Virtual Library

Abstract
Thymic epithelial tumors represent a wide range of anatomical, clinical, histological, and molecular malignant entities, which may be aggressive and difficult to treat. The histopathological classification distinguishes thymomas from thymic carcinomas. Our current understanding of the carcinogenesis of these tumors remains limited; only few well-characterized preclinical models - mostly consisting of thymic carcinoma cell lines -, have been established. Several biomarkers studies have been reported; while tumor stage, completion of surgical resection, and - to a lesser extent – histology have been shown to significantly predict the outcome of patients, the expression of EGFR, KIT, or IGF1-R by immunohistochemistry failed to consistently demonstrate a prognostic value on the survival of patients. More recently, expression signatures have been reported, both for thymomas and thymic carcinomas, to correlate with metastasis-free survival; these results remain to be validated in separate cohorts, while prospective study to understand the clinical significance of these results will be required. The management of thymic epithelial tumors is a paradigm of cooperation between clinicians, surgeons, and pathologists from establishing the diagnosis to organizing the multimodal therapeutic strategy. Chemotherapy may be used in two clinical scenarios in thymic epithelial tumors: 1) chemotherapy may constitute primary part of the multimodal curative-intent treatment of locally-advanced tumors, and is subsequently combined with surgery or radiotherapy; the main objective is to achieve long-term survival with no evidence of tumor recurrence; 2) chemotherapy may be delivered as the sole treatment modality in unresectable, advanced, metastatic, or recurrent tumors; then a palliative-intent treatment, the aim is to improve tumor-related symptoms through achievement of tumor response, while no prolonged survival is expected. Novel treatment strategies are needed, especially for refractory, recurrent tumors, and thymic carcinomas, which carry a poor prognosis despite multimodal treatment. Potentially druggable targets are emerging, laying the foundations to implement personalized medicine for patients. Given the currently available targeted agents outside of a clinical trial, the signaling pathways that are relevant in the clinical care of patients are the KIT and the Vascular Endothelial Growth Factor (VEGF)-R (Receptor) pathways. Several tyrosine kinase inhibitors targeting KIT have been developed - including imatinib (Novartis, Basel, Switzerland), sunitinib (Pfizer, New-York, NY), and sorafenib (Bayer, West Haven, CT) - , most of which also potently inhibiting other kinases, including VEGFRs and Platelet-Derived Growth Factor Receptors. Collectively, KIT is overexpressed in 80% of thymic carcinomas, while KIT gene mutations are found only in 9% of cases. Clinically, KIT-mutant thymic carcinomas then represent a small molecular subset of thymic tumors. The clinical relevance of KIT mutations is more limited in thymic carcinoma than in other cancers like gastro-intestinal stromal tumor (GIST), as 1) KIT mutations are far less frequent; 2) KIT expression does not correlate with the presence of KIT mutation; and 3) non-pretreated KIT mutants are not uniformly sensitive to imatinib, based on the clinical and/or the preclinical evidence in thymic carcinoma and/or other KIT-mutant tumors. These findings may explain why the 2 reported phase II trials with imatinib, where patients were not selected, or selected based upon histologic type (B3 thymomas and thymic carcinomas) or KIT staining by immunohistochemistry, and not upon KIT genotyping, were negative. Multi-kinase inhibitors may also be of interest to target neoangiogenesis. The most potent pro-angiogenic molecules are those of the VEGF/VEGFR signaling pathway. VEGF-A and VEGFR-1 and -2 are overexpressed in thymomas and thymic carcinomas. Micro-vessel density and VEGF expression levels have been shown to correlate with tumor invasion, aggressive histology and clinical stage. In a phase II trial, bevacizumab was tested in combination with erlotinib in thymomas and thymic carcinomas. No tumor response was observed. Interestingly, despite the large tumor burden of thymic tumors and the frequent abutment to mediastinal vascular structures, no hemorrhagic side effect has been reported with the use of these drugs in these studies. Beyond the inhibition of KIT, sunitinib and sorafenib also inhibit VEGFR-1, VEGFR-2, VEGFR-3 at the nanomolar range. The effect of these drugs, especially in KIT-wild-type thymic carcinoma tumors may then be partially related to an anti-angiogenic effect. Promising new targets in thymoma and thymic carcinoma include IGF-1R and histone-deacetylase. Cixutumumab, an IGF1-R directed monoclonal antibody was recently reported to produce a promising 90%-disease control rate in refractory thymomas. Belinostat, a histone deacetylase inhibitor was evaluated in thymic malignancies in a recently completed phase II trial enrolling 41 patients (25 thymomas and 16 thymic carcinomas). Response and 2-year survival rates were 8% and 77% in thymomas. Given the rarity of the tumor, translation of pre-clinical findings to the clinic may be quick; several strategies have been implemented. A pragmatic approach is the recommendation for KIT genotyping in clinical practice, what represents a model of n-of-one trial approach in the field. Another approach to validate the concept of personalized medicine in thymic malignancies includes the development of open-label multicentric phase II trials, using high throughput genome analysis (CGH array, next generation sequencing) as a therapeutic decision tool, to compare a medical treatment administered according to the identified molecular anomaly of the tumor with a medical treatment administered without considering the genome analysis; thymic tumors have been integrated in some ongoing trials using such methodology. A third approach is to promote the enrollment of patients with refractory thymic tumor in phase I trials, what may lead to identify new molecular pathways of therapeutic interest; mTOR is emerging as a potential target, following tumor responses observed in phase I trials, with recent data from several groups. Along with the large variety of questions relative to the treatment strategy, thymic epithelial tumors represent a model of therapeutic implementation and achievement; in this setting, regional and international collaborative initiatives are mandatory to progress both in the understanding of the biological mechanisms underlying the development of thymic malignancies, and in the identification and validation of new targets with prognostic and predictive value.

Abstract
Malignant mesothelioma is an uncommon malignancy that may involve the pleura, peritoneum or pericardium. Mesothelioma may assume a variety of histological appearances ranging from epithelial to sarcomatoid to mixed. Because of its histologic similarity to a variety of other malignancies and because other malignancies may metastasize to serosal membranes, the diagnosis of mesothelioma may be challenging. A variety of immunohistochemical stains have been developed to aid in the distinction between mesothelioma and other malignancies with which it may be confused. It is currently recommended that at least two positive (mesothelial) and two negative (carcinomatous) markers should be used. The differential diagnosis may differ depending on the particular histologic pattern under consideration and the location of the tumor. This will in turn affect what immunohistochemical panel is chosen. It is also recommended that the markers used should have a specificity and sensitivity of at least 80%. For epithelial pleural mesothelioma, the main differential diagnostic consideration is adenocarcinoma of the lung. For such cases the author uses a panel of four positive markers and two negative markers. Positive markers include calretinin, cytokeratins 5/6, WT-1 and D2-40. Negative markers include carcinoembryonic antigen (CEA) and TTF-1. The author has used this panel since 2004 and has had experience with more than 500 mesotheliomas to date. Epithelial pleural mesotheliomas were positive for calretinin in 98% of cases, and this was the most sensitive mesothelial marker. Cytokeratins 5/6 and D2-40 were each positive in 92% of cases, and WT-1 was positive in 91% (nuclear staining only counted). For the negative markers, pCEA was positive in 10% and TTF-1 in 0.5% (nuclear staining only counted). More recently we have used mCEA and thus far have not identified any positive cases. We have seen some cases where mesothelioma was confused with squamous cell carcinoma. Cytokeratins 5/6 are not useful in this scenario since nearly all squamous cell carcinomas are positive for this marker. Similarly, TTF-1 is not useful since squamous cell carcinomas are expected to be negative. For this differential diagnosis, WT-1 and p63 (or p40) are a good pair of nuclear markers to use. Biphasic pleural mesotheliomas are more poorly differentiated so it is not surprising that one often observes a drop-out of one or more of the positive mesothelial markers in these tumors. We find calretinin positivity in 92% with positive staining for D2-40 in 83%. The corresponding values for WT-1 and cytokeratins 5/6 are 80% and 78%, respectively. CEA was positive in only 4% and none was positive for TTF-1. If one grades the intensity of staining, the epithelial tumors show higher scores on average than the biphasic. Peritoneal mesotheliomas usually involve a different set of diagnostic considerations. Since lung cancer is unlikely to be in the differential, TTF-1 has limited usefulness in this setting. Similarly, many of the tumors in the abdomen with which mesothelioma may be confused are negative for CEA. Therefore our panel for peritoneal tumors includes the same four positive markers noted above, and we substitute B72.3 and BerEP4 as negative markers. Epithelial peritoneal mesotheliomas are positive for calretinin in 98% of cases, identical to pleurals. Positive staining for WT-1 and D240 is seen in 91% of cases, and 89% are positive for cytokeratins 5/6. BerEP4 was focally positive in 9% of peritoneal mesotheliomas but B72.3 was uniformly negative. For biphasic peritoneal mesothelioma, calretinin is positive in 80% and cytokeratins 5/6 positive in 75%. Only 60% were positive for D2-40 and 50% for WT-1. None of the biphasic peritoneal cases was positive for B72.3 or BerEP4. Peritoneal mesotheliomas in women present a special diagnostic problem since many of the malignancies involving the peritoneal cavity in women stain for one or more of the mesothelial specific markers. For this reason, we add estrogen and progesterone receptor immunohistochemical staining to the panel for peritoneal mesothelioma in women. Strong diffuse nuclear staining for these markers would favor a diagnosis of ovarian carcinoma over mesothelioma. Also, histochemical staining for PASD is useful in this setting. We also use a broad spectrum cytokeratin marker in our panel, with cytokeratins positive in 100% of our epithelial or biphasic pleural and peritoneal mesotheliomas. The intensity of staining for these broad spectrum cytokeratins was a little greater for epithelial as compared to biphasic mesotheliomas. Complete absence of staining for cytokeratins in an epithelioid serosal based malignancy should alert the pathologist to the possibility of melanoma, lymphoma or epithelioid hemangioendothelioma and appropriate follow-up immunostains examined to rule out these considerations. Sarcomatoid mesotheliomas account for 16.5% of our cases, and these are the least differentiated form. This is reflected in their immunohistochemical profile, which is typically negative for mesothelial specific markers. For example, in 76 cases of sarcomatoid mesothelioma, we found positivity for calretinin in only 47% and D2-40 in 55%. The other two markers we use in the sarcomatoid panel are vimentin (95%) and broad spectrum cytokeratins (92%). Vimentin is important as an antigenic preservation marker: if vimentin is negative in a sarcomatoid tumor, then one should suspect poor tissue preservation with less concern for a negative keratin stain. Strong diffuse staining for cytokeratins is the most useful marker for sarcomatoid mesotheliomas. However, sarcomatoid carcinomas may be similarly positive, so one should pay close attention to the gross distribution of tumor in these cases. Although diffuse and moderate to strong staining for cytokeratins was seen in more than 70% of our sarcomatoid mesotheliomas, it should be noted that focal staining or less intense staining may also be seen in some cases. REFERENCES: Husain AN, Colby T, Ordonez N, et al. Guidelines for pathologic diagnosis of mesothelioma: 2012 update of the consensus statement from the International Mesothelioma Interest Group. Arch Pathol Lab Med 137: 647, 2013. Klebe S, Brownlee NA, Mahar A, et al. Sarcomatoid mesothelioma: a clinical-pathologic correlation of 326 cases. Mod Pathol 23: 470, 2010. Churg A, Cagle PT, Roggli VL. Tumors of the Serosal Membranes, AFIP Atlas of Tumor Pathology, Series IV, Fascicle 3, American Registry of Pathology: Washington, DC, 2006.

Abstract
Novel therapeutic targets have been revealed in lung cancer through the use of systematic large-scale sequencing approaches. While some of these are already part of routine clinical care, others still await confirmation in clinical trials. I will discuss basics of lung cancer genomics, differences in targets and target candidates in lung adenocarcinoma, squamous cell lung cancer and small cell lung cancer and problems arising during the path of translating basic discoveries into clinical practice.

Abstract
A Practical Approach to the Incidental Pulmonary Nodule David Midthun, M.D. David Milne, M.D. The finding of a pulmonary nodule (or multiple nodules) on an imaging study presents a decision point for the patient and physician. In the absence of a completely sensitive and specific non-invasive test for malignancy, the physician and patient must weigh the options for management. The vast majority of such nodules are benign; however the detection of a nodule may be the first and only point in time of a chance of cure in the patient with lung cancer. Guidelines for nodule evaluation by the American college of Chest physicians (ACCP) and the Fleischner Society may help guide the decision making. Studies of lung cancer screening have shown high rates of nodule detection and that the rate is related to the CT slice thickness (collimation) used. Screening with 10 mm collimation results in detection of one or more nodules in approximately 20-25% of participants, 5 mm collimation increases this to 40-50% of participants, and 1.25 mm collimation raises detection to as high as 60%. A review of the data from 8 CT studies in high risk patients (current or former smokers, age 50 or above) reported that likelihood of malignancy was 0 to 1% for nodules < 5 mm, 6 to 28% for nodules 5 to 10 mm, 33 to 60% for nodules > 11-20 mm, and 64 to 82% for nodules > 20-30 mm. The finding of a nodule on CT should first prompt review of any available old images that might include the nodule for comparison. Review of old images may show that the nodule is growing or, alternatively, establish that it has been stable for 2 or more years. Stability in size over a two-year period has been established as an excellent indicator of benignancy for a solid nodule. If old images are not available, nodules < 8 mm may be observed with follow-up CT at an interval determined by the nodule size. Evidence for nodule growth is a hallmark of malignancy and should lead to a staging PET-CT scan (in those who are candidates for surgery) and consideration of prompt resection. Calcification in a benign pattern is an excellent indicator that a nodule is a granuloma and needs no further pursuit. Eccentric calcification should maintain concern for malignancy. Ground-glass nodules (GGN) are nodules of low density (attenuation) that are generally only visible by CT scan. They deserve special mention as they may represent low-grade adenocarcinomas which behave differently than most malignancies presenting as solid nodules. Malignant GGNs typically exhibit slow growth with doubling times on average over 400 days and, for this reason, the 2-year stability rule for solid nodules doesn’t apply and a longer period of follow-up is needed. PET scanning is not helpful to distinguish malignancy due to the low density of the lesions, and needle biopsy is often nondiagnostic. GGNs may show growth or stay the same size yet develop a solid component in the process of progression. PET scanning uses the injection of the glucose analog 18F-2-fluorodeoxyglucose (FDG) and identifies elevated metabolic activity. Nodule enhancement is an indication that a nodule is more likely malignant than benign, and absence of enhancement is a strong predictor that a nodule is benign. In a multicenter prospective study reported that FDG-PET had an overall sensitivity of 92% and a specificity of 90% for detecting malignant nodules, yet the sensitivity fell to only 80% when nodules of 15 mm or smaller were analyzed. A meta-analysis of pulmonary nodules showed that PET had a sensitivity of 94% and a specificity of 86%. The lower limit of solid nodule size for PET applicability using current techniques is about 8-10 mm. A growing nodule that shows no enhancement on PET should still be considered suspicious for malignancy and prompt needle biopsy or resection. If multiple nodules are present, then evaluation is dictated by the largest nodule. Observation may be appropriate for patients with nodules that are larger than 8-10 mm and have a low likelihood of malignancy based lack of enhancement. Whether or not an indeterminate nodule > 8-10 mm should be biopsied is the subject of considerable debate and practices vary. The two biopsy techniques for assessment of nodules are bronchoscopy and transthoracic needle aspiration (TTNA). Bronchoscopy with fluoroscopy alone has a yield of less than 20% in the setting of malignant nodules less than 2 centimeters and in the range of 40-60% when the nodule is 3-4 cm. Studies of guided bronchoscopy using endobronchial ultrasound and/or electromagnetic guidance have shown marked improvements in diagnostic yield over standard fluoroscopic guidance. Studies using one or more of these techniques have shown yields of 60 to 80% of peripheral nodules of a mean diameter of 2- to 25 mm. Yields remain highest with TTNA; multiple studies report yields of 90% and above for nodules < 2 cm and 95% for nodules > 2 cm. Pneumothorax is the most frequent complication of TTNA. Likelihood of obtaining a specific diagnosis in the setting of a benign lesion is problematic for both bronchoscopy and TTNA. Decision as to the method of biopsy involves lesion size, location, presence of a bronchus leading to the lesion, and comorbidities. Preoperative diagnosis may not be needed for lesions that show growth or are nearing 3cm and are PET avid due to the high likelihood of malignancy and low likelihood a biopsy is going to provide a specific benign diagnosis. An exception would be in countries where there is a high prevalence of tuberculosis where sampling may remain appropriate. Resection is the ultimate management for many lesions that remain indeterminate after imaging evaluation especially in a high risk individual. There are currently too many benign nodules removed surgically. Series of video assisted thoracic surgery (VATS) have reported benign nodules representing as high as 50-86% of nodules resected. Reduction in benign nodule resections may be achieved by observing smaller nodules, by utilizing PET-CT, and by performing biopsy by TTNA or bronchoscopy when information is discordant.

Abstract not provided

Abstract
Malignant pleural mesothelioma (MPM) is an asbestos-related disease with a very poor prognosis because of late-stage diagnosis due to unspecific imaging techniques and non-specific symptoms resembling pleural effusions such as chest pain and dyspnea. Early diagnosis could improve the disease’s outcome by reducing the diagnostic delay. Serum tumor biomarkers for early MPM screening are attractive because of their non-invasive and low-cost nature. Nevertheless, MPM is a heterogeneous disease suggesting an ideal biomarker should be capable of capturing all the MPM subtypes and to distinguish MPM from benign or metastatic pleural conditions. To have a clinical impact, biomarkers should predict the development of the disease in which a positive test could enrich the at-risk population eligible for further screening and reduce the economical burden of wild screening. In order rule in MPM, a biomarker should have a positive predictive value of minimally 10%, and the low MPM prevalence urges the need for a high test specificity. Many candidate blood biomarkers have been studied such as hyaluronic acid, carcinoembryonic antigen, CYFRA 21.1 and cancer antigen 125. Nonetheless, their accuracy is insufficient and based upon small-sized retrospective studies without further validation. Hence, they are of little use in clinical practice. More recently, the cell adhesion proteins soluble mesothelin (SM), megakaryocyte potentiating factor (MPF) and osteopontin (OPN) were investigated. OPN functions in cancer progression, bone matrix formation and immunologic responses. Despite 78% sensitivity and 88% specificity found by Pass et al., subsequent validation studies revealed contradictory accuracy findings with AUCs ranging from 0.64 to 0.89. Furthermore, OPN lacks specificity because it is overexpressed in other tumor types, limiting its use as a diagnostic mesothelioma marker. SM has been found the best available serum marker for MPM. It is derived from a mesothelin gene-encoded precursor protein, which is cleaved into a soluble fraction (MPF) and a membrane-bound fraction (mesothelin) involved in cancerous growth, proliferation and migration and present on the pleural, peritoneal and pericardial mesothelial cells. SM enters the circulation by shedding of the membrane-bound mesothelin or frameshift mutations and is highly expressed in mesothelioma and other malignancies. The first determined serum MPF levels differentiated MPM from healthy controls with 91% sensitivity and 100% specificity. Subsequent studies resulted in diagnostic accuracies with AUCs between 0.79 and 0.88. MPF and SM are highly correlated and have equal diagnostic performances. SM was elevated in MPM patients compared to controls and had discriminative power with 84% sensitivity and 100% specificity. A commercial MesoMark[TM] ELISA assay for SM was developed and evaluated by different groups gaining diagnostic accuracies with AUCs from 0.72 to 0.81. The use of SM as diagnostic marker is still under debate because of large intergroup inconsistencies. An individual patient data meta-analysis was performed, yielding an AUC of 0.77 representing the overall SM diagnostic performance. When SM was used to rule in or rule out diagnosis, the specificities and sensitivities were respectively 95% and 32% and 22% and 95%. SM has low specificity and is only selective for epithelioid and biphasic MPM, hampering its use as a stand-alone marker and possible replacement of the current gold standard of invasive histopathologic diagnosis. Hence a combination of several tumor markers might improve the diagnostic performance. However, combining SM, OPN and MPF did not outperform the accuracy of SM alone and it is necessary to take GFR, BMI and age into account as confounding effects because they influence the biomarker levels. New markers like HMGB1 and fibulin-3 were investigated and were found upregulated in patients sera compared to sera from healthy controls. Although both are promising, a long and cumbersome validation process will need to be executed comparing the accuracy of both biomarkers with serum SM. In the past, several programs have been conducted to screen asbestos-exposed individuals for lung disease with annual chest X-rays. Besides demonstrating the presence of benign asbestos-related diseases, these modalities have not proven to be effective at detecting early malignancies. CT has a superior sensitivity, and its use in screening has been examined in several large-scale studies (Table). Results predominantly illustrate the low prevalence of mesothelioma, and the high background noise (non-calcified nodules) in asbestos-exposed populations. In addition, the standardization of reading both chest X-ray and CT has proven to be difficult, while the cost and radiation doses represent other problematic issues. PET and MRI are currently not applied in screening, and their cost is likely to be prohibitive. Altogether, it is now generally accepted that the use of imaging has not made any impact on the early detection of mesothelioma, and their use is not recommended by the different (inter)national guidelines. In the future, the ‘Holy Grail’ could be sniffed out thanks to the innovative field of exhaled breath analysis. Volatile organic compounds (VOCs) arise from the cells’ metabolism and are released in exhaled breath, making these promising diagnostic MPM markers obtained via high-throughput non-invasive techniques. Electronic nose (eNose) analysis has shown promising results for diagnosing MPM with diagnostic accuracies ranging from 80.8% to 95% in discriminating MPM patients from asbestos-exposed and healthy individuals. However, eNoses work as blackboxes and do not identify VOCs as possible biomarkers. GC-MS analysis identified cyclohexane to discriminate MPM patients from asbestos-exposed and healthy persons. Although this research field opens new perspectives for biomarker discovery, a lot of small-sized studies were performed in order to identify new biomarkers for screening, urging the need for large-scale validation studies. Although different routes of discovering biomarkers for early screening have been paved, the way to a validated and clinically useful screening biomarker panel still has to be constructed. Table: Prospective screening studies using chest CT in asbestos-exposed cohorts to detect asbestos-related malignancies.

Reference	Total (n)	Lung cancer (n)	Mesothelioma (n)
Tiitola et al.	602	5	1 peritoneal
Vierikko et al.	633	5	1 pleural
Fasola et al.	1045	9	-
Mastrangelo et al.	1119	5	-
Roberts et al.	516	6	2 pleural/2 peritoneal
Total	3915	30 (0.7%)	6 (0.2%)

Abstract
Treatment of Cancer Cachexia over the next 5 years. The focus of this session will be on the emerging evidence for therapeutic approaches, outside anti-cancer treatments, that help clinicians care for patients and their families with anorexia cachexia syndrome. An overview of the basic pathophysiology will be presented to enable understanding of the diverse interventions studied as well as those that are currently under investigation. The challenge and importance of defining the syndrome will be discussed. It will also be important to discuss the drivers that are moving the sector towards the concept of categorising patients into pre-cachexia, established cachexia and refractory cachexia. This has implications about when to intervene and how the goals of interventions may change over time. This may help design better matched cohorts for clinical trials, influence design of service models and alert us to the possibility of previous false negative studies. There will be a review of the prevalence of anorexia cachexia in various tumour streams and at different time points over the illness trajectory. The scope and scale of the clinical impact will be discussed in detail including the potential importance of symptom clusters, association with total symptom burden, links to fatigue and the evolving picture of the psycho-social distress that is common for suffers and their families or carers. Significant time will be dedicated to summarising the evidence base for pharmacological, nutritional and exercise approaches. The author will share his own experience in leading a dedicated inter-disciplinary cancer cachexia clinic over the last 9 years in Victoria and why he feels a multi-modality approach should be adopted. Lastly an outline of what might represent best practice for the management of this syndrome over the next five years will be presented for open discussion.

Abstract
Background Standard treatment for stage IIIA or IIIB non-small cell lung cancer (NSCLC) is chemoradiotherapy (CRT). In patients with infiltrative stage III (N2) NSCLC and PS 0-1 being considered for curative-intent treatment, platinum-based chemotherapy and radiotherapy (60-66 Gy) are recommended based on the revised American College of Chest Physicians (ACCP) guidelines for lung cancer. This guideline adopted new criteria for clinical N2 status, discrete and infiltrative N2 status. Infiltrative N2 means no radiolucency between nodes and surrounding organs, and usually considered to be unresectable unless combined resection not performed. Discrete N2 was defined as nodes having surrounding radiolucency on computerized axial tomogram (CAT). Surgery alone is not recommended for both clinical N2 diseases, based on several retrospective studies reported in 1990’s. Some cases would be candidates for upfront surgery followed by adjuvant chemotherapy, but approximately 40% of patients with resected lung cancer cannot tolerate postoperative adjuvant chemotherapy. ACCP guideline does not recommend surgery followed by chemotherapy, if N2 was verified preoperatively, which is reasonable for considering Thus preoperative treatment should be suitable for most clinical N2 NSCLC, if trimodal treatment is taken into consideration. As to a suitable preoperative treatment, induction chemotherapy is generally preferred to CRT, for pulmonary resection following CRT is believed to be more difficult and higher complication rate. However one of the major disadvantages for induction chemotherapy should be incomplete resection, which means losing local control for patients with N2 disease. In this respect induction treatment should include radiotherapy for infiltrative or discrete N2 NSCLC. The significance of surgery in N2 NSCLC is controversial. Randomized controlled trial definitive CRT vs induction CRT followed by surgery, i.e. “Intergroup 0139 trial”, resulted in negative. Overall survival was not improved for surgery group. While in subset analysis lobectomy appeared to improve survival, pneumonectomy did not. This is partly due to a high mortality rate in pneumonectomy group, which is a matter of debate. Some authors insisted that pneumonectomy is feasible following CRT with a low surgical mortality rate. Lessons from “Intergroup 0139 trial) were described in the recent ACCP guidelines as follows; 1) patients preferences and characteristics should be considered; 2) highlight the importance of minimizing harms, and surgery should be undertaken in a center with experience; 3) if there are reasons to be concerned about the ability of radiotherapy to achieve local control, surgery may have a benefit provided R0 resection is likely to be achieved; 4) lobectomy is preferred to pneumonectomy following induction CRT. Dose of irradiation in surgery group is 45Gy, and in the other side of definitive CRT group 61Gy. Considering tumor biology under radiotherapy, the difference of the doses cannot be negligible. Cancer cells tend to response in the late phase of radiation rather than early phase. Thus ideal radiation dose would be 60 to 66Gy, and lobectomy rather than pneumonectomy should be performed in trimodal treatment for stage IIIA N2 NSCLC. In this setting surgery is performed following definitive CRT, and called as “salvage surgery.” There are few reports on the significance and feasibility of salvage surgery for lung cancer, and the indications should be limited. On the other hand, definitive CRT can control local disease at most 40% resulting in approximately 20% of long-term survival. Not a few patients die due to local disease without distant metastasis. Recent randomized study on the feasibility of very high dose radiotherapy, RTOG 0617, showed inferiority of 74Gy to 66Gy, and better local control should be obtained only with additional surgical resection, i.e. salvage surgery. Methods Comprehensive review of the literature will be presented. Experience of salvage surgery at Juntendo Hospital was investigated and presented. Between February of 2008 and July of 2013, 1240 patients with lung cancer underwent surgical resection, which includes 20 salvage surgeries. Salvage surgery was performed in 10 patients following chemotherapy, and 10 CRT. Majority of the mode of surgery was 6 pulmonary lobectomies, including 3 sleeve resection, and 3 pneumonectomies, including 1 sleeve pneumonectomy. The dose of radiation was 60Gy in 8 patients, 45Gy in 1 patient, and 140Gy in 1 patient who underwent proton and heavy particle radiotherapy. All cases underwent irradiation to the primary tumor and hilar and mediastinal region. Surgical outcome was investigated. Results There were no surgical mortality. There were 3 postoperative morbidity including empyema, pneumonia and bronchopleural fistula (BPF). BPF developed following right side pneumonectomy following proton and heavy particle radiation for the primary and hilar region. This patient underwent open window and remain alive without evidence of disease for 19 months. Sleeve bilobectomy of right middle and lower lobectomy following 45Gy radiation was performed in 1 patient, sleeve right upper lobectomy combined resection of the superior vena cava following concurrent CRT with 60Gy was performed in 2 patients. Postoperative course for those patients were no problems at all. Postoperative bronchial healing was excellent. Conclusion Salvage lobectomy following high dose radiation could be feasible even performing radical resection of the superior vena cava or pulmonary artery.

Age/Gender	Regimen	Rt (Gy)	Stage & Hist-ology	Discrete?	Timing	Operation	Compli-cations
73 M	CBDCA+PAC×3　CBDCA+GEM ×5	60	IIIB Sq	infiltrative	12 months	Rt. Sleeve pneumonectomy	No
68 M	CBDCA+PAC×3　DOC×4 PEM×13	60	IIIA Ad	infiltrative	11 months	WWR	No
74 M	CDDP+VNR×2	60	IIIA NSCLC	discrete	7 months	RUL & S6 segmentectomy	Pneumonia
63 M	CBDCA+PAC×5	60	IIIB Sq	Infiltrative	9 months	RULobectomy	Non
56 M	CDDP+PEM×3　PEM×2	60	IIIA Ad	Discrete	12 months	Rt pneumonectomy	Non
40 M	CDDP+PEM×5 CDDP+DOC×2 VNR+Bev×2	140	IV Ad	Infiltrative	4 months	Rt pneumonectomy	BPF
72 F	CDDP+VNR×2	60	IIIA Ad	Discrete	84 months	RULobectomy	Empyema
61 M	CDDP+VNR×3	60	IIIA Ad	Infiltrative	15 months	Sleeve RUL SVC replacement	None
61 M	CDDP+VNR	60	IIIA NSCLC	Discrete	2 months	Sleeve RUL SVC PA plasty	None
60 F	CDDP+DTx	45	IIIA Ad	Discrete	3 months	Sleeve Middle and lower lobectomy	None

Abstract
Smoking is the most important cause of lung cancer and of death and morbidity from a wide range of causes. In most of Europe, North America and Australia recent results show that two-thirds of all deaths among male and female smokers in their 50s, 60s and 70s are caused by smoking. Smokers lose at least 10 years of lifespan. It takes decades for the full effects of smoking to emerge in a population. The hazards of smoking have been well described in men, but only recently have sufficiently large numbers of women in western countries been smoking for long enough for the full effects to be evident. For example, in the Million Women Study, a prospective study of 1.3 million UK women, lung cancer death rates in smokers were 20 times higher than in never smoker. Even in women who smoked fewer than 10 cigarettes per day, lung cancer rates were increased 10 fold, whereas among women who smoked about 25 cigarettes per day the risk was increased almost 40 fold. Risks were greater the younger women were when they started smoking. In western countries women began smoking in large numbers more than 50 years ago, and it is only now that results show clearly that when women smoke as much as men their risk of lung cancer and other conditions are similar. In Asia the full effects of smoking are still not evident, since large proportions of the population smoking began only a few decades ago. Stopping smoking substantially reduces the risks of lung cancer and of other conditions that would have occurred with continued smoking. Stopping at any age is beneficial. In the Million Women Study stopping at age 40 years was associated with a 3 fold increase in lung cancer, thus avoiding about 90% of the 20-fold excess mortality from lung cancer caused by continued smoking. Stopping smoking at around age 30 avoided about 97% of the excess risk of lung cancer. Similar findings have been reported from the USA and elsewhere. The avoidance of most of the excess risk of lung cancer and other adverse effects of continuing smoking by quitting before about age 40 years has major public health implications. It is estimated that during the 20th century smoking caused about 100 million deaths worldwide, but that it will cause ten times as many - 1000 million deaths - in the 21st century if current smoking patterns continue. If current smoking patterns continue almost all the smoking-related excess lung cancers and deaths in the next 50 or so years will occur in people who are already smoking. Much of the predicted epidemic of smoking-related deaths in the next 50 could be avoided if people who now smoke stopped, preferably before age 40.

Abstract not provided

Abstract not provided

Abstract
The purpose of this presentation is to define a potential role for biomarkers 1) in the risk assessment of lung cancer among individuals considered for screening and 2) in the diagnostic evaluation of screening detected lung nodules. We will distinguish types of biomarkers design for the screening of lung cancer and their desired performance characteristics. We will describe current front runner candidates biomarkers and discuss how one could envision using those in the clinic. Suggested reading: The state of molecular biomarkers for the early detection of lung cancer. Hassanein M, Callison JC, Callaway-Lane C, Aldrich MC, Grogan EL, Massion PP. Cancer Prev Res (Phila). 2012 Aug;5(8):992-1006.

Background
Background: Thymidylate synthase (TYMS) is an important chemotherapeutic target in non-small cell lung cancer (NSCLC). Recently, arsenic trioxide (ATO) has been shown to downregulate TYMS in a colonic cancer model. Therefore, we examined the effect of TYMS suppression by ATO in NSCLC.

Methods
Methods: A panel of seven NSCLC cell lines (NCI-H23, NCI-H358, HCC827, NCI-H1650, NCI-H1975, HCC2935 and HCC4006, from ATCC) was used to determine the effects of ATO treatment on cell viability, TYMS protein and mRNA expression, E2F1 protein expression and TYMS activity. TYMS knockdown and overexpression were performed to confirm the importance of TYMS in cell survival and resistance to ATO respectively. Tumor growth inhibition in vivo was studied using a nude mice xenograft model.

Results
Results: ATO showed antiproliferative effects with clinically achievable concentrations (around 1.1-6.9 μM after 72 hours incubation) in NSCLC cell lines. Baseline TYMS protein expression was detected in H23, H358, H1650 and H1975 cells. Among them, downregulation of TYMS protein and mRNA expression, reduced total TYMS activity, and suppressed E2F1 expression were demonstrated in H23, H358, and H1650 cells with ATO. Cell viability was reduced by 15-50% with TYMS knockdown (to less than 30% protein expression) (Fig. 1). Overexpression of TYMS led to a 2.7-fold increase in IC~50~ value with ATO treatment for 72 hours in H358 cells, but not H23 cells. Using a xenograft model with H358 cell line, relative tumor volume was reduced to 44% that of control after 8 days of treatment with 7.5 mg/kg intraperitoneal ATO, and associated with significant downregulation of TYMS protein expression in tumor xenografts (Fig. 2). Figure 1 Figure 2

Conclusion
Conclusion: ATO has potent in vitro and in vivo activity in NSCLC, and is partially mediated by transcriptional downregulation of TYMS.

Background
Lung cancer is the most widespread cause of cancer death in the world. COX-2 derived prostaglandin E~2~ (PGE~2~) is well known to induce tumor growth and metastasis, and thus is associated with a poor prognosis. Lymph node metastasis is also one of the major determinant of the prognosis and is facilitated by lymphangiogenesis, however the precise of the mechanisms is not well understood. In the presenr atudy, we investigated the role of COX-2-derived PGE~2~ on formation of premetastatic niche facilitate the lymph node metastasis in Lung Cancer. Lung cancer is the most widespread cause of cancer death in the world. COX-2 derived prostaglandin E~2~ (PGE~2~) is well known to induce tumor growth and metastasis, and thus is associated with a poor prognosis. Lymph node metastasis is also one of the major determinant of the prognosis and is facilitated by lymphangiogenesis, however the precise of the mechanisms is not well understood. In the presenr atudy, we investigated the role of COX-2-derived PGE~2~ on formation of premetastatic niche facilitate the lymph node metastasis in Lung Cancer.

Methods
Male C57BL/6 (6-8 weeks-old, Wild type; WT) and EP3 knockout mice (EP3[-/-]), one of PGE~2~ receptor subtype, were used. Orthotopic intrapulmonary implantation model was made by injecting Lewis Lung Carcinoma (LLC 5.0X10[6] cells/ml) cells into the lung. We estimated the expressions of lymphangiogenic factors, SDF-1/CXCR4, subtypes of EP receptors by immunehistochemical staining, ELISA and real time PCR. The accumulation of immature dendritic cells (iDCs) and regulatory T cells (Tregs) were estimated by immunofluolecent analysis. Furthermore, we estimated the role of iDC on lymph node metastasis by injecting iDCs.

Results
The expression of COX-2 and SDF-1, and accumulation of iDCs and Tregs were increased before LLC cell accumulated in lymph node. Compared to vehicle mice, mediastinal lymph node metastasis formations were suppressed with a COX-2 inhibitor (celecoxib, 100mg/kg/day, p.o., P<0.05). Compared to other PGE~2~ subtype receptors, the expression of EP3 receptor was significantly suppressed in celecoxib treated mice (P<0.05). The expression of lymphangiogenesis markers and lymph node metastasis were suppressed in EP3[-/-]. The expression of SDF-1 and accumulations of CD11c[+]DCs and FOXP3[+]Tregs in lymph nodes were significantly suppressed in EP3[-/-] (P<0.05). In vitro study, under PGE~2~ stimulation, the SDF-1 conentration and expression of EP3 in culutured iDCs were significanltly enhanced compared to control. Furthrmore, WT transplanted with EP3[-/-]BM were significantly suppressed mediastinal lymph node metastasis formation compared to WT transplanted with WT mice.

Conclusion
These results suggested that premetastatic niche formation in mediastinal lymph node was induced by bone marrow derived immatured dendritic cells via PGE~2~-EP3 signaling by induction of SDF-1-expression. Thus, COX-2 inhibitors, CXCR4 antagonists and/or EP3 antagonists may become one of the options to suppress the lymph node metastasis.

Background
A hierarchical model in which a small population of stem cells or Cancer Initiating Cells (CIC) may be responsible of tumour growth has been proposed. CIC is defined as a cell able to regenerate detectable neoplastic populations in xenografted immunodeficient mice. By this experimental approach, convincing data have led to the identification of CIC in solid tumours including breast, brain, colorectal, melanoma and lung. However, the autonomous growth ability of CIC is allowed in specialized structures, called niches, in which a microenvironmental control is exerted by several stromal elements including mesenchymal cells. These attractive pathogenetic mechanisms encompass innovative therapeutic implications because the limited success of the actual treatment of lung cancer may be attributable to the lack of targeting on CICs and their niches. The aim of the present study was to identify mesenchymal stem cells (MSC) from Non Small Cell Lung Cancer (NSCLC) and to investigate their biological properties and in vivo tumorigenic potential.

Methods
Small fragments of NSCLC (53 Adeno- and 24 Squamous carcinoma) surgically removed from 77 patients (46 male) were subjected to enzymatic digestion whereas most of the original sample was processed for diagnostic purpose. To identify mesenchymal cells from human lung cancer (hLc-MSC) we used different conditioning media and immunomagnetic selection to separate neoplastic epithelial cells frequently growing together with adherent stromal cells at early passages. To determine their tumorigenic properties, female immunodeficient BALB/c nude mice were subcutaneously injected with either 10[6] hLc-MSCs (from male donors) alone or in combination with 10[6] Calu-3, a human adenocarcinoma cell line lacking Y chromosome and able to reproducibly induce xenografted tumours.

Results
We have been able to obtain stable primary cultures of hLc-MSC, expandable for at least 14 passages, from 62% of NSCLC samples. This cell population uniformly expressed the typical MSC markers CD90, CD105, CD73, CD13 and CD44 at FACS analysis. hLc-MSC were negative for EpCAM, CD45, CD14 and CD34 whereas CD117 and CD133 were expressed at low frequency. Clones and subclones were documented by limiting dilution analysis. The nuclear expression of transcription factors involved in stemness (OCT 3/4 and SOX2), and in bronchio-alveolar (TTF1) or squamous (p63) lineage commitment was consistently present in hLc-MSC isolated from Adeno- or Squamous cell carcinomas. In vivo xenotransplantation demonstrated that although hLc-MSC administered alone did not generate tumours, their co-injection with Calu-3 developed in all animals subcutaneous nodules that were 10-fold larger than those induced by equal doses of Calu-3 injected alone. To detect the fate of both injected cells on sections of xenografted tumours, FISH analysis of human specific sex chromosomes was combined with immunohistochemistry. We documented that cytokeratin positive adenocarcinoma structures showing X chromosomes only (Calu-3 derived) were surrounded by a thin layer of non neoplastic cells carrying XY chromosomes (hLc-MSC derived). The cross talk and phenotypic changes of E-Cadherin positive neoplastic cells co-cultured with hLc-MSC were also documented.

Conclusion
Our study demonstrates that the reciprocal inductive interactions between the mesenchyme and the neoplastic epithelium are essential for the fate induction of lung cancer.

Background
Activating mutations of K-ras are one of the most common alterations associated with tobacco exposure-related lung cancer(LC). There are two major types of LC - small-cell LC(SCLC) and non-small cell LC(NSCLC) - and the later accounts for 80% of the cases. NSCLC can be divided into three histological subtypes: Adenocarcinoma, large cells and squamous cell carcinoma. First-line treatment comprises surgical resection of tumor, followed or not by chemo/radiotherapy. In non-surgical cases, platinum compounds remain the cornerstone for both early and advanced NSCLC stages management, in spite of its toxicity, high rate of chemo-resistance and poor long term results. Several attempts to develop therapies based on molecular targets, such as K-ras, have been developed and thus far failed, clearly stating the need for new approaches to bring clinical benefits to patients. Among its several aberrations, NSCLC harbors alterations in the apoptotic pathways, leading to impaired pro-apoptotic signaling and positive modulation of anti-apoptotic pathways. Therapeutic strategies targeting such pathways can emerge as an alternative to the cytotoxic therapies selected to wild-type EGFR-patients – especially for K-ras mutated patients, comprising about 20% of this population.

Methods
Here we have analyzed a representative panel of NSCLC cell lines (A549, H460, Calu-1 and LC319), that display distinct sensitivities towards cisplatin, mutational profiles and histological subtypes, to test a pre-clinical therapeutic strategy engaging the TNF-related apoptosis-inducing ligand (TRAIL) receptor (Death receptor 4 and 5).

Results
TRAIL is a member of the TNF family of cytokines that induces apoptotic cell death in a variety of tumor cells by means of activation of its specific receptors, DR4 and DR5, while displaying low toxicity towards normal cells. We sought to investigate if DRs activation could subvert the relative resistance to cisplatin intrinsically presented by NSCLC cell lines. NSCLC cell lines treated with suboptimal concentrations of cisplatin (IC30) for 48h had their gene expression analyzed by qPCR and the results showed an increased expression of DR4 and DR5 in these cells. These results were confirmed at protein level by Western blot analysis. NSCLC cells are naturally regarded as resistant to TRAIL-induced cell death. Such resistance can rise, among other reasons, from low expression of DRs or increased expression of decoy receptors and/or anti-apoptotic proteins. LBY135 (Novartis) is a DR5 agonist monoclonal antibody that mimics TRAIL and induces cell death in DR5 expressing cells. As cisplatin modulated DR5 protein expression, we have combined it with LBY135 as a strategy to improve cell death induction upon NSCLC cells. LBY135 monotherapy did show some effects when analyzed by MTT assay, although it did not induce cell death accessed by Annexin/PI FACS analysis. However, when cisplatin IC30 and LBY135 were combined, we observed a significant decrease in MTT measurements and increased cell death incidence, suggesting a synergistic effect of these drugs. Such pro-apoptotic effect was blocked by zVAD-fmk, a pan-caspase inhibitor.

Conclusion
The synergistic effect observed was more pronounced in cell lines bearing the classical G12K-ras mutation, suggesting an alternative way to subvert chemo-resistance of K-ras mutated NSCLC and restore cisplatin-induced apoptotic signaling leading to cell death.

Background
Arsenic trioxide (As~2~O~3~), an ancient drug used in traditional Chinese medicine with the characteristics of high effectiveness and low toxicity, was used to treat acute promyelocytic leukemia at the beginning . Over the past decade, in our clinical practice, local injection of As~2~O~3~ into pleural cavity has been successfully used for more than 80 patients with malignant pleural effusions(MPE) caused by pleural metastasis of lung cancer. And interestingly, after local administration of As~2~O~3~, the composition of blood cells in pleural effusion was reduced and the color of MPE changed from red to light yellow in the majority of patients, indicating that As~2~O~3~ could be considered as an effective approach for the treatment of MPE, but the underlying mechanism remains unclear . The aim of this study was to investigate the mechanism of arsenic trioxide (As~2~O~3~) in the treatment of malignant pleural effusion (MPE) caused by pleural metastasis of lung cancer.

Methods
A mouse model of MPE caused by pleural metastasis of lung cancer was first established in our study, and then As~2~O~3~ was intraperitoneally injected to treat MPE. Those mice treated with bevacizumab and bleomycin were included as positive control animals, and placebo equivalents were also employed as negative control. The effects of As~2~O~3 ~on MPE volume, pleural vessel density, vascular permeability, expression of angiogenic function related factors including VEGF and TNF-α, as well as NF-κB activity in pleural carcinomatosis were observed.

Results
Intraperitoneal injection of As~2~O~3~ could reduce the volume of MPE and decrease vascular density and vascular permeability in pleural metastatic nodules, in a dose-dependent manner. Moreover, the dose-dependent decrease in VEGF and TNF-α expression in MPE and NF-κB activity in pleural carcinomatosis were also found after As~2~O~3 ~treatment.

Conclusion
Our work demonstrated that As~2~O~3~ could down-regulate the VEGF expression through NF-κB inhibition, and then decrease vascular density and permeability in pleural metastatic nodules, thereby playing its effects on MPE caused by pleural metastasis of lung cancer. Our results could provide a foundation for As~2~O~3~-based clinical treatment program.

Background
Small-cell lung cancer (SCLC) is the most aggressive subtype of lung cancers, with no early detection strategy or targeted therapy currently available. We hypothesized that difference gel electrophoresis (DIGE) may allow the identification of membrane-associated proteins specific to SCLC, the advancement of our understanding of SCLC biology, and the discovery of new candidate diagnostic or therapeutic biomarkers.

Methods
Membrane-associated protein lysates were prepared in quadruplicate from three SCLC, three non-small-cell lung cancer (NSCLC), and three immortalized normal bronchial epithelial cell lines. The 36 samples were co-analyzed by DIGE and subsequent protein identification was performed by mass spectrometry (MS). Identified proteins were submitted to Ingenuity pathway analysis (IPA). Candidate biomarkers were validated by Western blotting (WB) and immunohistochemistry (IHC), and tested against clinical outcomes.

Results
Principle component analysis on the global DIGE dataset demonstrated that the four replicates derived from each of the nine cell lines clustered very closely from one another, as did samples within the same histological group. A total of 135 distinct protein features were differentially expressed in SCLC as compared with NSCLC and bronchial airway epithelial cell lines (P<0.001). Those included 137 different proteins identified by tandem MS. IPA revealed that these proteins were overrepresented in the cellular assembly, organization, morphology, and tissue morphology networks. DPYL2, GNAQ, RSSA, RUVB1, and STMN1 were found to be the most discriminatory candidate biomarkers among the membrane-associated proteins overexpressed in SCLC as compared with NSCLC and normal bronchial airway cell lines. Overexpression of DPYL2, GNAQ, and STMN1 was verified in SCLC cell lines by WB. Intense protein expression of DPYL2, GNAQ, RUVB1, and STMN1 was also confirmed in SCLC by IHC on tissue microarrays (TMA). These proteins’ expression levels measured by IHC were significantly associated with the SCLC subtype and survival in a univariable analysis but could not be verified as independent in a multivariable analysis.

Conclusion
Proteomic analysis of membrane-associated proteins in lung cancer and bronchial airway epithelial cell lines revealed 137 proteins differentially expressed in SCLC. These proteins were enriched for cellular assembly, organization, morphology, and tissue morphology networks. Of the five proteins selected for clinical validation, DPYL2, GNAQ, RUVB1, and STMN1 overexpression in SCLC was verified by WB and IHC, suggesting that these results need to be tested for functional implication in SCLC progression. The association with survival requires further validation in a larger clinical dataset. Funding This work was supported by a VA Merit review award 1I01CX000242 to PPM. SO was supported by a Télévie Grant, a Fondation Mont-Godinne Grant and a Clinician-Researcher Mandate from Secteurs des Sciences de la Santé, Université Catholique de Louvain (UCL), Belgium.

Background
Cancer cells have defects in regulatory circuits governing proliferation and homeostasis. Consequently, cell metabolism is altered to meet the demand for accelerated, deregulated growth. Metabolic perturbations arising from malignant transformation have not been well characterized in human lung cancers in situ. The most well known metabolic derangement(s) in tumors is that of enhanced glycolysis and a decrease in mitochondrial oxidative phosphorylation. We wanted to characterize this phenomenon more accurately in human lung adenocarcinomas by metabolomic profiling.

Methods
We performed metabolomic analysis of matched pairs of solid, non-small cell lung adenocarcinomas and normal lung tissue from 25 surgically resected patients. Metabolites were extracted by a methanol-chloroform-water technique. The resulting extracts were divided into multiple fractions. Ultrahigh performance liquid chromatography/ mass spectrometry coupled with tandem mass spectrometry and gas chromatography/ mass spectrometry experiments were conducted. Agilent MassHunter Qualitative software was utilized. The Molecular Feature Extractor was utilized to find features in raw data files. Extracted peaks were retention time aligned using Mass Profiler Professional and unique features detected by least squares analysis. The Agilent version of the Metlin database was utilized to identify metabolites. Matched pairs t-test identified biochemicals significantly altered between tumor and normal specimens. The false discovery rate method assessed for significance; p-value ≤ 0.05 and q-value < 0.10.

Results
Based on known library standards to identify biochemicals, our global metabolomic profiling found 204 overexpressed and 42 underexpressed metabolites in tumors relative to normal lung (p< 0.05). We observed altered metabolite levels in lung tumors that mapped to not one, but two glucose utilization pathways. Glucose-6-P (2.7-fold), fructose-6-P (2.6-fold), fructose-1,6-bisP (6.9-fold), lactate (2.7-fold), and NAD[+] (1.4-fold) were significantly upregulated in tumors consistent with an aerobic glycolysis (i.e. Warburg) biosignature, the major source of ATP. Concurrently, pentose phosphate pathway (PPP) metabolites were upregulated in tumors: ribulose-5-P (2.6-fold), ribulose (3.6-fold), ribitol (4.6-fold), ribose (4-fold), and sedoheptulose-7-P (3-fold). Our data reveals evidence of multiple active pathways to explain glucose utilization in lung adenocarcinomas. The PPP is important to protect against oxidative stress as it serves to generate NADPH, and is a key anabolic pathway of nucleotide synthesis by generating the ribose-5-P backbone for proliferating cells. Observing both pathways simultaneously in lung adenocarcinomas suggests they are coupled to give tumors a growth advantage over normal tissue. Consistent with this, we observed an overall increasing nucleotide biosynthesis signature in tumors: multiple metabolites (range 2 to 17-fold) in purine and pyrimidine pathways were significantly elevated.

Conclusion
Metabolomic analysis identified a unique glucose energetic biosignature in lung tumors that is more complex that just a single process/ pathway. Our results suggest a specific strategy to target lung adenocarcinomas by exploiting their high glucose uptake.

Background
The etiology of malignant pleural mesothelioma (MPM) is closely linked with asbestos exposure. Asbestos is capable of inducing chronic inflammation which potentiates tumour suppressor gene silencing. Epigenetic silencing of the Wnt pathway, well characterized in the progression of colon cancer, is associated with chronic inflammation. As antagonists of Wnt pathways, the SFRPs are functional tumour suppressors of colon, gastric, breast, ovarian and lung cancers, with some members methylated in mesothelioma. In this study, we aimed to investigate the functional significance of the SFRP2 and 5 in MPM, and the effect of long-term asbestos exposure on epigenetic alteration in the immortalised mesothelial cell MeT-5A.

Methods
Gene expression and promoter DNA methylation of the SFRP family were analysed in MPM lines and MeT-5A with and without 5’Azacitidine treatment using RT-qPCR, MSP and COBRA. The effect of SFRP2 and SFRP5 re-expression on MPM cells was determined by cell growth and clonogenic assays in 2D and 3D culture. The expression and promoter DNA methylation of SFRP genes was also assessed in MPM patient samples using RT-qPCR and MSP. MeT-5A cells were cultured long-term (12 months) in the presence of asbestos, and SFRP mRNA expression and promoter DNA methylation was analysed.

Results
SFRP2 and SFRP5 were either absent or down-regulated MPM lines, and restored after 5’Azacitidine treatment. SFRP1 was highly expressed and unmethylated in MeT-5A line. Expression of the SFRP family was down-regulated in MPM patient samples and this correlated with methylation of promoter CpG islands. Ectopic expression of SFRP2 or SFRP5 inhibited MPM cell growth and colony formation in both 2D and 3D culture. SFRP1 was down-regulated and methylated following prolonged asbestos exposure in MeT-5A cells.

Conclusion
Our results indicate that both SFRP2 and SFRP5 function as tumour suppressor genes in MPM and long-term asbestos exposure induce gene silencing via DNA hypermethylation.

Background
Most patients die from lung cancer because of chemoresistant metastatic disease. We have previously identified the importance of fibroblast growth factor 2 (FGF2) /S6 kinase 2 (S6K2) signalling in mediating multidrug resistance in lung cancer through enhanced translation of antiapoptotic proteins, such as BCL-XL and XIAP. Here we investigate the downstream mediator(s) of this translational response and demonstrate its relevance in a lung cancer tissue array.

Methods
We used tandem affinity purification using S6K2 as bait as well as quantitative phospho-proteomics in the presence and absence of FGF2 stimulation in small cell lung cancer and HEK 293 cells to identify new S6K2 interactors and downstream mediators of the FGF2 pathway. Interactors were validated by co-immunopreciptation experiments and their functional significance examined by siRNA and expression of wild-type or phosphomutant forms. The clinical significance of our in vitro findings were examined in lung cancer tissue arrays.

Results
Here, we show that S6K2 interacts with and phosphorylates the heteronuclear ribonuclear protein hnRNPA1 on Ser 4 and 6. This increases the association of this protein with BCL-XL and XIAP mRNAs to promote their nuclear export while de-repressing their translation. A non-phosphorylatible S4/6A hnRNPA1 mutant prevented this process from occurring and impaired the pro-survival activity of FGF2/S6K2 signalling. Following phosphorylation and transfer to the cytoplasm in complex with mRNAs, phospho-hnRNPA1 associates with 14-3-3 to be sumoylated on K189 within a multi-protein complex involving UBC9. This targets hnRNPA1 for re-import into the nucleus in a caryopherin-dependent manner, a step that is essential for translational derepression of target mRNAs. Our in vitro findings predicted that in cancer cells where FGF2/S6K2 signalling is switched on, hnRNPA1 would be predominantly localised to the nucleus and cytoplasmic expression of anti-apoptotic proteins such as BCL-XL would be increased. To test this hypothesis, we stained a NSCLC tissue array for S6K2, hnRNPA1 and BCL-XL expression. Strikingly, this revealed that increased S6K2 expression tightly correlated with decreased cytoplasmic hnRNPA1 and increased BCL-XL levels.

Conclusion
FGF-2/S6K2 signalling promotes the nucleo-cytoplasmic cycling of hnRNPA1 to promote tumour cell survival through increased translation of the anti-apoptotic proteins BCL-XL and XIAP. Our immunohistochemical findings in NSCLC suggests that tumours which show absence of cytoplasmic hnRNPA1 in the presence of increased S6K2 and Bcl-XL expression might respond better to FGF receptor inhibitors.

Background
Tumor-derived microvesicles (TMV) contain bioactive molecules including proteins, RNA, microRNA, and DNA, and shed TMV propagate the horizontal transfer of their cargo in tumor microenvironment. TMV play important roles in cancer biology such as tumor progression, angiogenesis, escape from immune surveillance, metastasis and drug resistance. In this study, we investigated whether T790M EGFR mutation-related gefitinib resistance could be transferred via TMV using lung cancer cells harboring exon 19 del mutation (PC-9 cells) ± exon 20 T790M mutation (PC-9/GR cells).

Methods
TMV were isolated from the culture medium by serial centrifugations. Isolated TMV were checked by electron microscopy for size and intact structure, and characteristic TMV markers (CD81 and ezrin) were verified by western blot analysis. Cytotoxic assay was done using XTT assay and EGFR mutation testing was done using PNA-clamping method.

Results
Incubation of PC-9 cells with TMV (40 μg/ml) derived from PC-9/GR cells confer PC-9 cells gefitinib-resistant. EGFR genotyping of TMV in both cells revealed as the same with cellular DNA and this means that TMV carry cell-specific DNA. Passage study after PC-9/GR cell derived TMV transfer showed, however, regained gefitinib sensitivity suggesting that TMV-mediated gefitinib resistance is phenotypically transient. Genotyping after T790M-containing TMV revealed EGFR WT.

Conclusion
T790M-associated gefitinib resistance in lung cancer cells might be horizontally transferred via TMV but this type transfer of gefitinib resistance was phenotypically transient in cell-line system and there was no genotype transfer. Further investigation for TMV-mediated transfer of gefitinib resistance is necessary.

Background
Lung adenocarcinoma (AC) and squamous cell carcinoma (SCC) tumours have a large variance in tumour cell content. This heterogeneity is a concern for genomic studies, as it is difficult to distinguish mutational differences between tumour and non-tumour if low percentage tumour is used for analysis. In addition to this, tumour samples are affected by the amount of necrosis present, as the overall number of viable cells is decreased. We assessed tumour and necrotic content in lung tumour specimens from AC and SCC patients and aimed to identify possible implications for the suitability of these samples in molecular characterisation studies using next generation sequencing technology.

Methods
Lung tissue specimens were collected during the period of 1990 to 2013 from patients at The Prince Charles Hospital who consented to donate their surgically resected lung tissues for research. Tissues were macroscopically dissected, snap frozen in liquid nitrogen and stored at -80°C. A tissue section was taken and stained with haematoxylin and eosin (H&E) for two pathologists to independently assess tumour cell and necrotic content. Tumour cell content (TC) in each specimen was scored as percentage of viable cells as seen on the H&E slide, where necrotic content (NC) was recorded as a percentage of the whole slide section. Statistics were calculated using SPSS v21 software. Tumour specimens screened for eligibility to The Cancer Genome Atlas sequencing project are presented here.

Results
Tumours from 62 AC and 104 SCC subjects were scored (specimen characteristics in Table 1). Scoring between the two pathologists was highly correlated, with a high intraclass reliability (0.94 and 0.96 for TC and NC respectively).

Table 1: Clinical and Pathological Characteristics of Specimens

	AC	SCC
Number of Specimens	384	609
Number of Males/Females	36/26	84/20
Median Specimens per Subject	4	4
Range of Specimens per Subject	1-25	1-27
Median TC	35%	30%
Range of TC	0-88%	0-90%
Median NC	0%	6%
Range of NC	0-90%	0-100%
Median Age	62 yrs	68 yrs
Range of Age	45-85 yrs	46-91 yrs
Median Smoking Pack Years	40	56
Range of Smoking Pack Years	0-115	0-158

TC varied from 0-~90% for both subtypes. Comparing AC and SCC, the median TC was higher in AC than SCC (35% vs 30% respectively, p<0.05). NC varied from 0-~100%, but was generally low. The median NC was statistically significantly different between AC and SCC (0% and 6% respectively, p<0.001). TC was weakly correlated with NC (Spearman Rank r = 0.32, p<0.01). There were no clinically important correlations between smoking pack years, gender or age with TC and NC of specimens.

Conclusion
Lung AC and SCC specimens are heterogeneous in terms of TC and NC. Therefore, only a small proportion of resected lung cancer specimens meet the criteria required for massively parallel sequencing projects that require high quality tumour DNA and RNA (ie low NC) and relatively low stromal contamination (ie high TC).

Background
PI3K pathway activation in NSCLC has been shown by us and others to lead to a more aggressive disease correlating to poor prognosis for patients. Unfortunately, the success of PI3K targeted inhibition has been hampered by a high rate of innate and acquired resistance. Mutations in KRAS and B-RAF, ERK hyperactivation as well as extensive PI3K-MEK pathway cross-talk allow the MEK pathway to provide a bypass track. Preclinical studies demonstrate a rationale for a PI3K-MEK co-targeted treatment strategy which may provide a more effective response. A Phase I clinical trial is underway investigating the combination of GDC-0941, a pan-PI3K inhibitor, with GDC-0973, a MEK inhibitor. GDC-0980 is a dual PI3K-mTOR inhibitor which may offer improved pathway inhibition compared to GDC-0941. No data has been published to date on the combination of GDC-0980 and GDC-0973, which we believe may offer improved overall inhibition of survival signaling in NSCLC cells. We aim to elucidate the role of mutation status in response to this co-targeted inhibition approach in vitro, as well as investigating the frequency of PI3K and MEK pathway mutations in a well characterized Irish NSCLC patient cohort.

Methods
The effects of GDC-0941, GDC-0980 and GDC-0973 on proliferation and apoptosis in a panel of four NSCLC cell lines were analysed by BrdU Assay and HCA Apoptosis Assay, respectively. The four cell lines investigated were H460 (adenocarcinoma, PIK3CA mutant & KRAS mutant), A549 (adenocarcinoma, PIK3CA wild type & KRAS mutant), H1975 (adenocarcinoma, PIK3CA mutant, KRAS wild type & EGFR TKI resistant) and SKMES-1 (squamous cell carcinoma, PIK3CA wild type & KRAS mutant). Further investigation involved expression analysis of pAkt, pGSK-3β, pp70S6K, pS6RP, ERK and pERK in cell lines treated with each inhibitor alone or in combination using Mesoscale technology and Western blot. DNA was extracted from 120 NSCLC patient tissue samples, and screened for 547 mutations in 59 genes (including PI3K and MEK pathway members) using the Sequenom.

Results
GDC-0941 and GDC-0980 treatment induced dose-dependent anti-proliferative and pro-apoptotic responses across all four NSCLC cell lines, while GDC-0973 treatment induced only anti-proliferative responses. Protein expression analysis showed that GDC-0980 & GDC-0973 combination treatment induced significantly improved phosphoprotein inhibition compared to treatment with either inhibitor alone in cell lines harbouring PIK3CA mutations, while in one cell line bearing WT PIK3CA (SKMES-1), combination treatment actually increased pathway signalling. NSCLC patient mutational profiling data will be presented.

Conclusion
This research underpins the importance of mutation status in sensitivity to targeted therapies. While combination treatment approaches may be beneficial in certain molecular subtypes, in others they may be detrimental. In the era of personalised medicine, patient genotyping is crucial to improve patient survival and reduce toxicities.

Background
Malignant pleural mesothelioma (MPM) is a rare aggressive cancer of the pleura. Asbestos exposure (through inhalation) is the most well established risk factor for mesothelioma. The current standard of care for patients suffering from MPM is a combination of cisplatin and pemetrexed (or alternatively cisplatin and raltitrexed). Most patients however, die within 24 months of diagnosis. New therapies are therefore urgently required for this disease. Lysine acetyltransferases (KATs) including KAT5 have been linked with the development of cisplatin resistance. This gene may therefore be altered in MPM and could represent a novel candidate target for intervention

Methods
A panel of MPM cell lines inluding the normal pleural cells LP9 & Met5A were screened for expression of KAT5 variants by RT-PCR. Levels were subsequently examined in a cohort of snap-frozen patient samples isolated at surgery comprising benign, epithelial, biphasic, and sarcomatoid histologies by RT-PCR. The effects of a small molecule inhibitor of KAT5 (MG-149) on cellular proliferation were examined.

Results
We show that the expression of KAT5A is dramatically increased in MPM. When separated according to histological subtype, KAT5A was significantly overexpressed in both the the sarcomatoid and biphasic subgroups for all transcript variants. Treatment of cells with the small molecule MG-149 caused significant inhibition of cellular proliferation (p<0.0001). We continue to asess this compound by other methodologies to confirm its potential utility in the treatment of MPM.

Conclusion
KAT5, a lysine acetyltransferase associated with cisplatin resistance in cancer is significantly altered in MPM. A small molecule inhibitor of this protein shows significant anti-proliferative effects in MPM cell lines. Targeting this protein may have important future implications for the management of MPM.

Background
Background: The PI3K/AKT signaling pathway is aberrantly active and has an important biologic impact in malignant pleural mesothelioma (MMe) cell cycle progression and chemo-resistance. Akt consists of three isoforms, Akt1, Akt2, and Akt3. Despite the growing amount of research demonstrating the existence of isoform-specific regulation, many papers still draw generalized conclusions about AKT without focusing on functional specificity of each isoforms. Recent data have clearly demonstrated a role of SIRT1 in the modulation of AKT1 activation and a role of PARP1 as a gatekeeper for SIRT1 activity by limiting NAD+ availability.

Methods
Methods: We explored the expression of AKT isoforms in MMe in vivo and in vitro and the balancing between their acetylation and phosphorylation status in human MMe derived cell lines in vitro.

Results
Results: We firstly described that MMe tumors in vivo and MMe derived cell lines express both AKT1 and AKT3 isoforms but not AKT2, and their expression results significantly increased in the biphasic histotype. Furthermore, we demonstrated an inverse correlation between AKTs acetylation and phosphorylation modulated by PARP1/SIRT1 activation status. By immunoprecipitation experiments, we evidenced that in basal conditions AKT1 is in part acetylated and in part phosphorylated and became highly phosphorylated and completely de-acetylated upon PARP1 inhibition. Interestingly, AKT1 activation, related to PARP1 inhibition, is unable to modulate pro-survival signals, because the downstream pathway is interrupted at the level of its effector mTOR. Conversely, SIRT1 inhibition or silencing result in a more evident AKT1 and its interactors acetylation.

Conclusion
Conclusions: In conclusion, our results clearly show how both PARP1 and SIRT1 affect critical cellular pathways involved in MMe progression and offer a model of a regulatory inter-relationship between these proteins. These data could be helpful for designing new effective therapeutic strategies.

Background
REV3, the catalytic subunit of the translesion synthesis (TLS) polymerase x, can continue replication past DNA adducts. Depletion of REV3 sensitizes A549 lung cancer cells to cisplatin. REV3 expression is part of a gene signature that predicted pemetrexed sensitivity in 17 NSCLC cell lines. BRCA1, TS, AEG1 and RAP80 are involved in DNA damage repair through homologous recombination. The homologous recombination and TLS pathways have non-redundant functions in response to cisplatin. We hypothesized that low mRNA expression of these genes – either alone or in combination – could confer improved outcome to cisplatin/pemetrexed in NSCLC patients.

Methods
REV3, BRCA1, RAP80, TS and AEG1 mRNA expression was examined by quantitative RT-PCR and categorized by terciles. Expression of each gene was correlated with outcome in 47 cisplatin/pemetrexed-treated NSCLC patients.

Results
63.8% male; 47% smokers; 80.9% ECOG PS 1; 80.8% adenocarcinoma. Overall response rate was 51%, with no differences according to expression levels of any of the genes. Progression-free survival (PFS) for patients with low, intermediate and high BRCA1 levels was 13.4, 5.5 and 3.9 months (m), respectively (P=0.005). Similar differences in PFS were observed according to TS (P=0.003) and AEG1 (P<0.001) expression. The hazard ratio (HR) for PFS for patients with high BRCA1 levels was 4 (P=0.002). Overall survival (OS) for patients with low, intermediate and high BRCA1 levels was 29.7, 7.4 and 6.3 m, respectively (P=0.05). Similar differences in OS were observed according to TS (P=0.005) and AEG1 (P=0.001) expression. HR for OS for patients with high BRCA1 levels was 3.6 (P=0.004). There were no differences in PFS or OS according to REV3 or RAP80 levels. However, the joint effect of BRCA1 and REV3 was significant for predictive modeling. PFS for patients with low, intermediate and high levels of both genes was 14.9, 7.2 and 2.8 m, respectively (P=0.001). OS for patients with low, intermediate and high levels of both genes was 29.7, 7.8 and 6.3 m, respectively (P=0.04).

Conclusion
Low BRCA1 expression predicts longer PFS and OS in pemetrexed/cisplatin-treated NSCLC p. Low TS and AEG1 levels have similar predictive value. The combination of low BRCA1 and REV3 expression confers longer PFS and OS. Analysis of these genes could be useful for customizing pemetrexed/platinum chemotherapy.

Background
Background: Small cell lung cancer (SCLC) is notoriously a highly fatal disease, with rapid emergence of resistance to first-line chemotherapy. Arsenic trioxide (ATO) has been used as a standard treatment for acute promyelocytic leukemia for the past decade. Our recent data have demonstrated in vitro activity of ATO, with clinically relevant concentrations, in SCLC cell line model. Although induced oxidative stress has been implicated as a possible mechanism of cytotoxicity, the exact mechanism of action of ATO in SCLC is not fully elucidated. In this study, we further explore the potential genetic changes and biological properties of acquired resistance to ATO in SCLC.

Methods
Methods: Using H841 SCLC cell line as a model, a daughter cell line resistant to ATO (H841AR) was established by culturing H841 cells in medium with progressively increasing concentrations of ATO for 12 months. Resistant clones were then selected and cultured in HITIS medium containing 20 μM ATO. Total RNA was extracted from both H841 and H841AR cells, followed by reverse transcription and hybridization with Affymetrix EXON 1.0 ST array. Gene chip signals were analyzed by Gene Spring 12 software. Cell proliferation (MTT) assay, wound healing (migration) experiment and invasion assay were performed to compare between H841 and H841AR cells.

Results
Results: Exponential growth of H841AR cells was shown in HITIS medium with 20 μM ATO. Comparing with H841 cells, 17 genes were upregulated by at least 5-fold in H841AR cells, consisting of stress response factors, immune system regulators and proliferation-associated or transcriptional factors. Similarly, 45 genes were downregulated by more than 5-fold in H841AR compared with H841 cells, mainly involving angiogenesis, signal transduction and neurodifferentiation. Apart from resistance to ATO (resistant index (RI) = 12.5 comparing H841AR cells with H841 cells), H841AR cells were also slightly resistant to cisplatin (RI = 2) compared with H841 cells. H841AR cells demonstrated faster proliferation, with greater migration and invasion properties.

Conclusion
Conclusion: There are distinct genetic alterations and biological properties in SCLC cell line (H841AR) with acquired resistance to ATO. Further analysis of the functional significance of various genetic alterations in ATO-resistant SCLC cell line is warranted. This study is partly funded by the CRCG Small Project Funding from the University of Hong Kong

Background
Recent studies in our laboratory show that tumor cells exposed to progressively higher concentrations of DETA-NONOate grow more aggressively than the parent cells. The currently accepted tumor stem cell model proposes that rare tumor stem cells produce both tumor cells (TCs) and rare tumor stem cells (TSCs). A number of TSC markers have been proposed, including CD antigens expression, Hoerscht 33342 exclusion, ALDH1/2 expression, and ALDH1/2 activity. In this study, we compare the TSC-like properties of parent and HNO cells. We hypothesized that human lung tumor cells adapted to high nitric oxide (HNO) levels will exhibit enhanced tumor stem cell-like properties relative to analogous cells not adapted to NO (“parent” cells).

Methods
The human lung adenocarcinoma cell line A549-Parent and the previously reported A549-HNO adapted cell lines were used in these studies. Immunohistochemistry was used to measure the expression of ALDH1/2 and six different CD antigens. Gene expression was measured using chip microarray analysis. FACS analysis was used to measure the population of H33342-positive and H33342-negative cells. ALDH activity was assessed using the ALDEfluor assay.

Results
The ALDEFLUOR assay demonstrated that ALDH activity of A549-HNO has increased compared to that of A549-parental. Relative to the A549-parental cells, A549-HNO cells showed markedly higher H33342-exclusion in the FACS analysis and greater ALDH1/2 expression in the immunostaining studies. Shocking these cells with 700 mM NO donor prior to treatment produced even more pronounced stem cell-like properties, suggesting that NO might be able to influence TCs into becoming more TSC-like. Gene chip results show that CD44, ALDH1A1, and ALDH2 are more highly expressed in A549-HNO cells than in A549-Parent cells. Cytospin results showed that ALDH1/2, CD44, and CD133 were more highly expressed in the A549-HNO cells, relative to A549-Parent cells.

Conclusion
The TSC markers tested in this study are consistent with the A549-HNO cells being more TSC-like than the A549-Parent cells. Having a large population of TSC-like cells would allow for improved drug screening of therapeutic candidates specifically designed to target TSCs.

Background
Hypothesis: Tumor stem cells (TSCs) are not rare cells and tumor cells TCs) have the plasticity to become TSCs. Objectives: Currently the rate limiting step in these efforts is the use of fluorescence-activated cell sorting (FACS) instrumentation to physically sort TSCs from TCs. We have shown that tumor cell lines could be adapted to comparatively high Nitric Oxide (NO) levels and that these cells had properties similar to TSCs. These cells were tested for various TSC properties using cellular, immunological, and molecular biology methods which included ALDH expression, ALDH activity, CD antigen expression, and H33342 dye exclusion. Proving that these are in fact TSCs would mean that we can produce large quantities of these cells to study. In a similar fashion, proving that TCs can become TSCs, would be very important.

Methods
We used the A549 and several Head & Neck SCC cell line in this study. They were FACS sorted five to seven times consecutively using H33342, to remove any chance of TSCs being present. The resulting TCs were cultured, and retested using H33342 at various times (several days to several weeks), for the presence of TSCs. The cells were cultured with or without NO.

Results
The FACS-sorted “pure” TCs, after being cultured, resulted in a TSC population re-immerging within days.

Conclusion
Collectively, these data support the idea that TSCs are not rare cells and that NO can drive TC populations to express properties of TSCs. This plasticity means that any TC can become a TSC.

Background
Malignant mesothelioma (MM), strongly associated with exposure to asbestos, is a growing worldwide problem (1). This aggressive tumour is largely resistant to oncological treatments and new approaches to therapy are urgently needed. Integrins are a class of adhesion molecules composed of an α and a β chain. Combinations of 18 α and 8 β subunits form the 24 members of the integrin family. The αv subunit can dimerize with β1, β3, β5, β6 and β8. Aberrant expression of αv integrins was reported in MM, and the integrins αv β3 and αv β5 have been implicated in tumour progression and metastasis. We have investigated the expression and function of αv integrins in MM cell lines and the effect of gene knockdown on cell invasion.

Methods
Expression of the integrin (ITG) genes was analysed by qPCR in 7 MM cell lines. Expression of the heterodimers was determined by Western blot, immunofluorescence and immunocytometry (monoclonal antibodies kindly provided by Simon Goodman, 2). In addition, we knocked down the genes potentially involved in tumour progression (ITGB3 and ITGB5) and analysed the in vitro 2D and 3D invasiveness with an agarose spot invasion assay (3) and MM spheroids.

Results
All 7 MM cell lines showed high ITGB1 expression, moderate ITGB5 expression, and a general low ITGB6 and ITGB8 expression. ITGB3 was expressed in one cell line, which accordingly had high αv β3 expression. ITGB3 knockdown of this cell line resulted in suppression of invasion both in 2D and 3D cultures.

Conclusion
We have found evidence that integrin αv β3 may play a role in MM invasion. Presently, we are testing cilengitide, a peptide antagonist of integrin αv, in MM cell lines. References: 1. Delgermaa F, Takahashi K, Park EK, Le GV, Hara T, Sorahan T. Global mesothelioma deaths reported to the World Health Organization between 1994 and 2008. Bull World Health Organ. 2011, 89:716-724. 2. Goodman SL, Grote HJ, Wilm C. Matched rabbit monoclonal antibodies against αv-series integrins reveal a novel αvβ3-LIBS epitope, and permit routine staining of archival paraffin samples of human tumors. Biol Open. 2012, 1:329-340. 3. Wiggins H and Rappoport J. An agarose spot assay for chemotactic invasion. Biotechniques. 2010, 48:121-124.

Background
Lung cancer remains the most common cause of cancer-related death in the world. The major advances in treatment of lung cancer have brought only minor improvements in survival; therefore novel strategic treatment approaches are urgently needed. Accumulating data allocate a central role for the cancer microenvironment including mesenchymal stem cells (MSCs) in acquisition of drug resistance and disease relapse. Therefore, we decided to study the effect of bone marrow (BM) MSCs on non small cell lung cancer (NSCLC) cell lines. Recent studies indicate that translation initiation factors are over expressed in multiple cancers and negatively impact NSCLC prognosis. Interestingly, translation initiation is highly modulated by microenvironmental cues. Hence, special emphasis will be attributed to the role of translation initiation in the crosstalk between the malignant lung cells and the BM-MSCs.

Methods
Specifically, we will determine the influence of MSCs (co-culture) and MSCs conditioned media on the proliferation, viability, migration of NSCLC cell lines (A549, H1299, and H460). We will also explore the effect on translation initiation factors implicated in cancer progression (eIF4E, eIF4GI, DHX29).

Results
Preliminary results demonstrate that 48h co-culture of NSCLC cell lines with MSCs conditioned media significantly decreased cell number (A549: 60%, H1299: 50%, H460: 10%, p<0.05) and viability (A549: 70%, H1299: 50%, H460: 60%, p<0.01) yet had no effect on cell cycle or death. Moreover, MSCs conditioned media decreased NSCLC cells motility (H1299: 60%, H460: 70%, p<0.05) and attenuated eIF4GI expression and activity as indicated by the levels of its targets (H1299: peIF4G 65%, total eIF4G 40%, SMAD5 30%, cMyc 80%; H460: peIF4G 90%, SMAD5 85%, cMyc 40%, p<0.05).

Conclusion
Ongoing work in our lab is aimed at dissecting whether the effects are dependent on direct cell-cell contact or mediated by soluble factors. Results that indicate translation initiation is modulated by MSCs derived components will mark new therapeutic targets for lung cancer treatment.

Background
A3 adenosine receptor mediates apoptosis in cancer cells via diverse signaling pathways. The present study examined A3 adenosine receptor-mediated apoptosis in Lu-65 cells, a human giant cell lung carcinoma cell line.

Methods
MTT assay, TUNEL staining, real-time RT-PCR, Western blotting, and assay of caspase-3, -8, and -9 activities were carried out in Lu-65 cells, and A3 adenosine receptor or p53 was knocked-down by transfecting each siRNA into cells.

Results
Extracellular adenosine induces Lu-65 cell apoptosis in a concentration (0.01-10 mM)-dependent manner, and the effect was inhibited by the A3 adenosine receptor inhibitor MRS1191 or by knocking-down A3 adenosine receptor or p53. Like adenosine, the A3 adenosine receptor agonist 2-Cl-IB-MECA also induced Lu-65 cell apoptosis. Adenosine upregulated expression of p53 and Noxa mRNAs and activated caspase-3 and -9, but not caspase-8. Those adenosine effects were still inhibited by knocking- down A3 adenosine receptor or p53.

Conclusion
The results of the present study show that adenosine upregulates p53 expression via A3 adenosine receptor, to promote p53-dependent Noxa gene transcription, causing activation of caspase-9 and the effector caspase-3 to induce Lu-65 cell apoptosis.

Background
Recently, a novel fusion gene resulting from linkage between the kinesin family member 5B gene (KIF5B; 10p11.22) and rearranged during transfection gene (RET; 10q11.21) was identified in non-small cell lung cancer (NSCLC). RET translocation was previously reported in thyroid cancer, as CCD6/RET translocation. However, the correlation between KIF5B/RET fusion gene status and clinicopathologic features of surgically-treated lung cancer has not been well characterized.

Methods
We have investigated KIF5B/RET fusion gene status in 371 surgically treated NSCLC (270 were adenocarcinoma and 101 were squamous cell carcinoma), 60 breast cancers, 11 metastatic lung cancers from colon cancers and thyroid papillary adenocarcinoma case at Nagoya City University Hospital. The fusion gene and CCD6/RET statuses were analyzed by RT-PCR based assay and direct sequencing. We have performed immunohistochemical (IHC) analysis using C-ternimal specific anti RET antibody (EPR2871, 1:250) (Epitomics Inc., Burlingame, CA, USA, n=86) with Dako linker kit using intercalated antibody-enhanced polymer (iAEP) method. Cytoplasm was stained either granular (G1) or diffuse (G2). G2 staining was defined as positive staining.

Results
We detected 3 of 270 cases of KIF5B/RET fusion genes in adenocarcinomas (1.5 %) from the present study; all were mixed subtype adenocarcinomas and three were female and never-smokers. The fusion genes were exclusive with the other mutations, such as EGFR, Kras, Braf, erbB2 mutations, and EML4-ALK fusion. KIF5B/RET fusion was not detected in the cases with squamous cell carcinoma or other types of cancers. Of the 3 cases, 2 cases were KIF5B (exon15): RET (exon12) fusions with papillary dominant and 1 cases were KIF5B (exon22): RET (exon12) fusion with solid dominant adenocarcinoma. Matched normal lung tissues did not show translocation. In the present study, we did not detect CCD6/RET fusion genes. as a driver somatic mutation of lung adenocarcinomas. Although all 3 had positive IHC staining, 35/86 had more than 10% staining and 15/86 had more than 50% staining.

Conclusion
In the present study, we reported KIF5B/RET fusion genes as a possible new driver somatic mutation of lung adenocarcinomas. Cinico-pathological backgrounds of the KIF5B/RET fusion positive patients were similar with that of the EML4/ALK fusion positive patients. The chimeric oncogene might be as a promising molecular target for the personalized diagnosis and treatment of NSCLC. However, the chimeric oncogene might not be determined using IHC analysis.

Background
Small cell lung cancer (SCLC) accounts for 13-15% of new lung cancer cases in worldwide and has the poor therapeutic outcomes with a median survival of just over one year. A CpG island methylate phenotype (CIMP) is well known as a methylator phenotype with characteristic promoter DNA methylation alterations, in colorectal cancers, glioblastoma and breast cancers, although there has been no report about any CIMP of SCLC. We investigated whether DNA methylation profiles can provide useful molecular subtyping of SCLC in terms of etiology and prognosis of SCLC.

Methods
We analyzed 28 fresh frozen samples from pure SCLC patients and 13 noncancerous lung tissues. All patients underwent surgical lung resection at the Cancer Institute Hospital, nine patients among them were treated with chemotherapy before surgery. After genomic DNA was treated with sodium bisulfite, bisulfite-converted genomic DNA was analyzed using Illumina’s Infinium HumanMethylation27 BeadChip. And, total RNA was extracted from twenty-five SCLC tumor samples and mRNA expression of these samples were analyzed by Agilent’s SurePrint G3 Human CGH Microarray. Next, we matched these two data sets by Gene Symbol, and identified fifty-five most differentially methylated CpG sites (corresponding to 46 genes) with a FDR p value cut off of 0.05. Gene ontology analysis was performed using DAVID Bioinformatics Resources.

Results
We selected a total of 1741 most differentially methylated CpG sites (s.d. > 0.20) across the 28 SCLC tumor tissues in each DNA methylation platform, after an elimination of the probes related with the X- and Y- chromosome. Unsupervised hierarchical clustering of methylation data from SCLC samples reveals two major subgroups with different prognosis: the 5 years disease-free interval (DFI) rate of patients in cluster 1 (11.1%) was lower than that of patients in cluster 2 (61.57%) (p = 0.001). By multivariate analysis for DFI, both postoperative chemotherapy and cluster 1 were a significant prognostic factor (p = 0.002 and 0.002; respectively). Next, among 1220 genes with methylation and expression data both available, the CpG sites were ranked on the basis of the spearman’s correlation coefficient between cluster 1 and cluster 2 into an ascending order. Finally, we identified that fifty-five CpG sites were nagetively correlated and found that apoptosis pathway was a most differentially expressed.

Conclusion
By comprehensive DNA methylation profiling, two distinct subgroups with different molecular and clinical phenotype were identified to evoke a CIMP of SCLC. We found some promoter markers in the apoptosis pathway could make a difference between the two groups, and we hope that our data can contribute to provide a useful resource for the construction of therapeutic strategy and the development of a new chemotherapeutic agent.

Background
Genetic variability can influence the pharmacokinetics and pharmacodynamics of paclitaxel and carboplatin in patients with non-small cell lung cancer (NSCLC). Additionally, the prevalence of genetic variations often differs between ethnic groups and may account for observed interethnic variability in drug efficacy. We aimed to undertake a pharmacogenomic investigation to account for patient variability in order to improve dosing and patient selection in NSCLC.

Methods
76 advanced NSCLC patients from Caucasian (n = 50) and Asian (n = 26) ethnicity were prospectively recruited into the study at Concord Repatriation General Hospital from 2007-2011. All patients received paclitaxel (175mg/m[2]) and carboplatin (target AUC 6 mg/mL•min) for an intended 6 cycles. A candidate gene approach was used to select single nucleotide polymorphisms (SNPs) for pharmacogenomic analysis. HPLC with population pharmacokinetic modelling (NONMEM) was used to obtain pharmacokinetic data (n = 61). Toxicities were assessed according to CTCAE v 4.0 and response was measured by CT scans according to RECIST criteria. Blood DNA was genotyped by the Australian Genetics Research Facility using a MassARRAY® iPLEX Gold system on a Sequenom mass spectrometer. SNPs were assessed for linkage disequilibrium, Hardy-Weinberg equilibrium and interethnic differences in allele frequency. Regression analysis was undertaken to associate SNPs with drug pharmacokinetics, toxicities and response.

Results
42 SNPs from 21 genes were selected and genotyped in patients. Regression analysis identified 12 SNPs significantly associated to paclitaxel pharmacokinetics, patient toxicities and response. SNPs rs2306283 and rs11615, in SLCO1B1 and ERCC1 respectively, associated with paclitaxel clearance. Previously published SNPs in CYP2C8, CYP1B1, ABCB1, ABCC10, SLCO2B1, GSTP1, ATP7A, ERCC1 and CCND1 were found to be associated with patient toxicities. Three SNPs, rs2306283, rs1056836 and rs2306168, encoded by SLCOB1, CYP1B1 and SLCO2B1 respectively, had interethnic difference in SNP prevalence and associated to patient outcomes. SNP rs2227291 in ATP7A associated with patient response and to nausea and anaemia.

Conclusion
The study associated various SNPs to drug pharmacokinetics, patient toxicities and response which could be integrated into future personalisation efforts of paclitaxel and carboplatin. Furthermore, SNPs with interethnic differences may provide clues to understand interethnic variability in drug efficacy. Finally, a novel SNP identified in ATP7A associated with patient response and toxicity. The ATP7A gene encodes a protein that has been linked to resistance to platinum based drugs, however, the SNP has unknown functional effects which require further elucidation.

Background
Malignant pleural mesothelioma (MPM) is a highly aggressive tumor caused by asbestos exposure after a long latency. The neurofibromatosis type 2 (NF2) tumor suppressor gene is mutated in around 50% of MPM cases, which encodes Merlin that regulates the Hippo tumor-suppressive signaling pathway. We previously reported occasional genetic alteration in the LATS2 and SAV1 genes in MMs, which encode components of the Hippo pathway. The LATS2 inactivation was shown to lead to constitutive activation of YAP, a prooncogenic protein and transcriptional coactivator, which enhances multiple cell cycle regulation genes including cyclin D1 (CCDN1). To further delineate the exact inactivation mechanism of this pathway in MMs, we analyzed MM cell lines for other multiple components that regulate activation or inactivation of this pathway.

Methods
We studied several new components that have been identified to be involved in the Hippo signaling pathway including a LIM protein Ajuba. Expression analyses such as conventional western blot and real time RT-PCR were performed. Using MPM cell lines and their transplanted mouse models, biological assays were conducted to study the effects of cell proliferation, motility and invasion after the induction of overexpression or knockdown of candidate genes. Immunohistochemical analysis with primary MPM tissues and xenografts was also carried out.

Results
We found that the expression of Ajuba was significantly down-regulated in 6 of 20 MM cell lines compared to an immortalized normal mesothelial cell line, MeT-5A. Interestingly, the 6 cell lines with low Ajuba expression showed decreased phosphorylation (activation) levels of YAP. We infected the MM cells with Ajuba-expressing lentivirus and found that exogenous Ajuba increased phosphorylation (inactivation) of YAP and inhibited cell proliferation. Dual-luciferase reporter assays demonstrated that Ajuba suppressed the promoter activities of YAP-transcriptional targets including CCND1. Knockdown of LATS2 effectively increased these promoter activities, suggesting the mediation of LATS2 for Ajuba to inhibit this cascade. Immunohistochemical analysis also showed frequent downregulation of Ajuba in primary MMs.

Conclusion
Inactivation of Hippo signaling is a key mechanism for MPM cell proliferation and progression. Our findings suggest that Ajuba is also one of the target components for Hippo signaling inactivation; Ajuba negatively regulates YAP in the presence of LATS2, and thus the downregulation of Ajuba serves to enhance constitutive activation of YAP in MM cells. Our study also indicates that a strategy to normalize this signaling cascade may be the rationale for developing a new target therapy against MPMs.

Background
Most of small cell lung cancer (SCLC) cases are diagnosed after metastatic spread of the diseases, and only 5% of SCLC patients survive beyond 5 years after diagnosis. Therefore, for the improvement of patients’ outcome in this disease, it is necessary to identify druggable targets that are activated by genetic alterations in SCLC cells.

Methods
To obtain a landscape of gross genetic alterations in SCLC, genome-wide copy number analysis and whole-transcriptome sequencing were performed in 58 and 42 SCLCs, respectively.

Results
Focal amplification of known oncogene loci, MYCL1 (1p34.2), MYCN (2p24.3), and MYC (8q24.21), was frequently and mutually exclusively detected. MYCL1 and MYC were co-amplified with other regions on either the same or the different chromosome in several cases. In addition, the 9p24.1 region was identified as being amplified in SCLCs without amplification of MYC family oncogenes. Notably, expression of the KIAA1432 gene in this region was significantly higher in KIAA1432 amplified cells than in non-amplified cells, and its mRNA expression showed strong correlations with the copy numbers. Thus, KIAA1432 is a novel gene activated by amplification in SCLCs. By whole-transcriptome sequencing, a total of 60 fusion transcripts, transcribed from 95 different genes, were identified as being expressed in SCLC cells. However, no in-frame fusion transcripts were recurrently detected in 2 SCLCs, and genes in the amplified regions, such as PVT1 neighboring MYC and RLF in MYCL1 amplicons, were recurrently fused with genes in the same amplicons or with those in different amplicons on either the same or different chromosome.

Conclusion
Amplification and fusion of several genes on chromosomes 1 and 8 were likely to occur simultaneously but not sequentially through chromothripsis in the development of SCLC. Amplification rather than fusion of genes was indicated to play an important role in its development.

Background
The epidermal growth factor receptor (EGFR) signaling pathway is involved in numerous biological processes including proliferation and apoptosis, migration/invasion, and angiogenesis, and has emerged as one of the most important and frequently deregulated pathways in NSCLC. The discovery of oncogenic, activating mutations in the tyrosine kinase domain of EGFR and DNA amplification of EGFR have led to the development of multiple targeted therapeutics against this pathway. While effective at prolonging survival, these targeted therapies are only applicable to a subset of patients (~15-20%) that harbour these alterations and resistance to treatment ultimately develops. As multiple genomic and epigenomic mechanisms can disrupt genes, a comprehensive understanding of the genetic alterations affecting genes within this pathway is required. An integrative, multi-dimensional genomics approach can detect genes disrupted by multiple mechanisms which may otherwise be overlooked if only a single genomic dimension were assessed, improving the ability to identify causal genetic events and decipher downstream consequences.

Methods
A multi-dimensional integrative analysis of copy number, DNA methylation and gene expression profiles on 77 adenocarcinomas and matched non-malignant tissue, was performed to investigate the complement of genetic alterations affecting the EGFR pathway. Novel candidate genes were validated in external datasets and immunohistochemical analysis of a tissue microarray was used to verify disruption at the protein level and to correlate expression with clinical features. The tumor suppressive effects of SIRPA were assessed by stable knockdown and in vitro assays on a panel of lung cancer cell lines. The effect of SIRPA downregulation on TKI sensitivity was assessed by dose response assays.

Results
Of the 35 genes examined, 11 were aberrantly expressed in over 50% of tumors, with 6 (RRAS, SIRPA, PIK3R1, TGFA, ERBB2 and EGFR) ranking in the 95th percentile of altered genes. Of these genes, all but SIRPA are known to be frequently disrupted in NSCLC and play a role in tumorigenesis. SIRPA is a transmembrane protein that negatively regulates receptor tyrosine kinsase activity and is frequently downregulated at both the mRNA and protein level in NSCLC tumors and cell lines. Underexpression of SIRPA is associated with EGFR mutations and is more prominent in adenocarcinoma than squamous cell carcinoma. Downregulation of SIRPA enhanced colony formation and wound healing but impaired viability and suppressed proliferation. Interestingly, SIRPA knockdown induced a senescent phenotype through the accumulation of p27 and Rb in its unphosphorylated state thereby blocking progression of the cell cycle. These results suggest senescence induced by SIRPA downregulation is a tumor suppressive mechanism that must be overcome to develop tumors.

Conclusion
Our integrative analysis of the EGFR pathway revealed SIRPA as one of the most frequently deregulated genes within the pathway. SIRPA functions as a tumor suppressor gene, controlling a number of biological functions through the inhibition of singaling pathways downstream of EGFR. To our knowledge, this is the first study to report a role for SIRPA in NSCLC tumorigenesis.

Background
MET is a tyrosine kinase receptor expressed on the cell surface of numerous cell types, including normal and malignant epithelial cells. MET is expressed on a proportion of non-small cell lung cancer (NSCLC) cases of adenocarcinoma and squamous-cell differentiation. In a phase II clinical trial evaluating the anti-MET monovalent antibody onartuzumab, patients with advanced MET-positive NSCLC – as determined by immunohistochemistry (IHC) – derived clinical benefit from onartuzumab and erlotinib. The prevalence of MET expression in two separate sample cohorts of NSCLC – vendor-procured and phase II clinical trial samples – is compared and correlated with MET gene copy number.

Methods
Formalin-fixed, paraffin-embedded NSCLC tissues were obtained from commercial sources or from the phase II clinical trial OAM4558g. Tumor histology was determined by morphology; IHC for TTF1 and p63, respectively; or both. MET IHC was performed with the SP44 rabbit monoclonal antibody (CONFIRM anti-Total c-MET) on a Ventana Benchmark XT platform. Patients were selected based on expression of MET by IHC as defined by moderate or strong staining in at least 50% of tumor cells (clinical score 2+/3+); MET gene copy number was determined by fluorescence in-situ hybridization (FISH) or by chromogenic in-situ hybridization (CISH) (INFORM® MET DNA probe, Ventana Medical Systems).

Results
Adenocarcinomas and squamous-cell carcinomas accounted for 46% (n=95) and 54% (n=110) of the vendor-procured samples and for 59% (n=75) and 28% (n=36) of clinical trial samples, respectively. In total, 16% (n=33) of vendor-procured and 52% of clinical trial samples were scored as MET-positive. The proportions of MET-positive cases among adenocarcinomas and squamous-cell carcinomas were 24% and 9% for vendor samples and 61% and 28% for clinical trial samples. FISH analysis of the OAM4558g clinical trial samples showed MET gene amplification in 8 of 96 cases (8%) with a high positive correlation to MET-positive IHC status; 6 of 8 gene-amplified cases were MET-positive by IHC (p<0.0001). Evaluation of MET copy number by CISH is ongoing and will be reported.

Sample Source	Clinical Trial OAM4558g		Vendor-Procured
Histology	Adenocarcinoma	Squamous-cell Carcinoma	Adenocarcinoma	Squamous-cell Carcinoma
Cases (n)	75	36	95	110
MET-positive (%)	61	28	24	9

Conclusion
These results demonstrate preferential MET expression for adenocarcinoma compared with the squamous-cell carcinoma subtype of NSCLC. The lower proportion of MET-positive cases in the vendor samples suggests that tissue quality is crucial for adequate MET assessment by IHC.

Background
Non-small-cell lung carcinoma (NSCLC) accounts for 85% of all malignant lung tumors. Our group previously identified Tripartite Motif-Containing 14 (TRIM14) as a component of a prognostic multigene expression signature for NSCLC patients. TRIM14 belongs to a conserved family of Tripartite Motif-encoding genes involved in a broad range of biological processes, such as transcriptional regulation, cell proliferation, and apoptosis. However, TRIM14 itself is poorly understood. Here we investigate the functional and prognostic role of TRIM14 in NSCLC.

Methods
Cox proportional hazards regression analysis was done on published mRNA expression datasets of primary NSCLCs to identify whether TRIM14 expression is associated with patient survival. The expression of TRIM14 was modified in human NSCLC cell lines (NCI-H1395, NCI-H1650, NCI-H520 and NCI-H157) by lentivirus-mediated overexpression and short-hairpin RNA (shRNA)-mediated silencing. Effects were assessed by examining in vitro cell proliferation and anchorage-independent growth, and in vivo tumorigenicity in mice.

Results
Univariate analyses demonstrated that TRIM14 expression is significantly associated with patient survival, with loss of expression correlated with poorer prognosis in resectable early stage NSCLC patients. In vitro studies showed that TRIM14 overexpression in H1395 cells suppressed proliferation and anchorage-independent growth, whereas shRNA knockdown of TRIM14 expression in H1650, H520, and H157 cells promoted cell proliferation and colony formation. In vivo studies demonstrated that suppressing TRIM14 expression in H1650 and H157 cells significantly promotes tumor growth in immunodeficient mice.

Conclusion
We show that TRIM14 may function as a tumor suppressor gene in NSCLC affecting cell proliferation and anchorage-independent growth in vitro and tumor growth in mice. We provide evidence that prognostic genes identified in microarray based gene expression analyses may have a strong biological rationale.

Background
Molecular cross-talk between EGFR and other signaling pathways creates alternative means of tumor cell proliferation and promotes resistance to single-agent erlotinib therapy in NSCLC driven by EGFR mutations. ROR1 knockdown inhibited the growth of NCI-H1975 cells (harboring EGFR L858R and T790M mutations). A pro-survival function for ROR1/MEK/ERK signaling in cooperation with AKT has been demonstrated. We have assessed ROR1 expression in 45 patients from the EURTAC trial (clinicaltrials.gov NCT00446225), 27 of whom harbored pretreatment concomitant EGFR T790M mutations, and correlated results with outcome.

Methods
ROR1 mRNA expression was examined by quantitative RT-PCR and categorized by terciles; patients were classified as having low/intermediate or high ROR1 expression. The T790M mutation was determined by Taqman with a PNA to inhibit amplification of the wild-type (wt) allele. Tumor samples were run in octuplicates; this method can detect 1 mutated allele among 10,000 wt alleles.

Results
Median age 65; 68.9% female; 57.8% never-smokers; 95.6% ECOG PS <2; 91.1% adenocarcinoma; 68.9% exon 19 deletion. No differences in baseline characteristics were observed according to ROR1 expression levels. 24 patients (53.3%) were treated with erlotinib and 21 (46.7%) with chemotherapy. 10 (41.7%) erlotinib-treated patients and 6 (28.6%) chemotherapy-treated patients had high ROR1 mRNA levels. Among erlotinib-treated patients, response rate (RR) was 40% for the 10 patients with high ROR1 levels vs 71.4% for the 14 with low/intermediate levels (P=0.058). Among chemotherapy-treated patients, RR for the 15 patients with low/intermediate ROR1 levels was 6.7%; the 6 patients with high ROR1 levels did not respond. Progression-free survival (PFS) was 11.8 months (m) for erlotinib-treated patients with low/intermediate ROR1 levels vs 5.8 m for those with high levels. PFS for chemotherapy-treated patients was 5.6 and 9 m, respectively (P=0.0165). 15 erlotinib-treated patients harbored concomitant T790M mutations; for these patients, PFS was10.8 m for those with low/intermediate ROR1 levels vs 2.7 m for those with high levels (P=0.0138).

Conclusion
ROR1 expression has a differential effect on outcome to erlotinib and chemotherapy in EGFR-mutant NSCLC patients. High ROR1 expression significantly limits PFS in erlotinib-treated patients with T790M mutations and ROR1-directed therapies can enhance the efficacy of treatment. In contrast, high ROR1 expression confers longer PFS to chemotherapy in the same group of patients. The role of chemotherapy and erlotinib in EGFR-mutant NSCLC patients with high ROR1 expression warrants further investigation.

Background
Cell transformation and tumour progression are associated with severe alterations of the epigenetic control of gene expression. Although the abnormal repression of tumour suppressor genes has been thoroughly investigated, the concept of tissue-specific gene aberrant reactivations in cancer is only starting to emerge. The extent of this process has not been reported yet, mainly because of the difficulties in detecting these expressions by the currently used transcriptomic analyses

Methods
We have developed a simple approach, exploiting genome wide expression data, which has enabled us to demonstrate that these "off-context" gene activations occur in any cancer, providing a universal source of cancer biomarkers. By applying this analysis to our series of lung cancer patients (n=297) with the corresponding precise clinical annotations and 5 to 10 years patient follow-up, we found that hundreds of germline-specific genes are ectopically expressed in the tumours and that a subset of 26 genes are specifically activated in a subgroup of highly aggressive metastatic-prone tumours, even at an early stage of the disease. This approach has enabled us to define and isolate a homogeneous group of very aggressive tumours called P3 with at least 3 genes ectopically expressed, independently of other prognosis parameters (histo-pathological or stage) (Rousseaux et al. Sci Transl Med 2013, 5 (186): 186ra66).

Results
Based on these data we setup a PCR based test to directly rank lung tumours by detection of the prognosis associated gene activations. This test was first validated on a series of 60 tumour samples from the same cohort (also analyzed by an affymetrix transcriptomic approach), which revealed the remarkable robustness of the PCR approach and showed a higher sensitivity of the PCR-based detections in tumour ranking. We then extended these PCR-based analyses to include additional patients with precise clinical and pathological annotations and 10 years follow-up but for whom transcriptomic data were not available. Here again we could show that our test successfully ranked the patients into groups with significantly different survival probabilities. we show here the intermediate analysis on a first group of 88 patients without lymph node metastasis treated by surgeryFigure 1

Conclusion
In conclusion we are ready to launch a prospective study based on this PCR test to evaluate its ability to predict tailored follow up of patients after surgery and also tumour sensitivity to design specific targeted therapies including immunotherapy since this ectopic activation may lead to very innovative treatment .

Background
Quantifying and comparing the degree of methylation in the promoter region of APC gene and coding region of FHIT gene in patients (p) with non small cell lung cancer (NSCLC) in different samples. Correlate these methylation patterns with clinical variables.

Methods
DNA was extracted from peripheral blood simples (B), sputum (S) and bronchial aspirate (BA), obtained from patients with NSCLC prior to the completion of surgery, as well as resected lung tumor tissue (TT) and normal lung tissue (TN). Methylation patterns were analyzed by bisulfito conversion and subsequent pyrosequencing (QIagen PyroMark System).We analyzed 5 CpG islands in the promoter region of the APC gene and 5 CpG islands in the coding region of the FHIT gene.

Results
We analyzed 20p, with a median age of 64 years (range 48-70), 16 men and 4 women. Smoking status: 2p never smokers, 11p former smokers, 7p current smokers. Histology: 10p adenocarcinoma, 9p squamous, 1p large cell. Stage: I 8p, II 8p, III 4p. No statistically significant differences were observed between the samples studied to any of the islands analyzed. The degree of methylation in TT CpG1 was higher in smokers and former smokers <5 years, compared to never smokers and ex-smokers >5 years, mean 4 (0-10) vs 0, p=0.022. There was no other difference when analyzing the degree of methylation as a function of the variables age, sex, smoking status, cumulative tobacco consumption, histological type and clinical stage. Respect FHIT gene, no statistically significant differences were found between the tissues studied respect to any of the CpG islands analyzed and like wise, no differences were observed when analyzed for degree of methylation depending on the clinical variables studied.

Conclusion
The area of ​​the APC gen promoter and FHIT coding region analyzed showed a low degree of methylation, with no significant difference observed between the samples studied. The degree of methylation in the TT CpG1 island of the APC gene was higher in current smokers and former smokers <5 years. However, these findings have to be confirmed in a larger sample. *This study was funded by the Carlos III Health Institute (PS09/00308), Spain.

Background
Lung cancer is one of the most common causes of cancer-related deaths worldwide. Effective early diagnosis and targeted therapies for lung cancer to reduce mortality and incidence would benefit from in-depth study on molecular mechanism of lung carcinogenesis, but these are largely unknown. In our previous study, we reported a novel hyper-methylated gene, CDO1, encoding cysteine dioxygenase 1 that involved in the conversion of cysteine to cysteine sulfinate, and plays a key role in taurine biosynthesis. In this study, we evaluated the functional characteristics of CDO1 in Lung squamous cell carcinoma (LSCC).

Methods
We analyzed expression levels of CDO1 mRNA and protein in tumor and normal tissue pairs from 12 LSCC patients were determined by RT-PCR and Western blot analysis. In order to determine whether high levels of CDO1 expression contributed to the LSCC cell proliferation, migration, invasion and colony formation, we employed a CDO1-expressing vector pEGFP_C3_CDO1 to transfect CDO1 into LSCC cell lines (HCC-95, HCC-1588). Furthermore, we performed gene expression profile analysis using human whole genome oligonucleotide chip (Agilent).

Results
Downregulation of CDO1 mRNA level was observed in LSCC cell lines and tumors derived from patient tissues. In cell proliferation assay, the number of HCC-95 and HCC-1588 cells transfected with pEGFP_C3_CDO1 decreased to 58-70% compared with pEGFP_C3 (p<0.05). Moreover, the forces expression of CDO1 in two different types of LSCC cell lines significantly decreased the cell migration, invasion, and colony formation ability (p<0.05). Furthermore, we showed that overexpression of CDO1 make change of several signal molecules in lung cancer cells using microarray analysis.

Conclusion
In this study, we found that CDO1 expression was regulated by DNA methylation, and this epigenetic regulated CDO1 might be a novel tumor suppressor gene and potentially valuable biomarker for lung squamous cell carcinoma.

Background
ATF2 is a member of the basic helix-loop-helix (b-ZIP) transcription factor family playing important roles in Stress and DNA damage. Several studies reported correlation between ATF2/JNK-mediated activation and resistance to damaging agents. Celastrol is an ATF2 inhibitor extracted from a Chinese plant known as “Tunder of God Vine”. Celastrol is used in traditional Chinese medicine for anti-inflammatory properties. In addition, Celastrol is a triterpene with promising anticancer activity in several cancer models both in vivo and in vitro. The purpose of the present study was to investigate whether ATF2 might play a role in inducing drug resistance in Non-Small Cell Lung Cancer (NSCLC).

Methods
In NSCLC cell lines ATF2 expression levels were evaluated by quantitative PCR (qPCR) and Western Blot (WB) and correlated to cisplatin (CDDP) resistance. Celastrol mediated ATF2/cJUN activity was assessed by Luminometry, qPCR and western blotting (WB). Furthermore, matched tumors and corresponding normal tissues of 88 surgically resected NSCLC specimens, were collected and both qPCR and immunohistochemical analyses for ATF2 were performed.

Results
NSCLC cell lines CDDP-resistant (H522 and H1395) expressed high levels of ATF2 protein. Moreover, CDDP treatment increased ATF2 phosphorylation levels leading to an enrichment within the nuclear cell compartment. In our study, Celastrol reduced ATF2 activity by decreasing the production of ATF2 mRNA and blocking the CDDP-mediated phosphorylation/mRNA expression of cJUN, a main ATF2 partner. Furthermore, ATF2/cJUN functional inhibition mediated by Celastrol restored the response to CDDP in resistant lung cell Lines. ATF2, at both protein and mRNA level, was significantly up-modulated in NSCLC tumor samples compared to the paired normal lung tissue (mRNA: p<<0.01, mean Log2(FC)=+4.7). Moreover, high expression of ATF2 mRNA was correlated with the smoking status of the patients. Relevantly, smoker or former smoker patients expressed significantly high ATF2 mRNA levels compared to non-smokers (p=0.02 and p=0.04, respectively).

Conclusion
This study shows that in NSCLC cell lines a correlation between ATF2 protein expression and CDDP resistance occurs. Furthermore, our results suggest a potential increase of CDDP sensitivity, following the Celastrol-mediated ATF2/cJUN inhibition. For the first time it has been shown in NSCLC an up-regulation of ATF2 mRNA/protein levels compared to normal tissues and consistent with that detected in CDDP resistant cell lines. This data suggest a possible involvement of ATF2 in NSCLC CDDP-resistance.

Background
Majority of investigation is focused on lung tumorigenesis in smokers, little in nonsmokers. Previously, Benzo[a]pyrene (B[a]P)-DNA adducts levels in nonsmoking female patients were higher than in nonsmoking male patients. However, p53 mutation rate in female patients did not differ from male patients. Moreover, HPV16/18 infection rate in female patients was much higher than in male patients. We therefore hypothesized that HPV infection could synergistically increased chromosomal instability (CIN) induced by B[a]P-DNA adducts and might contribute to lung tumorigenesis.

Methods
Herein, three HPV-positive and five HPV-negative lung cancer cells were enrolled to test the hypothesis. FISH analysis was used to determine the micronuclei formation when these cells were treated with various concentrations of B[a]P for different time-intervals.

Results
The efficacy of micronuclei and DNA adduct formation induced by B[a]P in HPV-positive cells were significantly higher than HPV-negative cells. Mechanistically, the micronuclei and DNA adduct formation are dependent on HPV E6 oncoprotein expression. We next questioned whether B[a]P-induced CIN enhanced by E6 could be through altering DNA repair gene expressions. The promoter hypermethylation and mRNA expression of six DNA repair genes including hMLH1, hMSH2, BRCA1, and BRCA2, and XRCC3, and XRCC5 were evaluated by MSP and real-time RT-PCR, and data indicated that XRCC3 and XRCC5 expressions in E6-positive cells were markedly lower than in E6-negative cells and the reduction of both genes was caused by promoter hypermethylation.

Conclusion
Collectively, the promoter hypermethylation of XRCC3 and XRCC5 induced by E6 may increase B[a]P-induced CIN and contribute to lung tumorigenesis in nonsmokers.

Background
Background: Antofagasta region in northern Chile shows the highest lung cancer (LC) mortality rate in the country. This population was exposed to arsenic (As) in drinking water of concentrations as high as 870 mg/L. Between 2003 and 2007, Antofagasta region showed a mortality rate of 30.8/100,000. It has been suggested that variation in As susceptibility among individuals might be partly due to differences in As biotransformation in addition to other genomic factors associated to the carcinogenic process. Objective: To determine the association of genomic variations (SNPs, and DNA copy-number alterations (CNAs) and susceptibility to LC, in blood and induced sputum samples collected from a LC risk sample population.

Methods
Materials and Methods: The 400 high-risk volunteers from Antofagasta and Metropolitan region was split in 2 groups, controls and cases, according to the result of early detection assays including automatic quantitative cytometry (AQC), DR70 and autofluorescence bronchoscopy (AFB). Copy number variants (CNVs) and single nucleotide polymorphisms (SNPs) were determined in genomic DNA of peripheral blood using TaqMan QPCR assays. Odd ratios (OR) were estimated by conditional likelihood and significance as exact mid-p values. Whole genome CGH profiles were generated with HEEBO 70-mer oligonucleotide microarrays.

Results
Results: Among 8 polymorphisms assayed (5 SNPs and 3 CNVs), CYP1A1 SNP rs1048943 was associated both to pre-neoplastic lesions (PNL) and cancer histopathology (OR 2.66, p=0.02). The CYP1B1 SNP rs1056836 was associated to LC, although no significantly. Some differences in the AS3MT SNP rs11191439 were observed between individuals from Antofagasta and Santiago. Additionally, the genomic profiles of sputum DNA and circulating cell-free DNA (cfDNA) from serum samples showed discrete differences among LC cases, pre-neoplastic lesions (PNL) and healthy controls. Chromosome 7p arm showed a significant gain in sputum DNA from cancer patients.

Conclusion
Conclusions: Conclusions: Our results suggest that SNPs associated to CYP450, like to CYP1A1 and 1B1, might be related to LC etiology. Additionally, genomic profiles from sputum DNA and cfDNA might be useful to detect PNL as complementary tools. Finally, this panel of genomic biomarkers in addition to DR70, AQC and AFB could be helpful to identify individuals susceptible to develop LC, and as complementary tools for LC early detection. These findings might be relevant in prevention and early detection of LC. Supported by INNOVA CORFO Chile: Grants 07CN13B48 and 11IDL2-10634.

Background
Plasminogen activator inhibitor type 1 (PAI-1) plays an important role in tumor growth and metastasis formation, directly via specific urokinase complexing or indirectly due to its affinity to vitronectin.

Methods
The aim of this study was to analyze the impact of the mutant forms of PAI-1: very long half-life (VLHL PAI-1) or devoid of affinity towards vitronectin (Vn neg PAI-1) and wild form (wPAI -1) on proliferation of lung cancer (A549 and H1299 ) and prostate cancer (LNCaP and DU145) cells characterized by different proteinase (urokinase) production.

Results
The dose- and time-dependent inhibition of cell proliferation in the presence of VLHL PAI-1 was evident in A549 and LNCaP cultures. In H1299 cells inhibitory effect was only dose-depended (p<0.001), while in DU145 only 100 mg/ml of VLHL PAI-1 in 72 hrs cultures suppresed prostate cancer cells proliferative activity (p <0.001). No significant effect of Vn neg PAI-1 on the proliferation of A549 and H1299 was observed while in prostate cancer lines (DU145, LNCaP) only the inhibitory effect of the highest Vn neg PAI-1 concentration (100μg/ml) was evident (p <0.001) but not time-dependent. wPAI-1 did’t affect A549 and LNCaP proliferation while in highest concentration it had the stimulating effect on H1299 and Du145 (24,48 hrs cultures).

Conclusion
PAI-1 is a negative regulator of cancer cells proliferation due to its anti-proteinase activity. Its biological effect on lung cancer cells is time and dose-dependent.

Background
Alpha-catulin is a cytoskeletal linker protein. Integrin-linked kinase (ILK) is a serine/threonine protein kinase implicated in cancer cell proliferation, anti-apoptosis, invasion and angiogenesis. Our previous study found that α-catulin is upregulated in lung cancer cells and directly interacted with ILK. Here, we further investigated the effect of α-catulin/ILK interaction in lung cancer metastasis and evaluated whether this interacting axis could be a potential therapeutic target.

Methods
Knockdown and overexpression of α-catulin and ILK genes were performed in non-small cell lung cancer (NSCLC) cell lines. Cell invasion and migration assays were done in 24-well Transwell polycarbonate filters coated with or without Matrigel. Spontaneous and experimental metastasis assays were performed in xenograft mice models to determine the effects of α-catulin on lung cancer cell metastasis in vivo. The public NSCLC cohort datasets were used for validation of the expression of α-catulin/ILK associated with patient survival.

Results
In the present study, we demonstrated that α-catulin promoted the migration and invasion of NSCLC cells through the ILK/NF-κB/integrin network. α-Catulin directly interacted with ILK, which in turn activated the ILK/Akt/NF-κB signaling pathway. This led to increased expression of the NF-κB downstream genes fibronectin and integrin α~v~β~3~, which sequentially activated NF-κB signaling and resulted in cancer cell migration, invasion and metastasis. The ILK plus CTNNAL1 two-gene signature was even more strongly associated with clinical outcomes of NSCLC patients.

Conclusion
Our data suggest that the alpha-catulin/ILK signaling axis is strongly associated with lung cancer cell migration and invasion and correlated with clinical outcome of NSCLC. Thus, this novel signaling axis could be a potential therapeutic target for treating NSCLC metastasis.

Background
Tyrosine Kinase (TK) domains are central regulators of signaling pathways that control differentiation, transcription, cell cycle progression, apoptosis, motility and invasion. EGFR TK mutations represent bona fide somatic mutations in NSCLC. Patients with specific mutations respond better to targeted therapy.Mutation analysis is known to be evaluated by several methods. Real time PCR using allele specific TaqMan primer probes is a simple, robust, highly sensitive, and selective method that is compatible with standard processes used for known gene expression analysis. The main objective of our study was to detect EGFR mutations in NSCLC patients registered in Tata Memorial Hospital (TMH) by using rapid and sensitive technique of RQ-PCR with In-house TaqMan primer probes. There is limited data regarding EGFR mutation in Indian population with variable mutation rate reported. This is an attempt to get actual mutation rate in our population, correlate the frequency of EGFR mutation and their subtypes and analyze across different variables of age, gender, habits and histology with different ethnicity groups.

Methods
1018 patients of NSCLC were referred for EFGR testing as a routine service over a 1.5-year period. Extracted DNA from FFPE blocks was amplified for the exons 18, 19, 20 and 21 using the specific TaqMan primer probes with the End point genotyping method on LC-480 II platform. Chi-square test was performed to reveal any significant correlation between the mutation status and age, gender and habits of the patient and tumor histology.

Results
Overall Mutation Rate (MR) was 25.0% with a higher mutation rate in females than males (34 vs.21) with a p value of 0.002. Within the age group, MR was high in > 60 yrs group as compared to <60 years (47.6 % vs. 31%). Total population of smoker vs. never-smoker was 37.5% vs. 58.6 %. (p value <0.001). MR was significantly higher in Never Smoker (NS) as compared to Smoker (S) (31.5 % vs.16 %), MR in Adenocarcinoma (ADC) was significantly higher than squamous (27.7 %vs. 5.6%) with a p value of <0.001. MR is higher in ADC NS female as compared to ADC NS male (37.2 %vs.29 %). MR of Exon 19 , 21, 18, & 20 was 53%, 38 %, 5.8 % & 2.4 % respectively. In the cohort of ADC NS gender, Exon 19 positivity was predominantly higher in male than female (61.8% vs.34 %) , whereas Exon 21 was marginally higher in ADC NS female than male (39 % vs 34 %). Exon 20 expressed 2.4%.

Conclusion
There is a heterogenous EGFR MR in diferent geographical population. We are in close association with East Asian countries than Western countries. In our data though the MR is high in NS, 16 % of EGFR MR in smokers is also startling reality. Another possibility of variation in EGFR mutation rate could be due to application of different technologies. Each technique differs in their sensitivity and specifictiy. EGFR tyrosine kinase domain define a new molecular marker for lung carcinoma.

Background
The spectrum and frequency of oncogenes in squamous cell lung cancers (SQCLCs) is actively being defined. Amplification of fibroblast growth factor receptor 1 (FGFR1) is the most common targetable oncogenic driver in SQCLCs, occurring in ~20%. Clinical trials of FGFR1 inhibitors for advanced SQCLCs are ongoing. The frequency, clinicopathologic features, and prognosis of FGFR1 amplification in early-stage SQCLCs have been reported but with discrepant results.

Methods
A cohort of histopathologically-defined and clinically-annotated resected SQCLCs was tested for FGFR1 amplification by FISH (Zytovision Dual Color Probe). Amplification was defined by FGFR1 copy number ≥2.2x CEP8 control copy number and was assessed by two evaluators (MW, LW) who were blinded to clinical results. Disease-free survival (DFS) defined as date of surgical resection until disease recurrent, relapse, or death, which ever occured first. DFS was estimated using Kaplan-Meier method. The association between FGFR1 status and clinical features (unpaired T-test, Fisher’s exact, Chi-square tests) and DFS (log-rank test for unadjusted analysis; Cox proportional hazards regression for multivariate analysis) were assessed.

Results
63 resected SQCLCs were evaluated. FGFR1 amplification was detected in 16 (24%). 56% were stage I, 24% were stage II, and 20% were stage IIIA. There was no association between FGFR1 amplification and age (p=0.86), sex (p=0.80), smoking status (p=0.37), or stage of disease (p=0.16). Median DFS was significantly longer in FGFR1-amplified cases compared to non-amplified cases: not reached vs 2.3 yrs (95% CI 1.1-3.4 yrs), p=0.02, with a corresponding unadjusted hazard ratio of 0.41 (95%CI: 0.19-0.88). Adjusted for sex and stage, multivariate analysis found FGFR1 amplification significantly associated with improved DFS (HR 0.31, 95%CI 0.1-0.89, p=0.03). Figure 1

Conclusion
FGFR1 amplification is associated with improved prognosis in this cohort of resected SQCLCs. The distinctive natural history substantiates FGFR1amplified SQCLCs as a unique, oncogene-defined subgroup. There was no association between FGFR1 status and sex, age, smoking status, or stage. FGFR1 amplification is common in SQCLCs.

Background
Molecular therapy targeted on tumour driving mutations should improve quality of life, PFS and overal survival in NSCLC patients. While EGFR mutations are widely accepted as targets for gefitinib and erlotinib in the first line treatment of advanced NSCLC, recently, translocation of EML4-ALK as well as amplification of ROS1, are used as guides for indication of crizotinib. Some other genetic changes, such as EGFR amplification, c-Met amplification, PIK3CA mutations and KRAS mutations are expected to serve as prognostic or predictive factors of targeted therapy currently or in near future.

Methods
We analyzed molecular predictors which were routinely tested at our department starting in 2004. We focus on EGFR mutations, EGFR gene amplifications, KRAS mutations, EML4-ALK translocations, c-Met amplifications and PIK3CA mutations. We analyzed the frequency of genetic changes and , where applicable, their impact on PFS and OS. We have also analysed comparability of PCR based detection (multiplex ligation-dependent probe amplification, MLPA) an FISH for EGFR gene amplifications.

Results
In the group of 890 NSCLC patients we found EGFR mutations on exon 19 in 62 cases, EGFR mutations on exon 21 in 24 patients. KRAS mutations were found in 141/819. Translocations of EML4-ALK were found in 7/203, c-Met amplifications in 5/104, EGFR amplifications in 11/34 and PIK3CA mutations in 8/220 patients. We confirmed a high correlation between FISH and MLPA-based testing of EGFR amplifications, as well as agreement of PFS and OS of both groups. However, EGFR gene amplifications were not prognostic, nor predictive marker of TKI therapeutic efficacy. Positive predictors of such treatment were in accordance with the literature and our expectiations, EGFR mutations for the treatment with gefitinib and erlotinib, and EML4-ALK translocations for crizotinib. We identified KRAS G12C mutation as the only negative predictor of EGFR inhibitor treatment. One hundred patients were defined as triple negative tumors (EGFR, EML4-ALK, KRAS negative). Some of our patiens had some combinations of genetic changes, such as concurrent EGFR and KRAS mutations, mutations and amplifications of EGFR gene and, in one patient, mutations and amplifications of EGFR, EML4-ALK translocation and c-Met amplification.

Conclusion
Precious morphologic and genetic investigations of NSCLC tumour tissue should serve as a guide for targeted therapy. However, in our population of non-squamous tumour lesions, we found predominantly a triple negative tumour type which should continue to be treated by first line chemotherapy.

Background
Lung cancer is the leading cause of cancer-related mortality in the world. These patients usually present at an advanced stage where treatment is mostly palliative. Hence, there is an urgent need to investigate the various pre-neoplastic pathways to identify suitable therapeutic targets to decrease lung cancer mortality. The expression patterns of mucins are drastically altered during lung cancer development, and these alterations facilitate lung cancer cell proliferation and metastasis. MUC16 mucin and Testis specific Y-like protein TSPYL5 are two proteins that are overexpressed in lung cancer and appear to promote growth of lung cancer cells. In addition overexpression of TSPYL5 facilitates chemoresistance and regulates aromatase (CYP19A1) enzyme expression. This study was conducted to assess the interplay between these two markers and evaluate their effect of cisplatin induced cytotoxicity in lung cancer cells.

Methods
Expression of MUC16 in normal lung as well as lung carcinoma tissues in a commercially available tissue array was assessed by immunohistochemical (IHC) analysis using MUC16 antibody (DAKO Company-M11 clone). MUC16 levels were semi-quantified using a composite score based on the intensity and extent of staining. Endogenously expressed MUC16 and TSPYL5 were stably knocked down using a MUC16 shRNA construct (pSUPER-Retro-MUC16-sh) and TSPYL5 ShRNA construct in H292 and H827 lung cancer cells by stable transfection method for investigating the oncogenic functions. Lung cancer cells were treated with cisplatin to examine the role of MUC16 and TSPYL5 on chemoresistance and survival in lung cancer cells.

Results
MUC16 is highly expressed in lung carcinoma and is not expressed in non-neoplastic lung tissues. MUC16 composite IHC scores increased progressively from stage I to stage III NSCLC. Knockdown of MUC16 led to decreased proliferation (due to G1 accumulation), invasion and motility in H292 lung cancer cells. TSPYL5 was significantly downregulated in MUC16 knockdown cells. TSYPL5 knockdown in H292 lung cancer cells resulted in decreased expression of MUC16 as well as aromatase. The knockdown studies suggested that silencing of MUC16 and TSPYL5 affected each other. These results indicate that there may be a feedback regulation between these two molecules during lung cancer cell proliferation. Furthermore, cisplatin treatment of these cells significantly abrogated MUC16 and TSPYL5 expression.

Conclusion
MUC16 is overexpressed in non small cell lung cancer. MUC16 plays an important role in lung cancer proliferation, through rapid G1/S transition in the cell cycle. MUC16 and TSPYL5 regulate each other during lung cancer cell proliferation, possibly through the aromatase pathway. One of the possible mechanisms through which cisplatin may decrease lung cancer proliferation is by downregulation of MUC16 and TSPYL5.

Background
Ataxia telangiectasia-mutated (ATM) is a DNA repair protein that is functionally absent in patients with A-T. Individuals with heterozygous or somatic homozygous mutations are predisposed to developing various cancers. Previous attempts to sequence ATM in cancer cell lines or in patient tissue samples have not identified any mutational hot-spots that are linked to the A-T phenotype or cancer predisposition, however much of the sequencing analysis performed to date has used traditional Sanger-based methodology, which is limited in its depth of coverage. Our lab has identified two non-small cell lung cancer (NSCLC) cell lines, NCI-H23 and NCI-H1395, which appear to be deficient for ATM. While H23 cells display sensitivity to ionizing radiation, typical of ATM deficiency, H1395 do not and moreover show downstream activation of several ATM targets. These previous results are important because the hallmark radiation sensitivity, and possible chemosensitivity of ATM-deficient patients may suggest that these patients would be more responsive to low-dose therapies. However, if cells show no detectable ATM but maintain an intact DNA repair mechanism, the usefulness of ATM as a biomarker is questionable. To assess if these observed differences were intrinsic to the ATM gene, we performed next-gen sequencing (NGS) to characterize the mutational spectra of H23 and H1395.

Methods
We used the semiconductor-based Ion Torrent PGM to sequence the coding region of ATM from genomic DNA isolated from H23, H1395, and H460 (ATM normal) cells. Variants were identified and mapped to the ATM gene. Additional western blots were performed to confirm mutation analysis.

Results
Several missense SNP mutations were identified in both H23 and H1395, however only one was shared by both, c.5948A>G, corresponding to amino acid substitution N1983S, which has previously been identified as a phosphovariant due to its proximity to the activation phosphorylation site, S1981. Interestingly, this SNP was not identified in H1395 in past sequencing attempts, demonstrating the power of NGS. Few of the other SNPs have previously been annotated in the COSMIC database, and it is as yet unknown whether they confer functional disruptions in the protein. Interestingly, the N1983S SNP is located in the same region of ATM used as the epitope for antibodies to both the phosporylated and unphosphorylated forms of ATM, raising the question of whether the antibodies are unable to bind to these variants and thus limiting detection. We performed western blots with antibodies to different epitopes to address this question.

Conclusion
We have yet to determine whether the mutations we identified in H23 and H1395 NSCLC cells are the cause of deficiency in these cells. However, use of deep-sequencing methodology has rapidly identified previously unknown mutations in these cell lines, and expansion of these studies to include a cohort of NSCLC patient samples may identify hot-spot mutations that were previously undetected by traditional sequencing methods. Moreover, these results imply that current methods for ATM detection may be insufficient; however supplementary deep-sequencing of these samples could be used to determine the nature of ATM deficiency, and to predict response to therapeutic treatments.