- Tony S.K. Mok (Shatin, HK)
- Noemi Reguart (Barcelona, ES)
LBA50 - Mechanisms of acquired resistance to first-line osimertinib: preliminary data from the phase III FLAURA study
- Suresh S. Ramalingam (Atlanta, US)
In the Phase III FLAURA study (NCT02296125), osimertinib showed superior efficacy compared with standard of care (SoC) EGFR-TKIs in patients (pts) with previously untreated EGFRm advanced NSCLC. Here, we report preliminary data on mechanisms of acquired resistance to osimertinib in pts who progressed during the FLAURA study.
Pts with previously untreated EGFRm (tissue, ex19del/L858R) advanced NSCLC (N=556) were randomised 1:1 to osimertinib 80 mg once daily (QD; n=279) or SoC EGFR‑TKI (n=277, gefitinib 250 mg QD or erlotinib 150 mg QD). Paired plasma samples were collected at baseline and following RECIST progression and/or treatment discontinuation up to March 2018. Plasma samples were analysed using next generation sequencing (NGS, Guardant Health; Guardant360 73 gene panel or Omni 500 gene panel).
In the osimertinib and SoC EGFR-TKI arms, respectively, 113/279 (41%) and 159/277 (57%) pts had experienced a progression event and/or discontinued treatment and had paired plasma samples analysed by NGS. Only pts with detectable plasma EGFRm (ex19del/L858R) at baseline were evaluable for this analysis: 91/113 (81%; osimertinib) and 129/159 (81%; SoC). In the osimertinib arm, there was no evidence of acquired EGFR T790M and the most common acquired resistance mechanism detected was MET amplification (amp; 14/91; 15%), followed by EGFR C797S mutation (6/91; 7%); other mechanisms included HER2 amp, PIK3CA and RAS mutations (2–7%). In the SoC arm, the most common resistance mechanisms were T790M mutation (60/129; 47%), MET amp (5/129; 4%) and HER2 amp (3/129; 2%).
In this paired sample preliminary analysis of a subpopulation of pts (with detectable baseline plasma EGFRm) who had experienced disease progression and/or discontinued treatment, heterogeneous resistance mechanisms were detected with first-line osimertinib, with MET amplification and EGFR C797S mutation being the most commonly observed. In line with previous analyses, T790M was acquired in approximately 50% of SoC-treated pts, and none of the osimertinib‑treated pts; no unexpected resistance mechanisms were observed in osimertinib-treated pts. Exploration into novel acquired mutations is ongoing.
Clinical trial identification
We thank Ellen Maxwell, PhD, from iMed Comms, who provided medical writing support funded by AstraZeneca in accordance with Good Publications Practice (GPP3) guidelines ().
LBA51 - Analysis of resistance mechanisms to osimertinib in patients with EGFR T790M advanced NSCLC from the AURA3 study
- Vassiliki A. Papadimitrakopoulou (Houston, US)
In the Phase III AURA3 trial (NCT02151981), osimertinib had superior efficacy compared with platinum-based doublet chemotherapy (CT) in patients (pts) with T790M-positive advanced NSCLC, whose disease progressed on or after first-line epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (TKI) therapy. Here we report the ctDNA genomic profile of pts with T790M-positive advanced NSCLC, whose disease progressed on osimertinib treatment during the AURA3 trial.
Pts with EGFR T790M advanced NSCLC, whose disease had progressed on first-line EGFR-TKI therapy, were randomized 2:1 to osimertinib (80 mg once daily) or platinum-based doublet CT. Paired plasma samples were collected at baseline and following disease progression and/or treatment discontinuation. Plasma samples were analyzed by next generation sequencing (NGS; Guardant Health, Guardant360, 73 gene panel).
Among 279 pts randomized to the osimertinib treatment arm, paired plasma samples were available from 83 (30%) pts who had progressed and/or discontinued treatment (PD/DC). 73/83 (88%) pts had baseline detectable ctDNA EGFR mutations (L858R, exon 19 deletion or T790M) and were evaluable for this analysis. Among these 73 pts, 36 (49%) had no detectable T790M at PD/DC, and 11 (15%) acquired EGFR secondary mutation in C797 (C797S n=10; C797G n=1). Amplification of MET, HER2, and PIK3CA were detected in 14 (19%), 4 (5%), and 3 (4%) samples, respectively. Other mechanisms of acquired resistance included mutations in BRAF (V600E, n=3; 4%), KRAS (n=1; 1%) and PIK3CA (E545K, n=1; 1%), and oncogenic fusion mutations in FGFR3, RET and NTRK (n=3; 4%).
In this preliminary analysis from AURA3 of paired plasma samples from pts with detectable baseline plasma EGFR mutations and at PD/DC on osimertinib treatment, a diverse mixture of resistance mechanisms were detected, with MET amplification and EGFR C797S most common. No unexpected resistance mechanisms were observed in these second-line osimertinib-treated pts. Understanding resistance mechanisms in the first and second-line settings will help define appropriate combination therapies.
Clinical trial identification
We thank Natalie Griffiths, PhD, from iMed Comms, who provided medical writing support funded by AstraZeneca in accordance with Good Publications Practice (GPP3) guidelines (ismpp.org/gpp3).
Invited Discussant LBA50 and LBA51
- Charles M. Rudin (New York, US)
LBA52 - Results of the GEOMETRY mono-1 phase II study for evaluation of the MET inhibitor capmatinib (INC280) in patients (pts) with METΔex14 mutated advanced non-small cell lung cancer (NSCLC)
- Juergen Wolf (Cologne, DE)
MET mutations leading to exon 14 deletion (METΔex14) occur in 3-4% of NSCLCs. Capmatinib is a highly potent and selective MET inhibitor. GEOMETRY mono-1 is a phase II, multi-cohort, multicenter study (NCT02414139) evaluating capmatinib in pts with METΔex14 mutated or MET amplified advanced NSCLC. Here we present data from pts with METΔex14 mutation who received either 1–2 prior lines of therapy (Cohort 4) or were treatment-naive (Cohort 5b).
Eligible pts were ≥18 years of age, ECOG PS 0–1 and had ALK and EGFR wt, stage IIIB/IV NSCLC (any histology). Pts with METΔex14 mutation (centrally confirmed) were assigned to Cohorts 4 and 5b regardless of MET amplification status/gene copy number and received capmatinib tablets 400 mg twice daily. The primary endpoint was overall response rate (ORR) by BIRC per RECIST v1.1. The key secondary endpoint was duration of response (DOR) by BIRC.
As of 9-Aug-2018, 94 pts with METΔex14 mutated advanced NSCLC had ≥18 weeks of follow-up (or discontinued earlier) and were included in this analysis (Cohort 4: 69 of 69 pts in 2nd or 3rd line; Cohort 5b: 25 of 28 treatment-naive pts). Treatment was ongoing for 20.3% and 44.0% of pts in Cohorts 4 and 5b, respectively. ORR by BIRC assessment was 39.1% (95% CI, 27.6-51.6) in cohort 4 and 72.0% (95% CI, 50.6-87.9) in cohort 5b. All responses were confirmed. DOR data is not mature with a median duration of follow-up of 5.6 months. Preliminary activity in patients with brain metastases was also observed. The most common AEs (≥ 20% all grades) across all cohorts (1-6, n=302), regardless of causality, were peripheral edema (49.0%), nausea (43.4%), vomiting (28.5%), blood creatinine increased (24.5%), dyspnea (24.2%), decreased appetite (21.2%) and fatigue (20.9%). Majority of these AEs was grade 1/2.
Capmatinib has demonstrated a clinically meaningful response rate and a manageable toxicity profile in pts with METΔex14 mutated NSCLC, particularly in treatment naive pts where the ORR by BIRC is 72%. Differential benefit from 1L to pretreated pts seem to highlight the need for earlier diagnostic testing and prompt treatment of this challenging patient population.
Clinical trial identification
EudraCT: 2014-003850-15; NCT02414139. Release date 13-Feb-2018.
Invited Discussant LBA 52
- James Chih-Hsin Yang (Taipei, TW)
1377O - Phase 2 study of tepotinib + gefitinib (TEP+GEF) in MET-positive (MET+)/epidermal growth factor receptor (EGFR)-mutant (MT) non-small cell lung cancer (NSCLC)
- Yi-Long Wu (Guangzhou, CN)
NSCLC can acquire resistance to EGFR tyrosine kinase inhibitors (EGFR TKIs) via MET activation; dual MET/EGFR inhibition may have potential in EGFR TKI-resistant NSCLC. TEP is a potent, selective MET TKI. We report randomized phase 2 data from a phase 1b/2 signal detection trial of TEP+GEF vs chemotherapy (pemetrexed + cisplatin/carboplatin) in patients (pts) with MET+/EGFR+T790M- NSCLC (NCT01982955).
Asian pts with advanced MET + (IHC2+, IHC3+, gene amplification) NSCLC, acquired resistance to 1st-line EGFR TKI and ECOG performance status 0–1 were eligible. Tumors had an EGFR-activating mutation (T790M-). Pts received TEP+GEF 500/250mg once-daily. Primary endpoint: progression-free survival (PFS by investigator). Secondary endpoints: safety, antitumor activity, pharmacokinetics.
Due to low recruitment, enrolment was halted after 55 pts were randomized to TEP+GEF (n = 31) or chemotherapy (n = 24): male n = 23, median age 60.4 (range 42–82) years. There was a numeric trend towards TEP+GEF on PFS in the intent-to-treat analysis set (hazard ratio [HR]: 0.71 [0.36, 1.39]), driven by the IHC3 + (HR: 0.35 [0.17, 0.74]) and gene-amplified (HR: 0.17 [0.05, 0.57]) pts (Table) confirming these as predictive biomarkers as indicated by phase 1b data. All pts had treatment-related (TR) treatment-emergent adverse events (TEAEs). In the TEP+GEF vs chemotherapy arms, respectively, 9.7 vs 4.3% had TEAEs leading to permanent discontinuation, 3.2 vs 0% had TEAEs leading to death (none were TRTEAEs), 16.1 vs 30.4% had serious TRTEAEs, 51.6 vs 52.2% had Grade ≥3 TRTEAEs, 12.9 vs 8.7% had a TRTEAE of special interest (lipase/amylase increase ≥3).
TEP+GEF shows promising antitumor activity in pts with MET protein overexpression (IHC3+) and gene amplification EGFR-MT NSCLC and was generally well-tolerated. This positive signal warrants further exploration in this pt population.
Clinical trial identification
Legal entity responsible for the study
Medical writing assistance (funded by Merck KGaA, Darmstadt, Germany) was provided by Lisa Jolly PhD of Bioscript Science (Macclesfield, UK).
S. Lu: Research support: AstraZeneca; Speaker fees: AstraZeneca, Eli Lilly, Roche, Pfizer; Advisor, consultant role: AstraZeneca, Hutchison MediPharma, Simcere, BMS, Roche, Pfizer. R. Bruns, A. Johne, J. Scheele: Employee: Merck KGaA. Y-L. Wu: Speaker fees: AstraZeneca, Eli Lilly, Pfizer, Roche, Sanofi. All other authors have declared no conflicts of interest. MET+: Met overexpression by immunohistochemistry (IHC2+ or 3+) and/or MET gene amplification and/or increased gene copy number (GCN ≥5 or MET/CEP7 ratio ≥2) by in-situ hybridisation
Median PFS [90% CI], months Objective response rate, n (%) [90% CI] Tepotinib + gefitinib (n = 31) 4.86 [3.88, 6.87] 14 (45.2) [29.7, 61.3] Pemetrexed + cisplatin/carboplatin (n = 24) 4.37 [4.17, 6.80] 8 (33.3) [17.8, 52.1] Stratified HR [90% CI] or odds ratio (OR) adjusted by randomization strata [90% CI] HR: 0.71 [0.36, 1.39] OR: 1.99 [0.56, 6.87] Tepotinib + gefitinib (n = 19) 8.31 [4.11, 21.16] 13 (68.4) [47.0, 85.3] Pemetrexed + cisplatin/carboplatin (n = 15) 4.37 [4.11, 6.80] 5 (33.3) [14.2, 57.7] Unstratified HR [90% CI] or OR [90% CI] 0.35 [0.17, 0.74] 4.33 [1.03, 18.33] Tepotinib + gefitinib (n = 12) 21.16 [8.31, 21.16] 8 (66.7) [39.1, 87.7] Pemetrexed + cisplatin/carboplatin (n = 7) 4.21 [1.35, 6.97] 3 (42.9) [12.9, 77.5] Unstratified HR [90% CI] or OR [90% CI] 0.17 [0.05, 0.57] 2.67 [0.37, 19.56]
MET+: Met overexpression by immunohistochemistry (IHC2+ or 3+) and/or MET gene amplification and/or increased gene copy number (GCN ≥5 or MET/CEP7 ratio ≥2) by in-situ hybridisation
Invited Discussant 1377O
- Egbert F. Smit (Amsterdam, NL)