Articles
Outcomes in patients with non-small-cell lung cancer and acquired Thr790Met mutation treated with osimertinib: a genomic study

https://doi.org/10.1016/S2213-2600(17)30480-0Get rights and content

Summary

Background

Osimertinib is approved for the treatment of non-small-cell lung cancer in patients who develop the EGFR Thr790Met mutation after treatment with epidermal growth factor receptor (EGFR) tyrosine-kinase inhibitors (TKIs). We assessed outcomes in patients with non-small-cell lung cancer and the EGFR Thr790Met mutation who were treated with osimertinib, a third-generation EGFR TKI, after previous treatment failure with one or more other EGFR TKIs.

Methods

Eligible patients had been enrolled at one centre in the AURA study, had shown resistance to a previous EGFR TKI, and had EGFR-activating mutations and acquired Thr790Met mutation detectable in tumour tissue or plasma. Patients took 20–240 mg osimertinib per day until disease progression or development of intolerable side-effects. Plasma samples were collected every 6 weeks and tumour tissue biopsy was done at study entry and was optional after disease progression. We tested samples for resistance mechanisms, including EGFR-activating, Thr790Met, and Cys797Ser mutations, and assessed associations with overall survival, progression-free survival, and survival after disease progression.

Findings

Of 71 patients enrolled in AURA, 53 were eligible for this analysis. Median progression-free survival was 11·1 months (95% CI 8·4–13·9) and overall survival was 16·9 months (11·7–29·1). 47 patients had disease progression. Median overall survival after osimertinib progression was 5·4 months (95% CI 4·1–10·0). Plasma samples were available for 40 patients after disease progression. 12 (30%) of these had the Thr790Met mutation (four of whom also had Cys797Ser mutations). Patients without detectable EGFR-activating mutations in plasma before treatment had the best overall and post-progression survival (22·4 months, 95% CI 15·6–not reached, and 10·8 months, 7·2–not reached, respectively). Loss of the Thr790Met mutation but presence of EGFR-activating mutations in plasma were associated with the shortest progression-free survival (median 2·6 months, 95% CI 1·3–not reached). In 22 post-progression tumour samples, we found one squamous cell and two small-cell transformations. We detected Thr790Met in nine (50%) of 18 samples, Cys797Ser in two (17%) of 12, cMET amplification in five (50%) of ten, BRAF mutation in one (8%) of 13, and KRAS mutation in one (8%) of 13.

Interpretation

Heterogeneous resistance mechanisms developed in patients receiving osimertinib. Differences in resistance mechanisms might dictate future development strategies for osimertinib in clinical trials.

Funding

AstraZeneca, Taiwan Ministry of Science and Technology.

Introduction

Osimertinib is an epidermal growth factor receptor (EGFR) tyrosine-kinase inhibitor (TKI) designed to inhibit EGFR-activating mutations (exon 19 deletion and Leu858Arg) and the Thr790Met resistance mutation. It has high anticancer activity against EGFR mutations (exon 19 deletion, Leu858Arg, and Thr790Met) but low activity against wild-type EGFR.1 In the AURA and AURA2 studies,2, 3, 4 osimertinib has had high anticancer activity, with treatment responses seen in 61–70% of patients with EGFR-mutant non-small-cell lung cancer previously resistant to EGFR TKI treatment and acquired EGFR Thr790Met mutation. In a similar group of patients in the AURA3 trial,5 progression-free survival was longer with osimertinib than with pemetrexed and platinum (8·5 vs 4·2 months) and more patients responded to treatment (71% vs 31%), although disease progression seems to be inevitable after treatment with osimertinib.

Several studies of repeat biopsy of tumour tissue in patients who progress after treatment with osimertinib have shown emergence of KRAS mutations,6 BRAF mutations,7 MET amplification,6, 8 HER2 (also known as ERBB2) amplification,6, 8 and the EGFR Cys797Ser mutation.9 Next-generation sequencing showed several mutations of unknown relevance in 12 patients resistant to another third-generation EGFR-TKI, rociletinib.10 Analysis of plasma samples taken from patients in AURA study11 indicated that patients who had detectable EGFR-activating mutations and acquired Thr790Met mutation in plasma might respond to osimertinib.11 Additionally, analysis of plasma DNA from patients with disease progression taking osimertinib showed various combinations of EGFR-activating mutations and the resistance mutations Thr790Met and Cys797Ser.9 The available information on osimertinib resistance needs to be investigated in genomic studies in tissue and plasma, and clinical outcomes analysed in a large cohort of patients. We assessed clinical outcomes in patients with non-small-cell lung cancer and acquired Thr790Met mutation who were participating in the AURA study. We also tested samples for pathological changes by cytology, and genomic alterations, acquisition of the EGFR Thr790Met and Cys797Ser variants, molecular alterations, and their associations with patients' survival.

Research in context

Evidence before this study

Osimertinib is a third-generation, CNS-active epidermal growth factor receptor (EGFR) tyrosine-kinase inhibitor (TKI) that selectively inhibits EGFR-activating mutations, such as deletion in exon 19 or Leu858Arg, and resistance mutations, such as Thr790Met. Osimertinib is approved for the treatment of patients with non-small-cell lung cancer with the EGFR Thr790Met resistant mutation and previous resistance to EGFR TKIs. Progression-free survival with second-line osimertinib is superior to that achieved with gefitinib or erlotinib as first-line treatment for EGFR mutant non-small-cell lung cancer. Mechanisms underlying resistance to osimertinib and clinical outcomes after treatment with this drug are being intensively investigated. We searched PubMed for studies published in English between Jan 1, 2013, and July 31, 2017, using the term “osimertinib” OR “AZD9291” AND “non-small cell lung cancer” OR “NSCLC” AND “EGFR mutations” AND “T790M” AND “resistance”. We identified reports involving 31 patients in three case studies and five case series. None had accrued enough patients to assess distributions of specific resistance mechanisms or outcomes. We searched PubMed again with the term “osimertinib” OR “AZD9291” AND “non-small cell lung cancer” OR “NSCLC” AND “EGFR mutations” AND “liquid biopsy” OR “circulating tumor DNA”, but found no studies with serial measurements (before treatment or after disease progression) of circulating tumour DNA (ctDNA) in plasma to monitor osimertinib treatment outcomes.

Added value of this study

This retrospective analysis used tumour tissue and plasma samples taken from 53 patients from one centre in the AURA study who had data on EGFR-activating mutations and acquired Thr790Met mutations before and after osimertinib treatment. We investigated mechanisms of resistance, time to development of resistance, and clinical outcomes after development of osimertinib resistance, in enough patients to assess outcomes, and revealed genomic and clinical data not captured in the AURA study.

Implications of all the available evidence

We were able to assess mechanisms underlying resistance to osimertinib in patients with EGFR-mutant non-small-cell lung cancer who had acquired the Thr790Met mutation, and found that heterogeneous resistance mechanisms develop, even across different samples from individual patients. Testing of ctDNA for EGFR-activating mutations and the Thr790Met mutation has been widely used to identify patients who might benefit from treatment with EGFR TKIs. Our findings suggest that this approach could be used to follow up patients treated with osimertinib. We found that loss of Thr790Met but maintenance of EGFR-activating mutations in plasma correlated with shortened progression-free survival, which had not previously been reported. These patients should be assessed for the potential benefit of early intervention in future clinical trials. Testing for the presence of ctDNA after disease progression should be done with caution, because we found heterogeneity in resistance mechanisms and discordance in results between tumour tissue and plasma even within individual patients. Of ten patients tested for MET amplification, five were positive. Testing osimertinib combined with a c-MET inhibitor as preventive rather than salvage therapy in future clinical trials might be useful. We detected many known resistance mechanisms, such as the EGFR Cys797Ser mutation, MET amplification, and BRAF mutations, in tumour tissue. These data support serial testing of plasma for ctDNA in a broader gene panel in future clinical trials as a potential method to identify patients who will develop resistance, particularly because the discordant resistant mechanisms we noted in plasma and tumour samples suggest that repeat biopsy is necessary after disease progression. Combination of osimertinib plus another targeted therapy might be useful for these patients, but the anticancer efficacy of this approach needs to be shown in clinical trials. Since all the patients in this study had received multiple lines of treatment before osimertinib, our results need to be compared with resistance patterns after first-line osimertinib treatment; such information is being collected in the AURA and FLAURA studies. We hope that future studies will provide insight into the optimum time to use osimertinib, alone or in combination therapy, to prolong overall survival of patients with EGFR mutant non-small-cell cancer.

Section snippets

Patients and treatment

AURA was a phase 1 dose-escalation, phase 2 expansion, and extension trial for patients with EGFR-mutant lung adenocarcinoma. The data for response, duration of response, and other anticancer efficacy have been reported.2, 3 This analysis included patients accrued to AURA from the National Taiwan University Hospital who had shown resistance to a previous EGFR TKI and had EGFR-activating mutations and detectable acquired Thr790Met mutation in tumour tissue (analysed centrally or locally) or

Results

Of 71 patients enrolled in AURA, 53 with previous resistance to one or more EGFR TKIs and EGFR-activating mutations and detectable Thr790Met mutation in tumour or plasma were included in this study (figure 1). Of those excluded, 12 patients had not previously been treated with an EGFR TKI and six were negative for the Thr790Met mutation (figure 1). The first patient started treatment with osimertinib on July 31, 2013. The data cutoff for this analysis was Dec 31, 2016. The clinicopathological

Discussion

In our population of patients with non-small-cell lung cancer, no shedding to plasma of EGFR-activating mutations seemed to be a good predictive factor for overall survival, and possibly for progression-free survival while taking osimertinib. Before receiving osimertinib, most patients had been heavily treated with other anticancer drugs, although the duration of previous treatment had little effect on the duration of osimertinib treatment or the pathological or genomic disease evolution.

References (23)

  • TS Mok et al.

    Osimertinib or platinum-pemetrexed in EGFR T790M-positive lung cancer

    N Engl J Med

    (2017)
  • Cited by (126)

    • 20 years since the approval of first EGFR-TKI, gefitinib: Insight and foresight

      2023, Biochimica et Biophysica Acta - Reviews on Cancer
    View all citing articles on Scopus
    View full text