Molecular Analysis of the IDEAL/INTACT trials: EGFR Mutations and Gene Amplification

Reviewer: S. Jack Wei, MD
University of Pennsylvania School of Medicine
Last Modified: May 15, 2005

Presenter: T.J. Lynch
Presenter's Affiliation: Massachusetts General Hospital Cancer Center
Type of Session: Scientific


  • Previous studies have shown that patients with non-small cell lung cancer (NSCLC) who have mutations in the gene for epidermal growth factor receptor (EGFR) have dramatic response and improved survival when treated with gefitinib (Iressa), a small molecular inhibitor of the tyrosine kinase domain of EGFR.
  • The demographics of patients who are more likely to have EGFR mutations mirror patients who showed improved survival in phase II and III trials of gefitinib.
  • The IDEAL and INTACT trials were large randomized trials examining the role of gefitinib in the treatment stage IIIB and IV NSCLC as first-line and salvage chemotherapy.
  • The IDEAL 1 and IDEAL 2 trials randomized patients who had received 1 or 2, or >2 chemotherapy regimens, respectively, to two different doses of gefitinib (250 mg qd vs. 500 mg qd).
  • The INTACT 1 and INTACT 2 trials took previously untreated patients who received either gemcitabine (1250mg/m2)/cisplatin (80 mg/m2) or paclitaxel (225mg/m2)/carboplatin (AUC=6), respectively, and randomized them to chemotherapy + gefitinib (500 mg qd), chemotherapy + gefitinib (250 mg qd), or chemotherapy + placebo.
  • A retrospective analysis of the patients enrolled in these trials was undertaken to examine whether EGFR overexpression and mutations correlated with improved outcomes.

Materials and Methods

  • Paraffin blocks of either the original resection specimen or biopsy were obtained.
  • Microdissection of the samples was performed until the samples were >50% tumor cells.
  • Bi-directional sequencing of exons 18-21 of EGFR was performed and determination of the EGFR gene copy number was performed using Taqman quantitative real-time PCR.
  • Overall, mutations in exon 19 and 21 were the most common mutations seen from these samples, and the mutational analysis focused on mutations in these two exons.
  • The mutational status and gene copy number was correlated with response to gefitinib in the IDEAL and INTACT trials.


IDEAL Analysis
Mutational status
  • 79 evaluable specimens were obtained for mutational analysis from the IDEAL specimens.
  • 14 mutations were detected in these samples representing 18% of the evaluable patients.
  • 57% of the patients with mutations were women, and adenocarcinoma represented 86% of patients with mutations.  There was one patient with squamous cell histology whose tumor exhibited an EGFR mutation.
  • 46% of patients with mutations responded to gefitinib compared to 10% of patients with wild-type EGFR (p=0.005).
  • Patients with mutations had improved progression-free survival (PFS) (p<0.05).
  • Patients with mutations did not show improved overall survival (OS).
Amplification status
  • 80 patients were evaluable for EFGR amplification
  • 7 patients were found to have amplified EGFR (8% of evaluable patients).
  • Amplification was 4 to 90 fold in these patients.
  • 29% of patients with gene amplification responded to gefintib compared to 15% of patients with no amplification (p=0.319).
  • Patients with gene amplification showed an non-statistically significant improvement in PFS.
  • There was no improvement in OS for patients with gene amplification.
INTACT Analysis
Mutational status
  • 32 patients with mutations were detected representing 10.2% of the evaluable patients
  • 13/18 (72%) of patients with mutations who received gefitinib showed a response compared to 84/152 (55%) of patients with wild-type EGFR.
  • Patients with mutations showed a statistically non-significant improvement in PFS compared to patients with wild-type EGFR
Amplification status
  • 33 patients with amplification were found representing 7% of evaluable patients.
  • Mutational data was available on 14 patients with amplification, and 80% of patients with amplification were found to have wild-type EGFR.
  • Patients with EGFR amplification showed a statistically non-significant improvement in PFS compared to patients with no amplification.
Additional Analysis of the Data Set
  • The patients who have mutations are not the same patients as those with overexpression.
    • 25.5% of patients non-smokers had mutations compared to 6.3% who had amplifications (p<0.05).
    • 17.4% of patients with adenocarcinoma or bronchalveolar carcinoma had mutations compared to 6.9% who had amplifications (p<0.05).
    • 18.5% of Asian patients exhibited mutations compared to 5.1% of Asian patients who had amplifications (p<0.05).
    • 14.3% of patients less than age 64 exhibited mutations compared to 6.6% of these patients who had amplifications (p<0.05).
    • 18.5% of patients over age 64 had gene amplifications compared to 6.8% of these patients who had mutations (p<0.05).
  • There was no direct correlation with EGFR gene amplification and overexpression of the EGFR protein.

Author's Conclusions

  • EGFR mutation and amplification identify different groups of patients with the population of patients with mutations mirroring those that have been found to have significant benefit from therapy with gefitinib.
  • EGFR mutations were associated with improved outcomes.
  • EGFR mutations predict for improved response and PFS from gefitinib in the IDEAL trials.
  • EGFR amplification may identify another group of patients who benefit from therapy with gefitinib.
  • The small number of patient samples in this study limit conclusions regarding impact of mutational status and amplifications on OS.
  • Molecular profiling is essential in these patients.
  • Future trials should focus on patients with EGFR mutations and examine the role of tyrosine kinase inhibitors such as gefitinib in first-line studies of metastatic disease as well as in adjuvant trials after surgery or radiotherapy.

Clinical/Scientific Implications

The above analysis attempts to define the subset of patients who benefited from gefitinib in the IDEAL and INTACT trials.  Unfortunately, the analysis is severely limited by the small number of patients for whom the investigators were able to obtain evaluable tissue samples.  The four trials together enrolled thousands of patients, and it is possible that given the small number of patients that were evaluated in this analysis, a selected subset of patients were analyzed in this study from the overall study populations.  Nevertheless, several interesting conclusions can be drawn from this study.  First, it appears that EGFR mutation is more consistent with the populations of patients that benefit from gefitinib therapy in general.  Specific mutations within the tyrosine kinase domain of EGFR have been reported in the past year, and this study seems to support these findings.  Secondly, it does appear that patients with mutations are not the same patients who have amplifications.  Although the outcome data of patients with gene amplifications should be viewed with caution given the small number of patients in this study, they did appear to have improved, albeit non-significant, response rates and PFS compared to patients without gene amplification.  Overall, this study supports the notion that specific subsets of populations can be identified based on genetic analysis who will respond to targeted treatment.  However, specific conclusions about exactly what those patient subsets are cannot be determined from this study.