摘要目的:探讨化疗早期 18F－脱氧葡萄糖(FDG)正电子发射型计算机断层扫描(PET/CT)显像最大标准摄取值变化率(ΔSUVmax%)、Deauville评分和C－myc基因重排对弥漫性大B细胞淋巴瘤(DLBCL)的预后预测价值。 方法:2010年9月至2016年12月山西省肿瘤医院收治的、经病理证实的原发DLBCL患者83例，分别于化疗前1周和化疗早期(34例在化疗1个周期后的17~21 d，49例在化疗2个周期后的17~21 d)行 18F－FDG PET/CT扫描。勾画感兴趣区，计算治疗前后ΔSUVmax%。采用Deauville 5分法对患者化疗早期的PET/CT图像进行评分。采用原位荧光杂交检测C－myc基因重排情况。随访时间为36~111个月，研究终点为无进展生存时间(PFS)。采用受试者工作特征(ROC)曲线分析、 χ2检验、Spearman相关分析、Log rank检验、Cox比例风险模型进行统计分析。 结果:83例DLBCL患者中，19例在随访期内进展。化疗早期预测肿瘤进展的最佳界值ΔSUVmax%为62.59%，Deauville评分取5分，两种方法的灵敏度、特异度、准确性差异均无统计学意义(均 P>0.05)。化疗早期ΔSUVmax%与Deauville评分、C－myc基因重排均呈负相关( rs分别为－0.889和－0.862，均 P<0.001)，Deauville评分与C－myc基因重排呈正相关( rs＝0.781， P<0.001)。ΔSUVmax%≥62.59%组(57例)和<62.59%组(26例)的中位PFS分别为59.0和16.0个月，差异有统计学意义( P<0.001)。Deauville评分<5分组(61例)和＝5分组(22例)的中位PFS分别为59.0和15.0个月，两组差异有统计学意义( P<0.001)。C－myc基因重排阴性组(62例)和阳性组(21例)的中位PFS分别为59.0和15.0个月，两组差异有统计学意义( P<0.001)。ΔSUVmax%<62.59%并Deauvulle评分＝5分组、ΔSUVmax%<62.59%并C－myc基因重排阳性组、Deauville评分＝5分并C－myc基因重排阳性组患者的中位PFS分别为15.5、15.0和13.5个月，明显差于其他亚组(均 P<0.001)。 结论:化疗早期ΔSUVmax%、Deauville评分和C－myc基因重排均与DLBCL患者的PFS有关，两两联合对DLBCL的预后有很好的预测价值。

abstractsObjective:To explore the prognostic value of the maximum standardized uptake value reduction proportion (ΔSUVmax%) on 18F-fluorodeoxyglucose (FDG) positron emission tomography combined with computed tomography (PET/CT) imaging, Deauville scores and C-myc gene rearrangement for the prediction of diffuse large B-cell lymphoma (DLBCL) in early chemotherapy. Methods:A total of 83 primary patients with pathologically confirmed DLBCL admitted in Shanxi Provincial Cancer Hospital from September 2010 to December 2016 underwent 18F-FDG PET/CT 1 week before and after early chemotherapy. The patients underwent post-chemotherapy examinations between 17 to 21 days after one cycle ( n=34) or two cycles ( n=49). The region of interest (ROI) was drawn and the ΔSUVmax% was calculated. Deauville 5-point scale was used to score the PET/CT imaging in early chemotherapy. Fluorescence in situ hybridization (FISH) was used to detect C-myc gene rearrangement. The follow-up time was from 36 to 111 months. The primary end-point of the study was progression-free survival (PFS). Receiver operating characteristic (ROC) analysis, χ2 test, Spearman correlation analysis, Log rank test, and Cox regression analysis were used to analyze the data. Results:Of 83 DLBCL patients, 19 progressed during the follow-up period. The optimal cut-off value of ΔSUVmax% for predicting tumor progression in early chemotherapy was 62.59%, and the Deauville score was taken as 5. The differences in sensitivity, specificity, and accuracy between the two methods were not statistically significant ( P>0.05). The ΔSUVmax% were negatively correlated with C-myc gene rearrangement and the Deauville scores ( rs= -0.889, -0.862, P<0.001). However, the Deauville scores was positively correlated with the C-myc gene rearrangement ( rs=0.781, P<0.001). The median PFS were 59 months and 16 months in ΔSUVmax%≥62.59% ( n=57) and ΔSUVmax%<62.59% ( n=26), respectively, with significant difference ( P<0.001). The median PFS for the Deauville score <5 subgroup (61 cases) and =5 subgroup (22 cases) was 59.0 and 15.0 months, respectively, with statistically significant differences ( P<0.001). The median PFS for patients with C-myc rearrangement subgroup (62 cases) and without rearrangement subgroup (21 cases) was 59.0 and 15.0 months, respectively, with statistically significant differences ( P<0.001). The median PFS for ΔSUVmax%<62.59% and Deauville score=5 subgroup, ΔSUVmax%<62.59% and C-myc rearrangement subgroup, Deauville score=5 and C-myc rearrangement subgroup were 15.5 months, 15 months and 13.5 months, respectively, with statistically significant differences ( P<0.001). Conclusion:ΔSUVmax%, Deauville score and C-myc gene rearrangement in early chemotherapy are all associated with PFS in DLBCL patients, and the combination of the two has a good predictive value for the prognosis of DLBCL