[摘 要] 目的：探讨复发/转移性鼻咽癌（NPC）接受含PD-1单抗免疫治疗的临床特征和预后影响因素。方法：回顾性分析2019年3月至2024年7月期间南部战区总医院确诊的95例NPC患者的临床资料和外周血生化及免疫学指标。预后分析采用Kaplan-Meier曲线，组间比较使用Log-rank检验，采用Cox比例风险模型进行单因素和多因素分析。结果：95例患者中男性81例，女性14例，中位年龄49.72岁（16~74岁），Ⅳ期91例（95.79%），所有患者均采用免疫治疗，联合或不联合化疗方案治疗，中位无进展生存期（mPFS）为10.5个月，客观缓解率（ORR）70.53%，疾病控制率（DCR）89.47%，接受含铂治疗方案患者PFS相对更长，且差异有统计学意义。紫杉醇 + 顺铂 + 氟尿嘧啶（TPF）对比吉西他滨 + 顺铂（GP）和紫杉醇 + 顺铂（TP）显示出更长的PFS，但差异无统计学意义。不同PD-1单抗治疗组间的PFS未显示出有统计学意义的差异。单因素及多因素Cox回归分析结果显示，肿瘤复发状态、初始血浆EBV感染状态、治疗周期数、基线外周血SII是复发/转移性NPC患者接受PD-1抑制剂治疗疗效预测的独立相关因素（均P < 0.05），并且非复发患者、初始血浆EBV DNA阳性、接受 ≥ 4治疗周期、基线外周血SII < 772.81的患者接受PD-1抑制剂治疗预后相对更好。结论：在接受PD-1抑制剂治疗的复发/转移性NPC患者中，非复发患者、初始血浆EBV DNA阳性、≥ 4治疗周期且外周血SII < 772.81者PFS相对更长，可早期识别免疫治疗效果不佳患者并精准干预。

Esophageal cancer is a prevalent malignant tumor of the digestive tract in China, and radical surgery remains the cornerstone of its comprehensive treatment. However, multifactorial challenges such as postoperative gastrointestinal tract reconstruction, traumatic stress, and tumor-related metabolic disturbances render esophageal cancer patients highly susceptible to malnutrition. Perioperative nutritional support therapy plays a crucial role in enhancing surgical safety, improving clinical outcomes, and elevating patients' quality of life by regulating metabolic homeostasis, preserving organ function, and optimizing the immune microenvironment. This article reviews the mechanisms underlying malnutrition in esophageal cancer, methods for nutritional status assessment, and precision intervention pathways based on multi-omics evaluations. The aim is to strengthen clinicians' awareness of standardized perioperative nutritional management for esophageal cancer patients and promote its clinical implementation, thereby facilitating postoperative recovery and improving long-term quality of life.

BACKGROUND:The regeneration and remodeling of Achilles tendon rupture after surgery are difficulties in clinical treatment.Tissue engineering hydrogels afford the possibility on the healing of postoperative Achilles tendon. OBJECTIVE:To investigate the effect of tannic acid modified interpenetrating network hydrogel on tissue regeneration and remodeling of ruptured Achilles tendon in rats. METHODS:(1)The interpenetrating network hydrogel was prepared under the blue light and the immersion of CaSO4 solution.The micromorphology,mechanical properties,adhesion properties,in vitro drug release properties,and biocompatibility of hydrogels were characterized.(2)Thirty Sprague-Dawley rats were randomly divided into sham operation group,operation group,and hydrogel group,with 10 rats in each group.The animal model of Achilles tendon rupture was established in the latter two groups.In the operation group,the ruptured Achilles tendon was sutured using the modified Kessler method.In the hydrogel group,the ruptured Achilles tendon was repaired by the same method,and the tannic acid modified interpenetrating network hydrogel patch was completely wrapped around the joint of the broken end.Four weeks after the operation,imaging examination,histological evaluation,biomechanical test,and the level test of inflammatory factors were performed. RESULTS AND CONCLUSION:(1)Scanning electron microscope showed that tannic acid modified interpenetrating network hydrogels had porous microstructure with pore size of 3-10 μm,and the hydrogels had good in vitro drug release properties,adhesion strength and tensile strength.CCK-8 assay and live/dead staining showed that the hydrogel had no significant effect on the proliferation activity of rat bone marrow mesenchymal stem cells,and had good biocompatibility.(2)MRI imaging showed that compared with the operation group,the Achilles tendon in the hydrogel group showed uniform low signal,the thickness of the anteroposterior diameter of the Achilles tendon was reduced,and the boundary between the Achilles tendon and the surrounding tissue was more clear,and the performance was more similar to that of the sham operation group.Hematoxylin-eosin staining and Masson staining showed that the tendon fibers in the operation group were arranged in a loose and chaotic manner,with increased cell density and disordered arrangement,accompanied by obvious inflammatory cell infiltration,and intratendinous ossification appeared in some areas.In the hydrogel group,the tendon fibers were arranged in an orderly manner;the cell density was reduced and arranged orderly;the inflammatory cell infiltration was significantly reduced.The tensile strength of Achilles tendon in the operation group was lower than that in the sham operation group(P<0.05).The tensile strength of Achilles tendon in the hydrogel group was higher than that in the operation group(P<0.05).Compared with sham operation group,the mass concentration and mRNA expression of interleukin-1β,interleukin-6,and tumor necrosis factor α in Achilles tendon of rats were increased in the operation group(P<0.05).Compared with the operation group,the level and mRNA expression of three inflammatory factors were decreased in the hydrogel group.(3)It is concluded that tannic acid modified interpenetrating network hydrogel can inhibit the local inflammatory response and promote the tendon remodeling.

Background: A 24-item early-onset scoliosis questionnaire (EOSQ-24) has been developed as a valid tool for assessing the physical and emotional function of patients with early-onset scoliosis (EOS). Previous studies that conducted transcultural adaptation of the original EOSQ-24 into other languages have demonstrated the high reliability of the questionnaire. However, a Korean version of the EOSQ-24 is not available, limiting optimal patient assessment in this nation. Therefore, this study was conducted to develop and validate a Korean version of EOSQ-24. Methods: The original English version of the EOSQ-24 was cross-culturally adapted following standard guidelines. The final version of the Korean EOSQ-24 was prospectively applied to a group of patients who were native Korean speakers. Internal consistency was evaluated using the Cronbach α coefficient and item-total correlations. The mean, standard deviation, floor effect, and ceiling effect of each item were also assessed. Results: A total of 102 caregivers of patients with EOS (45 males and 57 females) completed the Korean EOSQ-24. The patients’ average age was 6.39 ± 2.16 years. The Cronbach α coefficient for the 24-item scale was 0.942, indicating excellent reliability. The corrected item-total correlation coefficients ranged between 0.449 and 0.788. The mean value of the EOSQ-24 questionnaire was 3.19. Floor and ceiling effects for all questionnaires were below 30%, which could be considered acceptable. Conclusions Our study successfully developed and validated the Korean version of the EOSQ-24, providing a reliable instrument for assessing the physical and emotional well-being of patients with EOS and their caregivers in the Korean context. The widespread adoption of the Korean EOSQ-24 in clinical practice and research settings can enhance the quality of care and improve outcomes for individuals affected by EOS in Korea.

Objective: To evaluate the safety and efficacy of coil embolization of venous segments in patients with Type IIa pelvic arteriovenous malformations (AVMs). Materials and Methods: A retrospective study was performed on 13 patients (median age, 43 years, range 20–62 years, 7 males) who underwent transvenous coil embolization for Type IIa pelvic AVM (characterized by multiple arterioles shunting to focal venous segments of a single draining vein) without the use of additional liquid embolic agents from March 2017 to February 2023. Treatment outcomes were analyzed based on clinical evaluations, post angiography findings, and follow-up CT. Results: Fourteen procedures were performed on 13 patients. Except in one patient, all treatments were completed in a single session. Transvenous access was employed in 10 procedures, whereas direct puncture was used in four sessions. The embolization procedures used an average of 55.7 ± 58.5 coils (range, 7–238) and lasted an average of 127.3 ± 39.5 minutes.The technical success rate was 92.9% (of 13/14). All patients reported symptom improvement. Follow-up CT scans showed complete occlusion of the AVM without recurrence in ten of the 13 patients. There was one minor adverse event: a small retroperitoneal hemorrhage, likely related to direct puncture, which resolved spontaneously. No other adverse events were observed. Conclusion Coil embolization of the draining vein segment, without the use of additional liquid embolic agents is a safe and effective method for managing Type IIa pelvic AVM.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: To evaluate the safety and efficacy of coil embolization of venous segments in patients with Type IIa pelvic arteriovenous malformations (AVMs). Materials and Methods: A retrospective study was performed on 13 patients (median age, 43 years, range 20–62 years, 7 males) who underwent transvenous coil embolization for Type IIa pelvic AVM (characterized by multiple arterioles shunting to focal venous segments of a single draining vein) without the use of additional liquid embolic agents from March 2017 to February 2023. Treatment outcomes were analyzed based on clinical evaluations, post angiography findings, and follow-up CT. Results: Fourteen procedures were performed on 13 patients. Except in one patient, all treatments were completed in a single session. Transvenous access was employed in 10 procedures, whereas direct puncture was used in four sessions. The embolization procedures used an average of 55.7 ± 58.5 coils (range, 7–238) and lasted an average of 127.3 ± 39.5 minutes.The technical success rate was 92.9% (of 13/14). All patients reported symptom improvement. Follow-up CT scans showed complete occlusion of the AVM without recurrence in ten of the 13 patients. There was one minor adverse event: a small retroperitoneal hemorrhage, likely related to direct puncture, which resolved spontaneously. No other adverse events were observed. Conclusion Coil embolization of the draining vein segment, without the use of additional liquid embolic agents is a safe and effective method for managing Type IIa pelvic AVM.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.