Applications of array-CGH for lung cancer

Lung Cancer Subtype Diagnosis by Fusing Image-Genomics Data and Hybrid Deep Networks

Accurate diagnosis of cancer subtypes is crucial for precise treatment, because different cancer subtypes are involved with different pathology and require different therapies. Although deep learning techniques have made great success in computer vision and other fields, they do not work well on Lung cancer subtype diagnosis, due to the distinction of slide images between different cancer subtypes is ambiguous. Furthermore, they often over-fit to high-dimensional genomics data with limited samples, and do not fuse the image and genomics data in a sensible way. In this paper, we propose a hybrid deep network based approach LungDIG for Lung cancer subtype Diagnosis by fusing Image-Genomics data. LungDIG first tiles the tissue slide image into small patches and extracts the patch-level features by fine-tuning an Inception-V3 model. Since the patches may contain some false positives in non-diagnostic regions, it further designs a patch-level feature combination strategy to integrate the extracted patch features and maintain the diversity between different cancer subtypes. At the same time, it extracts the genomics features from Copy Number Variation data by an attention based nonlinear extractor. Next, it fuses the image and genomics features by an attention based multilayer perceptron (MLP) to diagnose cancer subtype. Experiments on TCGA lung cancer data show that LungDIG can not only achieve higher accuracy for cancer subtype diagnosis than state-of-the-art methods, but also have a high authenticity and good interpretability.

Prognostic Impact of EGFR Amplification and Visceral Pleural Invasion in Early Stage Pulmonary Squamous Cell Carcinomas Patients after Surgical Resection of Primary Tumor

Over the last few decades, an increasing amount of information has been accumulated on biomarkers in non-small cell lung cancer (NSCLC). Despite these advances, most biomarkers have been identified in the adenocarcinoma histological subtype (AC). However, the application of molecular-targeted therapies in the prognosis and treatment of SCC in the clinical setting is very limited, becoming one of the main focus areas in research. Here, we prospectively analyzed the frequency of numerical/structural abnormalities of chromosomes 5, 7, 8, 9, 13 and 22 with FISH in 48 pulmonary SCC patients. From a total of 12 probes, only abnormalities of the 7p12 and 22q12 chromosomal regions were identified as unique genetic variables associated with the prognosis of the disease. The study for these two chromosomal regions was extended to 108 patients with SCC. Overall, chromosome losses were observed more frequently than chromosome gains, i.e., 61% versus 19% of all the chromosome abnormalities detected. The highest levels of genetic amplification were detected for the 5p15.2, 7p12, 8q24 and 22q11 chromosome bands, of which several genes are potentially involved in the pathogenesis of SCC, among others, include the EGFR gene at chromosome 7p12. Patients who displayed EGFR amplification (n = 13; 12%) were mostly older than 65 years (p = 0.07) and exclusively patients in early T-primary tumor stage (pT1−pT2; p = 0.03) with a significantly shortened overall survival (OS) (p ≤ 0.001). Regarding prognosis, the clinical, biological, and histopathologic characteristics of the disease that displayed a significant adverse influence on OS in the univariate analysis included patients older than 65 years (p = 0.02), the presence of lymph node involvement (p = 0.005), metastasis (p = 0.01) and, visceral pleural invasion (VPI) at diagnosis (p = 0.04). EGFR amplification also conferred an adverse impact on patient OS in the whole series (p = 0.02) and especially in patients in early stages (pT1−pT2; p = 0.01). A multivariate analysis of the prognostic factors for OS showed that the most informative combination of independent variables to predict an adverse outcome was the presence of VPI and/or EGFR amplification (p < 0.001). Based on these two variables, a scoring system was built to stratify patients into low- (no adverse features: score 0; n = 69), intermediate- (one adverse feature: score 1; n = 29) and high-risk (two adverse features: score 2; n = 5) groups, with significantly different (p = 0.001) OS rates at 50 months, which were as following: 32%, 28% and 0%, respectively. In the present study, we show that the presence of a high level of 7p12 (EGFR) amplification, exclusively detected in early stage SCC (pT1−pT2), is an independent adverse prognostic factor for OS. The identification of the EGFR gene copy number using FISH techniques may provide a more accurate diagnosis of high-risk populations after the complete resection of the primary tumor. When combined with VPI, three groups of pulmonary SCC were clearly identified that show the extent of the disease. This is of such importance that further prospective studies are necessary in larger series of SCC patients to be classified at the time of diagnosis. This could be achieved with the combined assessment of 7p12 amplification and VPI in primary tumor samples.

Do physicians think genomic medicine will be useful for patient care?

Significant technological improvements over the last decade have led to a vast expansion in the understanding of the genomic architecture of human disease. However, the use of genomic information, so-called genomic medicine, in routine clinical care, has been slow in comparison to the growth in genomic discovery. The uptake of genomic technology into clinical practice will depend on physicians' perspectives of its utility in patient care. We review recent literature addressing physician attitudes regarding the usefulness and limitations of genomic testing. We conclude by proposing research areas to better understand the role physicians will play in the uptake of genomic information into clinical medicine.