Squamous cell lung cancer

The treatment of lung cancer has been revolutionized by the discovery of specific targeted therapies such as erlotinib or gefitinib for EGFR-mutated lung cancer and crizotinib for ALK-translocated lung cancer. These successes have taught us that lung cancer is not just one disease, but a multitude of different diseases, best defined by the specific tumor genetic changes that are driving tumor growth and that can serve as targets for therapy. At Massachusetts General Hospital, we have been performing tumor genetic testing via SNaPshot, a panel of known oncogenic mutations, since 2009. We have found that approximately 40% of our patients do not have an identifiable mutation. For these patients, identifying what tumor genetic changes are driving their cancer will be critical to develop effective targeted therapy. This proposal therefore is focused on those patients who did not have an identifiable tumor mutation, with the hope that we can discover new targets for effective therapy. We plan to do this by two methods: First, we will perform whole exome sequencing, i.e. sequencing the coding regions of the tumor genome, on the tumors of patients who did not have any identifiable tumor mutations. Second, we will screen for chromosomal rearrangements involving tyrosine kinases. Tyrosine kinases are of particular interest in cancer, as they play a role in cell growth and proliferation, and more importantly are potentially “druggable.” With this two-pronged approach, we hope to identify novel tumor-related genetic changes that will lead to new targeted therapies for lung cancer.

Lung cancer is the leading cause of death for men and women in the United States. Contributing to the lethality of this disease is our inability to detect treatable early stage lung cancer resulting in the majority of lung cancer patients being diagnosed with incurable advanced disease. Computerized tomography (CT) screening has shown promising preliminary results in the detection of early stage lung cancer (with a 20% reduction in lung cancer mortality); however, it remains to be determined how it will be incorporated into standard practice. Biomarker-based screening methods that complement CT – and that are accessible to all current and former smokers – will be vital to making early lung cancer detection more attractive. Our goal at UC Davis is to combine our multidisciplinary leaders in lung cancer practice and research (Drs. Kelly, Gandara, Cooke, Yoneda and Murin) with “omics” experts (Drs. Miyamoto, Fiehn, and Lebrilla), who have developed sensitive technologies to detect and analyze sugars, lipids, and metabolites in human specimens, to identify blood biomarkers for early stage lung cancer that can be used in conjunction with CT screening. We will examine biomarker compositions of tumor tissue, non-malignant tissue, and blood in cancer patients, healthy controls and patients with benign lung nodules, to ultimately identify easily detectible blood markers that: 1) will detect early stage cancer and 2) are sensitive and specific. Such biomarkers will contribute to earlier, more informative lung cancer detection and make tailored lung cancer therapy and increased cure rates possible.

Lung cancer is the leading cause of cancer death in the world and has a dismal 16% 5 year survival rate. One reason for this discouraging statistic is that until recently, no early detection test had been validated. The National Lung Screening Trial (NLST) has recently demonstrated that CT screening results in a 20.3% decrease in lung cancer mortality compared to chest radiograph screening. It is now clear that CT scans will be increasingly used for lung cancer screening.

CT scans detect lung nodules in 20-50% of patients, while only 1-2% have lung cancer. Thus, CT scanning has low specificity (many false positive tests) for lung cancer. Patients and physicians are often put in the situation of having to decide between serial CT scans to look for nodule enlargement (which takes months) or an immediate biopsy strategy. Each approach has potential drawbacks: serial observation can result in delay in diagnosis and an immediate biopsy strategy can lead to unnecessary morbidity and expense. Biomarker based testing to determine risk for lung cancer is not currently available, but would be invaluable in aiding clinical decision making in this setting.

We propose to further refine and validate two promising risk biomarkers, abnormal chromosome numbers in sputum cells and a blood test assessing multiple bioactive proteins as guides to clinical decision making for patients and physicians confronted with indeterminate lung nodules and compare their performance to prediction models based on nodule and patient characteristics.

With the widespread availability of computed tomography (CT), an estimated 20 million chest CT-scans are now performed annually in the United States. From these studies about 4 million individuals have pulmonary nodules detected that require further evaluation. Only about 1-in-37 of these nodules will be lung cancer, even among high-risk individuals. Clinicians are charged with rapidly identifying the individuals with malignancy while minimizing the number of unnecessary testing to those without. The objective for the proposed studies is to develop a non-invasive test with excellent performance characteristics across the spectrum of clinical presentations of indeterminate pulmonary nodules would reduce the number of unnecessary invasive procedures and minimize the time from detection to diagnosis. Our hypothesis contends that, in the case of early-stage lung cancer, tumors possess a unique molecular phenotype that is discernable from highly-inflammatory, non-malignant/benign lesions commonly detected by CT-based imaging studies. A portion of these molecules are tumor-shed and provoke an immune response that can be exploited for diagnostic purposes. Our approach to address this matter is to use immunoproteomic methods to identify candidate biomarkers from clinical specimens that are useful for discerning non-neoplastic/benign nodules from malignacy. Next, we will develop and validate Luminex assays for no less than 20 candidate biomarkers and evaluate these internally to identify the optimal combination of biomarkers to form a diagnostic algorithm for this purpose. Finally, algorithm validation against external patient cohorts (in our possession) will be performed to evaluate the potential for future clinical application.

To significantly reduce the deaths from lung cancer, more accurate diagnostic tests are urgently needed to detect the disease early enough and benefit from a surgical cure. We recently identified 164 novel candidate protein markers in lung cancer from extensive shotgun proteomic profiling of Stage 1 lung cancers. These candidate protein biomarkers are abundantly expressed in Stage 1 lung cancer tissues but not in normal counterparts. While standard ELISA-based assays to validate these candidates can be done, this approach is low throughput, very slow and expensive. In this application, we propose to develop and validate a non-invasive diagnostic test for lung cancer based on the detection of autoantibodies directed against our candidate protein biomarkers that are markedly differentially expressed in lung cancer and present in the circulation. More specifically, we propose to establish an assay to detect and quantitate lung cancer autoantibodies directed against new proteomic biomarker candidates. We will validate the diagnostic efficacy of our autoantibody signature in blood samples taken from individuals with Stage 1 lung cancer or matched nodules controls and integrate these findings in the diagnostic setting of patients presenting with lung nodules. Establishing this new approach may ultimately lead to improving the accuracy of the diagnosis of lung cancer when the disease is surgically curable.

Lung cancer is the leading cause of cancer death. Recently, the National Lung Cancer Screening Trial reported a 20% reduction in lung cancer mortality and three lung cancer chemoprevention trials report reductions in markers associated with lung cancer development. These studies highlight the possibilities for further reductions in lung cancer mortality through early detection and chemoprevention strategies among high-risk smokers. Our group has previously identified smoking- and lung cancer-specific molecular alterations in the epithelial cells that line the major airways which can be obtained during bronchoscopy. In this study, we propose to use next generation sequencing technology to quantify RNA expression alterations (RNA-Seq) in normal airway epithelial cells in patients with and without precancerous lesions in their airways. We will characterize how RNA expression is affected by the presence of precancerous lesions and identify RNA expression changes that predict if the lesions will progress or regress. The predictors will be important in selecting which high-risk smokers to enroll in chemoprevention trials, identifying which smokers are benefiting from the therapy, and identifying new targets for chemoprevention. Finally, we will test whether our RNA expression predictors are associated with the presence or future development of lung cancer. Common biology between RNA expression alterations in normal airway epithelial cells of patients with precancerous lesions and patients with lung cancer may indicate early events in lung carcinogenesis and have additional utility as early markers of lung cancer.

There is a lack of strategies for detection of small cell lung cancer (SCLC), a difficult-to-treat lung cancer subtype, which we believe is due to our poor understanding of the molecular development of the disease. Moreover, while screening by clinical measures reduces lung cancer mortality, the detected lung cancers are less likely to be SCLCs warranting the need for additional detection methods (e.g., biological).

Our earlier work revealed cancer-associated markers and profiles in “normal” (to the “eye”) airways close to lung tumors, an effect we refer to as “molecular field cancerization,” in which there is field around the tumor that is enriched with malignant properties. We found field effects in “normal” airways that increased with time following tumor removal were different between patients with and without lung cancer or different subtypes of the disease, and were present in other respiratory tract compartments, including the nose.

Although we previously found that airways near SCLCs display excessive genetic abnormalities, the molecular field effect in SCLC and its relevance to detection of the disease is unknown. The proposed project aims to characterize, by high-throughput technologies, the molecular makeup of cells lining the nose and airways and of tumors from the SCLC patients compared to molecular profiles of samples from patients with benign disease or other lung cancer types, and then test these markers for their capacity to detect SCLC in easily accessible sites. We expect by using such unique resources and approaches to delineate novel markers for SCLC detection.