基于剂量组学预测肺癌患者放射性肺炎发生的研究

目的 探讨剂量组学在预测肺癌根治性放疗患者放射性肺炎发生中的应用潜能。方法 回顾性收集行根治性放疗的314例肺癌患者的临床资料、放疗剂量文件、定位及随访CT图像,根据临床资料及影像学随访资料对放射性肺炎进行分级,提取全肺的剂量组学特征,构建机器学习模型。应用1000次自助抽样法（bootstrap）的最小绝对值收敛和选择算子嵌套逻辑回归（LASSO‐LR）及1000次bootstrap的赤池信息量准则（AIC）向后法筛选与放射性肺炎相关的剂量组学特征,随机按照7∶3划分为训练集及验证集,应用逻辑回归建立预测模型,并应用ROC曲线及校正曲线评价模型的性能。结果 共提取120个剂量组学特征,经LASSO‐LR降维筛选得到12个特征进入“特征池”,再经过AIC向后法筛选,最终筛选出6个剂量组学特征进行模型构建,训练集AUC为0.77（95%CI为0.65~0.87）,独立验证集AUC为0.72（95%CI为0.64~0.81）。结论 利用剂量组学建立的预测模型具有预测放射性肺炎发生的潜力,但仍需继续纳入多中心数据及前瞻性数据进一步挖掘剂量组学的应用潜能。     

玻璃体切割联合内界膜剥除治疗糖尿病性黄斑水肿

评估玻璃体切割联合内界膜剥除对糖尿病性黄斑水肿（DME）的疗效。方法：回顾性病例对照研究。2014年6月至2017年1月间因糖尿病视网膜病变合并玻璃体积血或增殖病变于温州医科大学附属眼视光医院杭州院区行玻璃体切割手术治疗，且术前或术中经光学相干断层扫描（OCT）检查确诊合并DME的患者31例（33眼）纳入研究。16例（18眼）术中联合内界膜剥除作为剥膜组，15例（15眼）仅接受玻璃体切割手术治疗者作为对照组。所有手术均由同一医师主刀完成。术后1、3个月随访时复查OCT，对比观察黄斑中心厚度（CMT）和视力的术后变化情况。随访中CMT和最佳矫正视力（BCVA）比较采用重复测量方差分析，组间CMT和BCVA比较采用独立样本t检验。结果：手术前，手术后1、3个月2组间比较LogMAR视力总体差异有统计学意义（F=15.93，P<0.001）。术后 1个月时剥膜组BCVA高于对照组（t=2.55，P=0.02），但术后3个月时2组间差异无统计学意义（t=0.82， P=0.42）。手术前，手术后1、3个月CMT总体差异无统计学意义（F=2.85，P=0.065）。术后1、3个月时，剥膜组的CMT均低于对照组，2组间差异均有统计学意义（t=2.24，P=0.03；t=3.79，P=0.001）。术后1个月时，剥膜组有效（与术前比CMT减少20%以上）、无效（变化不超过20%）及恶化（增厚超过 20%）的例数分别为8、6、4例，术后3个月时则分别为11、5、2例，与对照组相比，术后1个月时组间差异无统计学意义（Z=-1.687，P=0.092），术后3个月时剥膜组DME改善有效比例明显高于对照组，组间差异有统计学意义（Z=-2.177，P=0.029）。结论：对于非牵拉性DME，内界膜剥除有助于术后早期DME消退。     

Craving-induced effects of different drug cues on persons abstaining from heroin

Background: The key factors of inducing drug cravings in persons abstaining from drug use remain a focus of addictions research. Given the accumulating evidences, the scope of cues investigated in the cue-reactivity paradigm has increased considerably. Yet, few studies have examined the effects of the intensity and endurance of different types of cues on their ability to induce craving. This study investigated differences among drug-cue words, negative physiological-cue words, and negative social-cue words in the induction of drug cravings among persons abstaining from heroin.

Methods: The sample consisted of 149 male abstinent heroin abusers from four addiction rehabilitation centers in China. Based on their abstinence lengths, they were labeled as short-term, medium-term, and long-term abstainer participants respectively. All participants completed a stress-imagery task and rated craving by visual analog scale.

Results: There was a significant interaction of cue type and abstinence length. There was no difference on the craving induced by three types of cue words in the short-term group. In the medium-term group, craving induced by negative social-cue words was significantly stronger than that by negative physiological-cue words, but not that by drug-cue words. In the long-term group, the craving induced by negative social-cue words remained the strongest, significantly stronger than that by both drug-cue words and negative physiological-cue words.

Conclusion: Negative social-cue words presented in the current study retain the ability to induce craving in heroin abstainers; this finding suggests that negative social cues encountered under more general circumstances could be a risk factor for relapse.     

Deferoxamine promotes recovery of traumatic spinal cord injury by inhibiting ferroptosis

Ferroptosis is an iron-dependent novel cell death pathway. Deferoxamine, a ferroptosis inhibitor, has been reported to promote spinal cord injury repair. It has yet to be clarified whether ferroptosis inhibition represents the mechanism of action of Deferoxamine on spinal cord injury recovery. A rat model of Deferoxamine at thoracic 10 segment was established using a modified Allen's method. Ninety 8-week-old female Wistar rats were used. Rats in the Deferoxamine group were intraperitoneally injected with 100 mg/kg Deferoxamine 30 minutes before injury. Simultaneously, the Sham and Deferoxamine groups served as controls. Drug administration was conducted for 7 consecutive days. The results were as follows:(1) Electron microscopy revealed shrunken mitochondria in the spinal cord injury group.(2) The Basso, Beattie and Bresnahan locomotor rating score showed that recovery of the hindlimb was remarkably better in the Deferoxamine group than in the spinal cord injury group.(3) The iron concentration was lower in the Deferoxamine group than in the spinal cord injury group after injury.(4) Western blot assay revealed that, compared with the spinal cord injury group, GPX4, xCT, and glutathione expression was markedly increased in the Deferoxamine group.(5) Real-time polymerase chain reaction revealed that, compared with the Deferoxamine group, mRNA levels of ferroptosis-related genes Acyl-CoA synthetase family member 2(ACSF2) and iron-responsive element-binding protein 2(IREB2) were up-regulated in the Deferoxamine group.(6) Deferoxamine increased survival of neurons and inhibited gliosis. These findings confirm that Deferoxamine can repair spinal cord injury by inhibiting ferroptosis. Targeting ferroptosis is therefore a promising therapeutic approach for spinal cord injury.