Biomarker analysis in oesophagogastric cancer: Results from the REAL3 and TransMAGIC trials

https://doi.org/10.1016/j.ejca.2013.02.007Get rights and content

Abstract

Background

REAL3 (Randomised ECF for Advanced or Locally advanced oesophagogastric cancer 3) was a phase II/III trial designed to evaluate the addition of panitumumab (P) to epirubicin, oxaliplatin and capecitabine (EOC) in untreated advanced oesophagogastric adenocarcinoma, or undifferentiated carcinoma. MAGIC (MRC Adjuvant Gastric Infusional Chemotherapy) was a phase III study which demonstrated that peri-operative epirubicin, cisplatin and infused 5-fluourouracil (ECF) improved survival in early oesophagogastric adenocarcinoma.

Patients and methods

Analysis of response rate (RR; the primary end-point of phase II) and biomarkers in the first 200 patients randomised to EOC or modified dose (m) EOC + P in REAL3 was pre-planned to determine if molecular selection for the on-going study was indicated. KRAS, BRAF and PIK3CA mutations and PTEN expression were assessed in pre-treatment biopsies and results correlated with response to mEOC + P. Association between these biomarkers and overall survival (OS) was assessed in MAGIC patients to determine any prognostic effect.

Results

RR was 52% to mEOC + P, 48% to EOC. Results from 175 assessable biopsies: mutations in KRAS (5.7%), BRAF (0%), PIK3CA (2.5%) and loss of PTEN expression (15.0%). None of the biomarkers evaluated predicted resistance to mEOC + P. In MAGIC, mutations in KRAS, BRAF and PIK3CA and loss of PTEN (phosphatase and tensin homolog) were found in 6.3%, 1.0%, 5.0% and 10.9%, respectively, and were not associated with survival.

Conclusions

The RR of 52% in REAL3 with mEOC + P met pre-defined criteria to continue accrual to phase III. The frequency of the mutations was too low to exclude any prognostic or predictive effect.

Introduction

Epidermal growth factor receptor (EGFR) is commonly over-expressed and/or amplified in human cancers, and capable of transforming cells in vitro.1 EGFR has been proven to be a clinically useful therapeutic target in EGFR-mutated lung cancer,2 head and neck3 and metastatic colorectal (mCRC) cancers.4 The efficacy of the anti-EGFR antibodies in mCRC is limited to patients without activating mutations in KRAS.4, 5 Mutations in BRAF and PIK3CA exon 20 are also predictive of resistance to cetuximab in chemotherapy-resistant mCRC.6

REAL3 was a phase II/III trial designed to evaluate the addition of panitumumab, a fully human IgG2 monoclonal antibody directed against EGFR, to EOC chemotherapy, initially in a molecularly unselected group of patients with advanced oesophagogastric cancer. Standard dose EOC combined with panitumumab resulted in unacceptable toxicity, leading to modification of the capecitabine and oxaliplatin doses, reported previously.7 The phase II analysis was conducted to exclude futility of the combination and assess potential predictive biomarkers for panitumumab efficacy for phase III.

In advanced gastric cancer, KRAS mutations are uncommon,8 and no association with response to cetuximab has been described.9, 10, 11, 12, 13 PIK3CA mutations have been reported in 16% of gastric cancers14 whereas BRAF mutations are rare, with a rate of 2·2% described in the largest series.8 Loss of expression of PTEN, is a further putative biomarker of resistance to anti-EGFR antibodies in mCRC,15 and is common in gastric cancer.16

Given the evidence available in support of the predictive role of KRAS, BRAF, PIK3CA and PTEN in mCRC, these were assessed in the phase II analysis of REAL3; to determine whether biomarkers used clinically for the selection of patients with mCRC to receive anti-EGFR therapies would be useful for the selection of patients to be enrolled in the phase III of REAL3. As survival was not an end-point of the phase II analysis, prognostic effect could not be assessed; therefore similar analyses were performed in patient-samples from the previously reported MAGIC trial,17 which compared surgery alone to surgery and peri-operative ECF in operable oesophagogastric cancer, to assess the prognostic impact of the biomarkers. It was hypothesised that biomarkers of poor prognosis would be more common in patients with advanced versus early disease, therefore comparison of frequencies in the MAGIC and REAL3 cohorts could theoretically provide information in support of their prognostic value.

Section snippets

Patients

Full eligibility criteria for REAL3 (NCT00824785) and MAGIC have been reported elsewhere.7, 17 All REAL3 patients gave written informed consent to the study and were required to give additional consent to be included in the biomarker study. Patients gave written informed consent to the MAGIC study, which included pathological review of tissue. Both trials had approval by the required Research Ethics Committees (REC) (REAL3: North West REC 08/H1010/6, TransMAGIC: South East REC, 11/LO/0566).

Toxicity in REAL3

Two hundred patients from 40 UK oncology centres were randomised to mEOC + P (n = 100) or EOC (n = 100) between May 2009 and September 2010. Toxicity results are recorded in Supplementary Table 4.

Response rate

Eleven mEOC + P and 14 EOC patients who withdrew or died prior to radiological (or clinical) assessment of response are analysed as no response. Overall RR in 100 patients randomised to mEOC + P was 52% (95% confidence interval (CI) 42–62%), comprising complete response (CR) in three patients, partial response

Discussion

To our knowledge, this study is the largest prospective evaluation of potential biomarkers of resistance to an anti-EGFR antibody in oesophagogastric cancer, however no association was found between the evaluated biomarkers and RR to mEOC + P. RR was similar in KRAS wild-type and mutant patients; however mutations were infrequent, consistent with previous reports of gastric cancer patients treated with chemotherapy plus cetuximab.10, 13, 27, 28 This contrasts the effect of KRAS mutations in mCRC4

Grant support

REAL3 is sponsored by the Royal Marsden NHS Foundation Trust and supported by a research Grant and provision of panitumumab from Amgen Ltd. REAL3 is endorsed by CRUK (CRUKE/07/049). The study was designed by Investigators at the Royal Marsden Hospital and the protocol was approved by Amgen Ltd. prior to submission for regulatory and ethics approvals. TransMAGIC is funded by CRUK (Grant reference No. C20023/A7217), with support from the Medical Research Council through the MRC Clinical Trials

Financial support

J.S. Reis-Filho is funded in part by Breakthrough Breast Cancer. Ruth Langley, Lindsay Thompson and Sally Stenning are employed by the Medical Research Council.

Role of the funding source

Amgen Ltd. had no role in collecting or analysing data, writing the report or the decision to submit.

Conflict of interest statement

Dr. Okines has previously received an honorarium and travel support from Roche and travel support from Amgen and Bayer. Professor Cunningham has served on an advisory board for Roche and Amgen, acted as expert witness for Amgen (uncompensated) and received research funding from Amgen, Merck Serono, Roche and Sanofi Aventis. Dr. Coxon has received research funding from Roche, Professor Ferry has received an honorarium and research funding from Amgen, plus honoraria and travel support from Roche

Acknowledgements

The REAL3 and TransMAGIC study teams acknowledge the support of the National Institute for Health Research, through the National Cancer Research Network and NHS Funding to the NIHR Biomedical Research Centre.

We would like to thank all the REAL3 and MAGIC trial principal investigators, histopathologists, research nurses, trial coordinators and patients for their important contributions to this work.

We thank Clare Peckitt and Swaranali Gupta for providing additional statistical support for REAL3.

References (36)

  • A.F. Okines et al.

    Epirubicin, oxaliplatin, and capecitabine with or without panitumumab for advanced esophagogastric cancer: dose-finding study for the prospective multicenter, randomized, phase II/III REAL-3 trial

    J Clin Oncol

    (2010)
  • S.H. Lee et al.

    BRAF and KRAS mutations in stomach cancer

    Oncogene

    (2003)
  • F. Lordick et al.

    Cetuximab plus oxaliplatin/leucovorin/5-fluorouracil in first-line metastatic gastric cancer: a phase II study of the Arbeitsgemeinschaft Internistische Onkologie (AIO)

    Br J Cancer

    (2010)
  • S.R. Park et al.

    Predictive factors for the efficacy of cetuximab plus chemotherapy as salvage therapy in metastatic gastric cancer patients

    Cancer Chemother Pharmacol

    (2010)
  • C. Pinto et al.

    Phase II study of cetuximab in combination with cisplatin and docetaxel in patients with untreated advanced gastric or gastro-oesophageal junction adenocarcinoma (DOCETUX study)

    Br J Cancer

    (2009)
  • D.A. Richards et al.

    Results of docetaxel plus oxaliplatin (DOCOX) with or without cetuximab in patients with metastatic gastric and/or gastroesophagel junction adenocarcinoma: Results of a randomized phase II study

    J Clin Oncol

    (2011)
  • S. Barbi et al.

    The analysis of PIK3CA mutations in gastric carcinoma and metanalysis of literature suggest that exon-selectivity is a signature of cancer type

    J Exp Clin Cancer Res

    (2010)
  • P. Laurent-Puig et al.

    Analysis of PTEN, BRAF, and EGFR status in determining benefit from cetuximab therapy in wild-type KRAS metastatic colon cancer

    J Clin Oncol

    (2009)
  • Cited by (0)

    o

    On behalf of the NCRI Upper GI Clinical Studies Group.

    View full text