MATERIALS AND METHODS: Contrast enhanced computed tomography (CT) images of 194 multi-racial NSCLC patients (79 EGFR mutants and 115 wildtypes) were collected from three different countries using 5 manufacturers' scanners with a variety of scanning parameters. Ninety-nine cases obtained from the University of Malaya Medical Centre (UMMC) in Malaysia were used for training and validation procedures. Forty-one cases collected from the Kyushu University Hospital (KUH) in Japan and fifty-four cases obtained from The Cancer Imaging Archive (TCIA) in America were used for a test procedure. Radiomic features were obtained from BN maps, which represent topologically invariant heterogeneous characteristics of lung cancer on CT images, by applying histogram- and texture-based feature computations. A BN-based signature was determined using support vector machine (SVM) models with the best combination of features that maximized a robustness index (RI) which defined a higher total area under receiver operating characteristics curves (AUCs) and lower difference of AUCs between the training and the validation. The SVM model was built using the signature and optimized in a five-fold cross validation. The BN-based model was compared to conventional original image (OI)- and wavelet-decomposition (WD)-based models with respect to the RI between the validation and the test.

RESULTS: The BN-based model showed a higher RI of 1.51 compared with the models based on the OI (RI: 1.33) and the WD (RI: 1.29).

EXPERIMENTAL DESIGN: Tumor tissue EGFRm status was determined at screening using the central cobas tissue test or a local tissue test. Baseline circulating tumor (ct)DNA EGFRm status was retrospectively determined with the central cobas plasma test.

RESULTS: Of 994 patients screened, 556 were randomized (289 and 267 with central and local EGFR test results, respectively) and 438 failed screening. Of those randomized from local EGFR test results, 217 patients had available central test results; 211/217 (97%) were retrospectively confirmed EGFRm positive by central cobas tissue test. Using reference central cobas tissue test results, positive percent agreements with cobas plasma test results for Ex19del and L858R detection were 79% [95% confidence interval (CI), 74-84] and 68% (95% CI, 61-75), respectively. Progression-free survival (PFS) superiority with osimertinib over comparator EGFR-TKI remained consistent irrespective of randomization route (central/local EGFRm-positive tissue test). In both treatment arms, PFS was prolonged in plasma ctDNA EGFRm-negative (23.5 and 15.0 months) versus -positive patients (15.2 and 9.7 months).

METHODS: Data from 585 eligible patients who received palliative radiotherapy between January 2012 and December 2014 were analysed. Median overall survival was calculated from the commencement of first fraction of the last course of radiotherapy to date of death or when censored. 30-DM was calculated as the proportion of patients who died within 30 days from treatment start date. Kaplan-Meier survival analysis was used to estimate survival. Chi-square test and logistic regression was used to assess the impact of potential prognostic factors on median survival and 30-DM.

RESULTS: The most common diagnoses were lung and breast cancers and most common irradiated sites were bone and brain. Median survival and 30-DM were 97 days and 22.7% respectively. Primary cancer, age, treatment course, performance status, systemic treatment post radiotherapy and intended radiotherapy treatment completed had an impact on median survival whereas mainly the latter three factors had an impact on 30-DM.