Normal epithelial cells often exert anti-tumour effects against nearby oncogenic cells. In the Drosophila imaginal epithelium, clones of oncogenic cells with loss-of-function mutations in the apico-basal polarity genes scribble or discs large are actively eliminated by cell competition when surrounded by wild-type cells. Although c-Jun N-terminal kinase (JNK) signalling plays a crucial role in this cell elimination, the initial event, which occurs at the interface between normal cells and polarity-deficient cells, has not previously been identified. Here, through a genetic screen in Drosophila, we identify the ligand Sas and the receptor-type tyrosine phosphatase PTP10D as the cell-surface ligand-receptor system that drives tumour-suppressive cell competition. At the interface between the wild-type 'winner' and the polarity-deficient 'loser' clones, winner cells relocalize Sas to the lateral cell surface, whereas loser cells relocalize PTP10D there. This leads to the trans-activation of Sas-PTP10D signalling in loser cells, which restrains EGFR signalling and thereby enables elevated JNK signalling in loser cells, triggering cell elimination. In the absence of Sas-PTP10D, elevated EGFR signalling in loser cells switches the role of JNK from pro-apoptotic to pro-proliferative by inactivating the Hippo pathway, thereby driving the overgrowth of polarity-deficient cells. These findings uncover the mechanism by which normal epithelial cells recognize oncogenic polarity-deficient neighbours to drive cell competition.

BACKGROUND:The secondary epidermal growth factor receptor (EGFR) mutation Thr790Met (T790M) accounts for approximately half of acquired resistances to EGFR-tyrosine kinase inhibitor (TKI). Recent reports have demonstrated that the emergence of T790M predicts a favorable prognosis and indolent progression. However, rebiopsy to confirm T790M status can be challenging due to limited tissue availability and procedural feasibility, and little is known regarding the differences among patients with or without T790M mutation. METHODS: The study investigated 78 EGFR-mutant patients who had undergone rebiopsy after TKI failure. The peptide nucleic acid-locked nucleic acid polymerase chain reaction clamp method was used in EGFR mutational analyses. Various patient characteristics and postprogression survivals (PPSs) after initial TKI failure were retrospectively compared in patients with and without T790M. RESULTS: The T790M mutation was identified in 4 (17%) of 24 central nervous system lesions, and in 22 (41%) of 54 other lesions (P 5.0417). No other characteristics had a statistical association with T790M prevalence. Median PPS was 31.4 months in 26 patients with T790M, and 11.4 months in 52 patients without T790M (P 5.0017). In the multivariate analysis, statistically significant factors for longer PPS included T790M-positive, good performance status, and no carcinomatous meningitis. CONCLUSIONS: The emergence of T790M in central nervous system lesions was rare, compared with other lesions. Patients with T790M after TKI failure appear to have better prognoses than those without T790M. TKI rechallenge or continuous administration beyond progression may be effective after initial TKI failure. Cancer 2013;119:4325-32.

PurposeTo conduct a retrospective multicenter trial to determine the significance of metastatic site as a predictor of nivolumab efficacy in patients with advanced non-small cell lung cancer.MethodsThis study was conducted across three medical centers in Japan. We retrospectively reviewed all patients who commenced nivolumab treatment at these centers between December 17, 2015 and July 31, 2016. Clinical data were collected, including age, sex, smoking status, Eastern Cooperative Oncology Group performance status, and metastatic site (lymph nodes, liver, brain, bone, lungs [intrapulmonary metastasis], and malignant pleural effusion) at the time of commencing nivolumab treatment. Patients were followed-up until March 31, 2017.ResultsTwo hundred and one patients were enrolled. The median age at the time of commencing nivolumab treatment was 68 (range, 27–87) years. One hundred and thirty-five patients were male, 157 patients had a history of smoking, 153 patients had a performance status of 0–1, and 42 patients had squamous cell carcinoma. The median progression-free survival of all patients was 2.5 months. In the univariate analysis, a performance status of ≥2 (hazard ratio [HR]: 1.89, 95.0% confidence interval [CI]: 1.33–2.69; p < 0.001) and liver (HR: 2.09, 95.0% CI: 1.35–3.25; p < 0.001) and lung (HR: 1.57, 95.0% CI: 1.14–2.16; p < 0.01) metastases correlated with a significantly shorter progression-free survival in nivolumab-treated patients. In the multivariate analysis, a performance status of ≥2 (HR: 1.54, 95.0% CI: 1.05–2.25; p < 0.05) and liver (HR: 1.90, 95.0% CI: 1.21–2.98; p < 0.01) and lung (HR: 1.41, 95.0% CI: 1.00–1.99; p < 0.05) metastases were independently correlated with a significantly shorter progression-free survival in nivolumab-treated patients.ConclusionLiver and lung metastases and a poor performance status are independent predictors of nivolumab efficacy in patients with advanced non-small cell lung cancer.

Complex EGFR mutations are not rare. Gefitinib has different efficacy according to the type of complex EGFR mutations. Patients with Del-19 and L858R mutations may benefit more from gefitinib than other types of complex mutations.

Although EGFR-TKIs seem to be generally less effective in EGFR-mutated SQC than in EGFR-mutated adenocarcinoma, some EGFR-mutated SQC patients can obtain clinical benefit from EGFR-TKIs. To better identify these patients, not only EGFR mutation status, but also clinical factors and pathologic findings should be taken into consideration.