Reliable entity subtyping is paramount for therapy stratification in lung cancer. Morphological evaluation remains the basis for entity subtyping and directs the application of additional methods such as immunohistochemistry (IHC). The decision of whether to perform IHC for subtyping is subjective, and access to IHC is not available worldwide. Thus, the application of additional methods to support morphological entity subtyping is desirable. Therefore, the ability of convolutional neuronal networks (CNNs) to classify the most common lung cancer subtypes, pulmonary adenocarcinoma (ADC), pulmonary squamous cell carcinoma (SqCC), and small-cell lung cancer (SCLC), was evaluated. A cohort of 80 ADC, 80 SqCC, 80 SCLC, and 30 skeletal muscle specimens was assembled; slides were scanned; tumor areas were annotated; image patches were extracted; and cases were randomly assigned to a training, validation or test set. Multiple CNN architectures (VGG16, InceptionV3, and InceptionResNetV2) were trained and optimized to classify the four entities. A quality control (QC) metric was established. An optimized InceptionV3 CNN architecture yielded the highest classification accuracy and was used for the classification of the test set. Image patch and patient-based CNN classification results were 95% and 100% in the test set after the application of strict QC. Misclassified cases mainly included ADC and SqCC. The QC metric identified cases that needed further IHC for definite entity subtyping. The study highlights the potential and limitations of CNN image classification models for tumor differentiation.

Abstrac t : High expression of SERPINA1 gene encoding acute phase protein, alpha1-antitrypsin (AAT), is associated with various tumors. We sought to examine the significance of SERPINA1 and AAT protein in non-small-cell lung cancer (NSCLC) patients and NSCLC cell lines. Tumor and adjacent non-tumor lung tissues and serum samples from 351 NSCLC patients were analyzed for SERPINA1 expression and AAT protein levels. We also studied the impact of SERPINA1 expression and AAT protein on H1975 and H661 cell behavior, in vitro. Lower SERPINA1 expression in tumor but higher in adjacent non-tumor lung tissues (n = 351, p = 0.016) as well as higher serum levels of AAT protein (n = 170, p = 0.033) were associated with worse survival rates. Specifically, in NSCLC stage III patients, higher blood AAT levels (>2.66 mg/mL) correlated with a poor survival (p = 0.002). Intriguingly, levels of serum AAT do not correlate with levels of C-reactive protein, neutrophils-to-leukocyte ratio, and do not correlate with SERPINA1 expression or AAT staining in the tumor tissue. Additional experiments in vitro revealed that external AAT and/or overexpressed SERPINA1 gene significantly improve cancer cell migration, colony formation and resistance to apoptosis. SERPINA1 gene and AAT protein play an active role in the pathogenesis of lung cancer and not just reflect inflammatory reaction related to cancer development.

Idiopathic pulmonary fibrosis (IPF) is a chronic progressive interstitial lung disease characterized by patchy scarring of the distal lung with limited therapeutic options and poor prognosis. Here, we show that conditional deletion of the ubiquitin ligase Nedd4-2 (Nedd4l) in lung epithelial cells in adult mice produces chronic lung disease sharing key features with IPF including progressive fibrosis and bronchiolization with increased expression of Muc5b in peripheral airways, honeycombing and characteristic alterations in the lung proteome. NEDD4-2 is implicated in the regulation of the epithelial Na⁺ channel critical for proper airway surface hydration and mucus clearance and the regulation of TGFβ signaling, which promotes fibrotic remodeling. Our data support a role of mucociliary dysfunction and aberrant epithelial pro-fibrotic response in the multifactorial disease pathogenesis. Further, treatment with the anti-fibrotic drug pirfenidone reduced pulmonary fibrosis in this model. This model may therefore aid studies of the pathogenesis and therapy of IPF.

CT screening for lung cancer is effective in reducing mortality, but there are areas of concern, including a positive predictive value of 4% and development of interval cancers. A blood test that could manage these limitations would be useful, but development of such tests has been impaired by variations in blood collection that may lead to poor reproducibility across populations.