超声血流参数联合血清CXCL8、IL-1β、IL-10诊断乳腺癌的价值研究

Value of ultrasound blood flow parameters combined with serum CXCL8, IL-1β and IL-10 in the diagnosis of breast cancer

摘要:

目的 探讨超声血流参数联合血清趋化因子配体8（CXCL8）、白细胞介素-1β（IL-1β）、白细胞介素-10（IL-10）诊断乳腺癌的价值。方法 回顾性分析2021年1月—2023年1月三亚中心医院收治的170例乳腺癌患者的临床资料，将其作为乳腺癌组。另取同期该院170例乳腺良性肿瘤患者作为良性组。比较两组超声血流参数［搏动指数（PI）、阻力指数（RI）］及血清CXCL8、IL-1β、IL-10水平，比较不同临床病理特征乳腺癌患者的PI、RI、CXCL8、IL-1β、IL-10水平，绘制受试者工作特征（ROC）曲线分析PI、RI、CXCL8、IL-1β、IL-10单独及联合诊断乳腺癌的价值，采用多因素逐步Logistic回归模型分析PI、RI、CXCL8、IL-1β、IL-10与乳腺癌的关系。结果 乳腺癌组PI、RI、CXCL8、IL-1β、IL-10水平均高于良性组（P <0.05）。临床分期Ⅲ、Ⅳ期，中低分化，淋巴结转移的乳腺癌患者PI、RI、CXCL8、IL-1β、IL-10水平分别高于临床分期Ⅰ、Ⅱ期，高分化，无淋巴结转移的患者（P <0.05）。ROC曲线分析结果表明，PI、RI、CXCL8、IL-1β、IL-10诊断乳腺癌的敏感性分别为75.3%（95% CI：0.724，0.791）、71.2%（95% CI：0.659，0.748）、82.4%（95% CI：0.792，0.886）、85.9%（95% CI：0.804，0.911）、75.3%（95% CI：0.711，0.789），特异性分别为87.1%（95% CI：0.831，0.902）、88.8%（95% CI：0.833，0.914）、85.9%（95% CI：0.816，0.897）、88.2%（95% CI：0.852，0.916）、89.4%（95% CI：0.847，0.922）；PI、RI、CXCL8、IL-1β、IL-10联合诊断乳腺癌的敏感性为87.1%（95% CI：0.825，0.912），特异性为91.8%（95% CI：0.887，0.936），优于各项指标单独诊断。多因素逐步Logistic回归分析结果显示：PI ≥ 1.185 ［O^R=1.753（95% CI：1.178，2.609）］、RI ≥ 0.745 ［O^R=1.861（95% CI：1.199，2.889）］、CXCL8 ≥ 39.535 ng/mL ［O^R=2.017（95% CI：1.288，3.159）］、IL-1β ≥ 4.100 pg/mL ［O^R=2.115（95% CI：1.304，3.430）］和IL-10 ≥ 31.205 pg/mL ［O^R=2.344（95% CI：1.398，3.930）］是乳腺癌发生的危险因素（P <0.05）。结论 PI、RI、CXCL8、IL-1β、IL-10对乳腺癌均有一定诊断价值，且联合检测诊断效能更高。

Abstract:

Objective To investigate the value of ultrasound blood flow parameters combined with serum CXC chemokine ligand 8 (CXCL8), interleukin-1β (IL-1β) and interleukin-10 (IL-10) in the diagnosis of breast cancer.Methods Clinical data of 170 patients with breast cancer admitted to our hospital from January 2021 to January 2023 were retrospectively analyzed as the breast cancer group, and another 170 patients with benign breast tumors in our hospital during the same period were selected as the benign group. The ultrasound blood flow parameters [pulsatility index (PI) and resistance index (RI)] and serum levels of CXCL8, IL-1β and IL-10 were compared between the two groups and also among breast cancer patients with different clinicopathologic characteristics. The values of PI, RI, CXCL8, IL-1β and IL-10 alone as well as their combination in the diagnosis of breast cancer were analyzed by the receiver operating characteristic (ROC) curve. The multivariable Logistic regression model was used to analyze the correlations of PI, RI, CXCL8, IL-1β, and IL-10 with the breast cancer.Results PI, RI, and levels of CXCL8, IL-1β and IL-10 in the breast cancer group were higher than those in the benign group (P < 0.05). Besides, PI, RI, and levels of CXCL8, IL-1β and IL-10 in breast cancer patients with TNM stage Ⅲ to Ⅳ tumors, moderately- and lowly-differentiated tumors and lymph node metastasis were higher than those in patients with TNM stage Ⅰ to Ⅱ tumors, highly-differentiated tumors and no lymph node metastasis (P < 0.05). The ROC curve analysis demonstrated that the sensitivities of PI, RI, CXCL8, IL-1β and IL-10 in the diagnosis of breast cancer were 75.3% (95% CI: 0.724, 0.791), 71.2% (95% CI: 0.659, 0.748), 82.4% (95% CI: 0.792, 0.886), 85.9% (95% CI: 0.804, 0.911), and 75.3% (95% CI: 0.711, 0.789), with the specificities being 87.1% (95% CI: 0.831, 0.902), 88.8% (95% CI: 0.833, 0.914), 85.9% (95% CI: 0.816, 0.897), 88.2% (95% CI: 0.852, 0.916), and 89.4% (95% CI: 0.847, 0.922). The sensitivity of combined PI, RI, CXCL8, IL-1β and IL-10 in the diagnosis of breast cancer was 87.1% (95% CI: 0.825, 0.912), with the specificity being 91.8% (95% CI: 0.887, 0.936), which were greater than those of each indicator alone. Multivariable Logistic regression analysis showed that PI ≥ 1.185 [O^R =1.753 (95% CI: 1.178, 2.609) ], RI ≥ 0.745 [O^R = 1.861 (95% CI: 1.199, 2.889) ], CXCL8 ≥ 39.535 ng/mL [O^R = 2.017 (95% CI: 1.288, 3.159) ], IL-1β ≥ 4.100 pg/mL [O^R = 2.115 (95% CI: 1.304, 3.430) ] and IL-10 ≥ 31.205 pg/mL [O^R = 2.344 (95% CI: 1.398, 3.930) ] were risk factors for breast cancer (P < 0.05).Conclusions PI, RI, CXCL8, IL-1β and IL-10 are all of certain value for diagnosing breast cancer, and the combination of the above indicators yields higher diagnostic efficacy.