Osimertinib is an oral, potent, CNS active, irreversible EGFR-TKI approved to treat pts with T790M-positive NSCLC. Non-invasive methods to confirm presence of T790M are needed to identify pts who could benefit.
AURA17 (NCT02442349) is a Phase II, single arm study investigating the safety and efficacy of osimertinib 80 mg once daily in an Asia-Pacific pt population with T790M positive advanced NSCLC, who had disease progression following EGFR-TKI therapy. Tumour tissue T790M status was centrally confirmed by cobas® EGFR Mutation Test (Roche Molecular Systems). Where possible, matched plasma ctDNA samples collected at screening were analysed for EGFR mutations using 3 tests: cobas® EGFR Mutation Test v2.0 (cobas plasma), AmoyDx SuperARMS EGFR T790M Mutation Detection Kit (SuperARMS) and droplet digital PCR (ddPCR; in-house research assay).
Table summarises concordance data.Table:
1331P Sensitivity and specificity of plasma tests using cobas tissue test as the reference
|% (95% CI)||cobas plasma (n = 240)||SuperARMS (n = 249)||ddPCR (n = 249)|
|T790M||PPA||42 (34, 50)||49 (41, 57)||56 (48, 64)|
|NPA||83 (72, 91)||78 (67, 86)||73 (62, 83)|
|L858R||PPA||65 (54, 75)||NA*||62 (51, 72)|
|NPA||100 (98, 100)||NA*||99 (96, 100)|
|Exon 19 deletions||PPA||86 (80, 92)||NA*||66 (58, 74)|
|NPA||97 (91, 99)||NA*||98 (93, 100)|
NPA, negative percent agreement (specificity); PPA, positive percent agreement (sensitivity); *SuperARMS used solely for detection of T790M Number of patients tested with cobas tissue test: 277 In the evaluable for response (EFR; 4 March 2016 data cut-off) set, pts with T790M-positive status by both tumour and plasma analysis had confirmed objective response rates (ORR) with osimertinib (RECIST 1.1 by blinded independent central review) of 56% (95% CI 43, 69; 36/64 pts) using cobas plasma, 64% (52, 74; 49/77 pts) using SuperARMS, and 56% (45, 67; 49/87 pts) using ddPCR. ORR in the overall EFR tumour T790M-positive population was 60% (52, 68; 100/166 pts).
Using cobas tissue test as the reference, sensitivity for plasma T790M detection slightly increased with superARMS and ddPCR compared to cobas plasma test. Conversely, specificity slightly decreased. In pts with tumour T790M positive status, ORR with osimertinib was consistent across plasma tests, and with the overall tumour T790M-positive population. Biopsy is recommended for pts with a plasma T790M-negative test result, where feasible.
Clinical trial identification
Legal entity responsible for the study
C. Zhou: Lecture honorarium: Eli Lily, AstraZeneca, Roche, Pfizer, Sanofi, Boehringer Iingelheim, Henrui Advisory Board: Roche, Boehringer Ingelheim, AstraZeneca. Y. Cheng: No financial interest in products or processes involved in our research. X. Ye, Y. Sun, X. Huang: Employee of AstraZeneca. S. Patel: Employee of, and shareholder in, AstraZeneca. Y-L. Wu: Speaker fees from AstraZeneca, Roche, Eli Lilly, Sanofi, Pfizer. All other authors have declared no conflicts of interest.