Objective:To investigate the role of three-dimensional dose distribution-based deep learning model in predicting distant metastasis of head and neck cancer.Methods:Radiotherapy and clinical follow-up data of 237 patients with head and neck cancer undergoing intensity-modulated radiotherapy (IMRT) from 4 different institutions were collected. Among them, 131 patients from HGJ and CHUS institutions were used as the training set, 65 patients from CHUM institution as the validation set, and 41 patients from HMR institution as the test set. Three-dimensional dose distribution and GTV contours of 131 patients in the training set were input into the DM-DOSE model for training and then validated with validation set data. Finally, the independent test set data were used for evaluation. The evaluation content included the area under receiver operating characteristic curve (AUC), balanced accuracy, sensitivity, specificity, concordance index and Kaplan-Meier survival curve analysis.Results:In terms of prognostic prediction of distant metastasis of head and neck cancer, the DM-DOSE model based on three-dimensional dose distribution and GTV contours achieved the optimal prognostic prediction performance, with an AUC of 0.924, and could significantly distinguish patients with high and low risk of distant metastasis (log-rank test, P<0.001). Conclusion:Three-dimensional dose distribution has good predictive value for distant metastasis in head and neck cancer patients treated with IMRT, and the constructed prediction model can effectively predict distant metastasis.

Chinese hamster ovary (CHO) cells are the preferred host cells for the production of complex recombinant therapeutic proteins. Adenine phosphoribosyltransferase (APRT) is a key enzyme in the purine biosynthesis step that catalyzes the condensation of adenine with phosphoribosylate to form adenosine phosphate AMP. In this study, the gene editing technique was used to knock out the aprt gene in CHO cells. Subsequently, the biological properties of APRT-KO CHO cell lines were investigated. A control vector expressed an enhanced green fluorescent protein (EGFP) and an attenuation vector (containing an aprt-attenuated expression cassette and EGFP) were constructed and transfected into APRT-deficient and wild-type CHO cells, respectively. The stable transfected cell pools were subcultured for 60 generations and the mean fluorescence intensity of EGFP in the recombinant CHO cells was detected by flow cytometry to analyze the EGFP expression stability. PCR amplification and sequencing showed that the aprt gene in CHO cell was successfully knocked out. The obtained APRT-deficient CHO cell line had no significant difference from the wild-type CHO cells in terms of cell morphology, growth, proliferation, and doubling time. The transient expression results indicated that compared with the wild-type CHO cells, the expression of EGFP in the APRT-deficient CHO cells transfected with the control vector and the attenuation vector increased by 42%±6% and 56%±9%, respectively. Especially, the EGFP expression levels in APRT-deficient cells transfected with the attenuation vector were significantly higher than those in wild-type CHO cells (P < 0.05). The findings suggest that the APRT-deficient CHO cell line can significantly improve the long-term expression stability of recombinant proteins. This may provide an effective cell engineering strategy for establishing an efficient and stable CHO cell expression system.
Adenine/metabolism* ; Adenine Nucleotides ; Adenine Phosphoribosyltransferase/genetics* ; Adenosine Monophosphate ; Animals ; CHO Cells ; Cricetinae ; Cricetulus ; Recombinant Proteins/genetics*