摘要目的:探讨弥漫大B细胞淋巴瘤（DLBCL）患者血清脂蛋白a[Lp（a）]的水平及其临床意义。方法:回顾性队列研究。回顾性分析2017年1月至2022年6月就诊于常熟市第二人民医院87例DLBCL患者的临床资料（初治DLBCL组），选择健康体检者78名为对照组。比较初治DLBCL组和对照组Lp（a）的水平，及治疗后获得不同疗效的DLBCL患者Lp（a）水平。绘制受试者工作特征（ROC）曲线，分析血清Lp（a）预测DLBCL患者疗效的效能，计算曲线下面积（AUC），确定最佳临界值。根据最佳临界值，将DLBCL患者分为低Lp（a）组和高Lp（a）组，比较不同Lp（a）水平患者的临床病理特征；采用Cox比例风险模型分析DLBCL患者预后的影响因素；采用Kaplan-Meier法比较不同Lp（a）水平DLBCL患者无复发生存（RFS）和总生存（OS）。结果:初治DLBCL组患者Lp（a）水平为（0.24±0.09）g/L，对照组为（0.09±0.06）g/L，DLBCL组高于对照组，差异有统计学意义（ t＝3.61， P＝0.019）。87例患者中，治疗后达完全缓解（CR）54例，部分缓解（PR）23例，疾病进展（PD）10例。CR、PR、PD患者Lp（a）水平分别为（0.09±0.09）g/L、（0.12±0.08）g/L、（0.25±0.15）g/L，CR、PR患者Lp（a）水平均低于初治DLBCL患者Lp（a）水平[（0.24±0.09）g/L]，差异均有统计学意义（均 P<0.05），PD组Lp（a）水平与初治DLBCL患者比较差异无统计学意义（ P>0.05）。ROC曲线结果显示，血清Lp（a）预测DLBCL患者疗效的最佳临界值为0.25 g/L，AUC为0.776（95% CI：0.676～0.876， P<0.05），其灵敏度及特异度分别为66.67%、82.76%。根据Lp（a）的最佳临界值（0.25 g/L），将患者分为低Lp（a）组（≤0.25 g/L）（57例）和高Lp（a）组（>0.25 g/L）（30例）。高Lp（a）组患者中，乳酸脱氢酶>227 U/L、Ann Arbor分期Ⅲ～Ⅳ期、结外器官受累>1个的患者比例均高于低Lp（a）组，差异均有统计学意义（均 P<0.05）。Cox多因素分析结果显示，Ann Arbor分期Ⅲ～Ⅳ期、国际预后指数（IPI）评分3～5分及Lp（a）>0.25 g/L是影响DLBCL患者OS的独立危险因素（均 P<0.05）；Ann Arbor分期Ⅲ～Ⅳ期、IPI评分3～5分是影响DLBCL患者RFS的独立危险因素（均 P<0.05）。低Lp（a）组中位OS时间未达到；高Lp（a）组中位OS时间为21个月，两组OS比较差异有统计学意义（ P＝0.001）。低、高Lp（a）组中位RFS时间均为未达到，两组RFS比较差异无统计学意义（ P＝0.102）。 结论:DLBCL患者Lp（a）水平升高，Lp（a）可能是影响DLBCL预后的因素。

abstractsObjective:To investigate the level of serum lipoprotein a [Lp (a)] in patients with diffuse large B-cell lymphoma (DLBCL) and its clinical significance.Methods:A retrospective cohort study was performed. The clinical data of 87 patients with DLBCL who were treated at Changshu No.2 People's Hospital from January 2017 to June 2022 (the newly treated DLBCL group) were retrospectively analyzed, and 78 healthy physical examination subjects were selected as the control group. The level of Lp(a) in the two groups and the level of Lp(a) in DLBCL patients achieving different therapeutic effects after treatment were compared. The receiver operating characteristic (ROC) curve was used to analyze the efficacy of serum Lp(a) in predicting the therapeutic effect of DLBCL patients, and the area under the curve (AUC) was calculated to determine the optimal critical value. Based on the optimal critical value, patients with DLBCL were divided into low Lp(a) group and high Lp(a) group, and the clinicopathological characteristics of DLBCL patients with different Lp(a) levels were compared. Cox proportional hazards model was used to analyze the factors affecting the prognosis of DLBCL patients. Kaplan-Meier method was used to compare the relapse-free survival (RFS) and overall survival (OS) of DLBCL patients with different Lp(a) levels.Results:The level of Lp (a) in the newly treated DLBCL group was higher than that in the control group[ (0.24±0.09) g/L vs. (0.09±0.06) g/L], and the difference was statistically significant ( t = 3.61, P = 0.019). Among 87 patients, 54 achieved complete remission (CR), 23 achieved partial remission (PR), and 10 achieved progression of the disease (PD). The Lp (a) levels of patients achieving CR, PR, and PD were (0.09±0.09) g/L, (0.12±0.08) g/L, and (0.25±0.15) g/L, respectively. The Lp (a) levels in patients achieving CR and PR were lower than those in the newly treated DLBCL patients [(0.24±0.09) g/L], and the differences were statistically significant (all P < 0.05). There was no statistically significant difference in the Lp (a) levels between patients achieving PD and the newly treated DLBCL patients ( P > 0.05). The ROC curve results showed that the optimal critical value of serum Lp (a) in predicting the efficacy of DLBCL patients was 0.25 g/L, AUC was 0.776 (95% CI: 0.676-0.876, P < 0.05), and its sensitivity and specificity was 66.67%, 82.76%, respectively. According to the optimal critical value of Lp (a) (0.25 g/L), patients were divided into the low Lp (a) group (≤ 0.25 g/L) (57 cases) and the high Lp (a) group (>0.25 g/L) (30 cases). The proportion of patients with lactate dehydrogenase level >227 U/L, Ann Arbor stage Ⅲ-Ⅳ, and extranodal organ involvement >1 in the high Lp (a) group was higher than that in the low Lp (a) group, and the differences were statistically significant (all P < 0.05). Cox multivariate analysis results showed that Ann Arbor stage Ⅲ-Ⅳ, international prognostic index (IPI) score 3-5, and Lp (a)>0.25 g/L were independent risk factors for OS in DLBCL patients (all P < 0.05); Ann Arbor stage Ⅲ-Ⅳ and IPI score 3-5 were independent risk factors for RFS in DLBCL patients (all P < 0.05). The median OS in the low Lp (a) group was not reached; the median OS of the high Lp (a) group was 21 months, and there was a statistically significant difference in OS between the two groups ( P = 0.001). The median RFS time was not reached in the low Lp (a) group and the high Lp (a) group; and there was no statistically significant difference in RFS between the two groups ( P = 0.102) . Conclusions:Lp(a) level of DLBCL patients is increased, and Lp(a) could be a factor influencing the prognosis of DLBCL.