinoma (mRCC) J Clin Oncol 29(suppl; abstr 4553):302s.
    Search Google Scholar
| Table of Contents

This Article

  1. Published online before print May 7, 2012, doi: 10.1200/JCO.2011.39.9824 JCO vol. 30 no. 17 2119-2127
  1. Abstract
  2. » Full Text
  3. PDF
  4. Protocol

Classifications

Navigate This Article

Current Issue

  1. July 20, 2013, 31 (21)
  1. Alert me to new issues of JCO
    • Advertisement
    • Advertisement
    • Advertisement