ESMO 2014: Continuation of Gefitinib Plus Pemetrexed/Cisplatin of No Clinical Benefit in NSCLC Patients with Acquired Resistance to Gefitinib

IMPRESS study indicates that doublet chemotherapy should remain a standard of care

The IMPRESS study is the first and only randomised phase III trial to confirm that continuation of gefitinib in addition to pemetrexed/cisplatin would be of no clinical benefit for patients with non-small cell lung cancer (NSCLC) and acquired resistance to gefitinib. The results were reported during the Presidential Symposium 1 at ESMO 2014 Congress, Madrid, Spain by Prof. Tony Mok of the Clinical Oncology Department, The Chinese University of Hong Kong, Hong Kong, China.

The study design was previously presented at the 10th Annual Meeting of the Japanese Society of Medical Oncology, 26–28 July 2012, Osaka, Japan. The latest findings presented as a late breaking abstract at ESMO 2014 Congress report on previously unpublished primary and secondary analysis data.

Most patients with epidermal growth factor receptor (EGFR) mutation-positive NSCLC respond to first-line EGFR tyrosine kinase inhibitors (TKIs), but later acquire resistance. Optimal management for patients with acquired resistance has yet to be defined, and options include:

  • Discontinuing EGFR TKI therapy and commencing platinum-based doublet chemotherapy
  • Continuing EGFR TKI therapy in combination with platinum-based doublet chemotherapy.

The  second option is suggested to be beneficial because of the potential tumour heterogeneity at the time of EGFR TKI resistance, and it is also supported by retrospective clinical studies.

The phase III, double-blind IRESSA Mutation Positive Multicentre Treatment Beyond ProgRESsion Study (IMPRESS) evaluated the efficacy and safety of continuing gefitinib plus pemetrexed/cisplatin vs placebo plus pemetrexed/cisplatin in patients with acquired resistance to first-line gefitinib.

Patients of or older than 18 years (in Japan ≥20 years) who were chemotherapy-naive, and who had cytological or histological confirmation of locally advanced/metastatic NSCLC other than predominantly squamous cell histology with an activating EGFR mutation as determined locally, and in whom a prior disease progression was recorded on first-line treatment with gefitinib, were eligible for the study.

Exclusion criteria included prior chemotherapy or other systemic anti-cancer treatment (excluding gefitinib); palliative bone radiotherapy was required to be completed at least 2 weeks before start of study treatment with no persistent radiation toxicity; past medical history of interstitial lung disease, drug-induced interstitial disease, radiation pneumonitis which required steroid treatment or any evidence of clinically active interstitial lung disease; other co-existing malignancies or malignancies diagnosed within the last 5 years, with the exception of basal cell carcinoma or cervical cancer in situ or completely resected intramucosal gastric cancer; any evidence of severe of uncontrolled systemic disease; and treatment with an investigational drug within 4 weeks before randomisation.

Patients from 71 centres in Europe and Asia Pacific were randomised to gefitinib or placebo. The primary endpoint was progression-free survival (PFS). Secondary endpoints included overall survival (OS), objective response rate (ORR), disease control rate (DCR), safety/tolerability, health-related quality of life. Exploratory analysis of biomarkers is not yet completed and will be reported separately.

The study researchers estimated that 250 patients would need to be randomised to achieve 190 PFS events (75% PFS maturity); 90% power to demonstrate superiority of gefitinib in combination with chemotherapy vs chemotherapy alone; and 2-sided 5% significance for assuming a hazard ratio (HR) of 0.63 for median PFS of 9.5 months vs 6.0 months.

Randomisation did not include stratification factors, but analysis was adjusted for two covariates: age (<65 versus ≥65 years) and prior response to gefitinib (stable disease vs partial response plus complete response).

From March 2014 to December 2013, 265 patients were randomised, 133 in the gefitinib arm and 132 in the placebo arm. Median follow-up in the study was 11.2 months.

The patient demographics in the two arms were well balanced. However, there were more patients ≥65 years in gefitinib arm and more patients with baseline brain metastases in the chemotherapy arm.

There was no statistically significant improvement in PFS for gefitinib vs placebo; hazard ratio (HR) 0.86; p = 0.273. Median PFS was 5.4 months in each arm.

The OS was immature (33% of patients had died), with better OS for placebo vs gefitinib (HR 1.62; p = 0.029).

Ad-hoc PFS and OS analyses included the addition of brain metastases at baseline as a covariate (brain metastases vs no brain metastases), but there was no difference in term of PFS.

No treatment differences were found in ORR and DCR.

The safety profile for gefitinib plus pemetrexed/cisplatin was in line with what is already known.

The most common adverse events in the safety population (132 patients in each arm) were nausea recorded in 64% in the gefitinib group and 61% in the placebo group and decreased appetite (49% in the gefitinib treated patients vs 34% in the placebo arm). No interstitial lung disease was noted. Gefitinib was associated with increased grade 1/2 gastrointestinal toxicities.

Adverse events with outcome of death were reported, in particular 2 casually-related in the gefitinib/chemotherapy arm and 1 casually-related in placebo/chemotherapy group.

Post-discontinuation therapy in intent to treat population was higher in placebo arm, where 17% of patients received platinum based regimens in comparison to 5% in the gefitinib arm, and 44% received EGFR TKI vs 30% of patients in the gefitinib arm.


Prof. Mok concluded that continuation of gefitinib in addition to pemetrexed/cisplatin would be of no clinical benefit for patients with acquired resistance to gefitinib. The IMPRESS study showed no statistically significant improvement in PFS with continuation of gefitinib in addition to chemotherapy beyond RECIST progression to first-line EGFR TKI for patients with EGFR mutation-positive NSCLC. The OS was immature and not conclusive. There were imbalances in post study treatment in favour of the placebo arm (more doublet platinum chemotherapies and more EGFR TKI rechallenge).

Dr Solange Peters, who discussed the study results, said that there are multiple mechanisms of resistance to EGFR TKIs reported in pre-clinical and clinical trials (MET amplification, HGF overexpression, PIK3CA mutations, PTEN loss, FGFR1/2/3 overexpression, AXL overexpression, CRKL amplification, NFkB activation, BRAF mutation, anti-apoptotic pathway (BIM deletion), loss of EGFR mutant gene, HER2 amplification, PDGFRb expression, and EMT). T790M mutation causes drug resistance by increasing the affinity for ATP. Dynamics of resistance evolution and the question of heterogeneity add to complexity of the problem.

Upon discussing treatment options at EGFR TKI resistance (chemotherapy, immunotherapy, EGFR TKI beyond progression, second generation TKI, third generation TKI, targeting bypass tracks, and chemotherapy plus EGFR TKI - which was tested in the IMPRESS study), she concluded that the IMPRESS study confirms that doublet chemotherapy should continue to be the standard of care for patients who develop resistance to first-line gefitinib. According to her, first-line EGFR TKI should be continued as long as possible and EGFR TKI should be subsequently rechallenged in the course of the disease.


The study was sponsored by AstraZeneca.