T790M Plasma Testing Methodology Comparison and Clinical Validation

Objective: The primary objective of this study is to evaluate concordance of T790M mutation plasma testing between the Cobas test and each of other platforms: Super-ARMS, digital PCR or NGS. And to assess the efficacy of AZD9291 monotherapy by assessment of PFS in adult patients with advanced or metastatic NSCLC, who have received prior EGFR-tyrosine kinase inhibitor (TKI) therapy and are T790M mutation positive detected by any one of the four plasma testing platforms: Cobas/Super-ARMS/ digital PCR/NGS.

Study number of patients planned: Approximately 250 patients will be recruited in China.

Study Design: This is an open-label, multi-center testing and treatment study.

Target patient population: 250 locally advanced or metastatic EGFR mutation positive NSCLC patients with progression on a previous EGFR-TKI will be recruited.

Investigational product (IP), dosage, and mode of administration: AZD9291 is an oral, potent, selective, irreversible inhibitor of both EGFR-TKI sensitizing and resistance mutations in NSCLC with a significant selectivity margin over wild-type EGFR. AZD9291 will be administered orally as one 80 mg tablet once a day. All AEs/SAEs would be reported in ASTRIS main study and would not be reported repeatedly in current study.

Duration of IP administration: Patients may continue to receive AZD9291 as long as they continue to show clinical benefit, as judged by the investigator, and in the absence of discontinuation criteria. The study will be closed in a maximum period of 18 months after the last patient is enrolled. Contingencies will be made to ensure continued drug supply for patients who are still deriving benefit from AZD9291 at that time.

Study measures: Data collected will include patient demographics, smoking history, information needed to determine patient eligibility (including medical history, past and current disease characteristics, and tumor EGFR mutations status, T790M and sensitizing mutations status results and type of test performed), AZD9291 exposure, investigator-reported efficacy (including tumor response and disease progression), overall survival (OS).

Statistical methods: The concordance of T790M resistance mutation testing between the Cobas test and each of other platforms will be calculated. The sensitivity, specificity, PPV and NPV of each testing platform (Super-ARMS, digital PCR, and NGS) will be calculated with the Cobas test as the reference. The Kappa coefficient will be calculated to measure the agreement of T790M mutation testing between the Cobas test and each of other platforms. Descriptive statistics will be provided for all variables, as appropriate. Continuous variables will be summarized by the number of observations, mean, standard deviation, median, interquartile range (Q1, Q3), minimum, and maximum. Categorical variables will be summarized by frequency counts and percentages for each category. The 95% confidence interval (CI) will be calculated as appropriate. PFS and OS, respectively, will be summarized using Kaplan-Meier estimates of the median time to event (progression and death) and quartiles together with their 95% confidence intervals.The chi-square test will be used to compare the sensitivity, specificity, and concordance between any of the two platforms using Cobas as reference testing in an exploratory manner.