Predicting prognosis of nasopharyngeal carcinoma based on deep learning: peritumoral region should be valued.

BACKGROUND: The purpose of this study was to explore whether incorporating the peritumoral region to train deep neural networks could improve the performance of the models for predicting the prognosis of NPC. METHODS: A total of 381 NPC patients who were divided into high- and low-risk groups according to progression-free survival were retrospectively included. Deeplab v3 and U-Net were trained to build segmentation models for the automatic segmentation of the tumor and suspicious lymph nodes. Five datasets were constructed by expanding 5, 10, 20, 40, and 60 pixels outward from the edge of the automatically segmented region. Inception-Resnet-V2, ECA-ResNet50t, EfficientNet-B3, and EfficientNet-B0 were trained with the original, segmented, and the five new constructed datasets to establish the classification models. The receiver operating characteristic curve was used to evaluate the performance of each model. RESULTS: The Dice coefficients of Deeplab v3 and U-Net were 0.741(95%CI:0.722-0.760) and 0.737(95%CI:0.720-0.754), respectively. The average areas under the curve (aAUCs) of deep learning models for classification trained with the original and segmented images and with images expanded by 5, 10, 20, 40, and 60 pixels were 0.717 ± 0.043, 0.739 ± 0.016, 0.760 ± 0.010, 0.768 ± 0.018, 0.802 ± 0.013, 0.782 ± 0.039, and 0.753 ± 0.014, respectively. The models trained with the images expanded by 20 pixels obtained the best performance. CONCLUSIONS: The peritumoral region NPC contains information related to prognosis, and the incorporation of this region could improve the performance of deep learning models for prognosis prediction.

In vitro assessment of the role of DpC in the treatment of head and neck squamous cell carcinoma.

The present study aimed to investigate the antitumor efficacy of di-2-pyridylketone-4-cyclohexyl-4-methyl-3-thiosemicarbazone (DpC) and di-2-pyridylketone-4,4,-dimethyl-3-thiosemicarbazone (Dp44mT) on head and neck squamous cell carcinoma (HNSCC) cells. The proliferation and apoptosis of HNSCC cells treated with the iron chelators DpC and Dp44mT were detected. The mechanism of DpC-induced apoptosis on HNSCC cells was investigated. The human HNSCC cell lines FaDu, Cal-27 and SCC-9 were cultured in vitro and exposed to gradient concentrations of DpC and Dp44mT. A Cell Counting Kit-8 assay was used to detect the viability of FaDu, Cal-27, SCC-9 cells. Double staining with annexin V and propidium iodide was performed for the detection of the proportion of apoptotic FaDu, Cal-27 and SCC-9 cells following treatment. The nuclear damage to Cal-27 cells that were treated with DpC was detected by Hoechst staining. Finally, western blot analysis was used to detect the expression of proteins associated with the DNA damage pathway in Cal-27 cells that were treated with DpC. The CCK-8 assay showed that treatment with DpC and Dp44mT was able to markedly inhibit the viability of FaDu, Cal-27 and SCC-9 cells in a concentration-dependent manner. In comparison to Dp44mT, treatment with DpC exhibited a more effective inhibitory effect on the viability of HNSCC cells. The proportion of apoptotic cells detected by flow cytometry increased in a dose-dependent manner in all cell lines following DpC and Dp44mT treatment, with the proportion of apoptotic HNSCC cells induced by DpC treatment being significantly higher compared with Dp44mT (P<0.05). The results of Hoechst staining revealed that the nuclei of Cal-27 cells exhibited morphological changes in response to DpC treatment, including karyopyknosis and nuclear fragmentation. The expression of DNA damage-associated proteins, including phosphorylated (p)-serine-protein kinase ATM, p-serine/threonine-protein kinase Chk1 (p-Chk-1), p-serine/threonine-protein kinase ATR (p-ATR), p-Chk-2, poly (ADP-ribose) polymerase, p-histone H2AX, breast cancer type 1 susceptibility protein, p-tumor protein P53, increased with increasing concentration of DpC in Cal-27 cells. Treatment with DpC and Dp44mT markedly inhibited cell viability and increased the apoptotic rates in human HNSCC cells in a concentration-dependent manner. DpC exhibited a stronger antitumor effect compared with Dp44mT, potentially inducing the apoptosis of HNSCC cells via the upregulation of DNA damage repair-associated proteins.