Characteristics of Inflammatory Cytokines and Prognostic Predictive Value in Breast Cancer Patients with Sleep Disorders
-
摘要:
目的 探讨乳腺癌患者睡眠障碍的血清炎症因子水平特征及其对新辅助化疗联合免疫治疗预后的预测价值。 方法 回顾性纳入2020年5月至2023年5月太原市中心医院肿瘤内科收治的397例激素受体阴性或HER-2阴性乳腺癌患者，其中2020年5月至2022年5月312例作为建模集，2022年6月至2023年5月85例作为时间验证集。所有患者于确诊后、新辅助化疗联合免疫治疗前1周内完成PSQI评分及血清炎症因子检测，并接受手术治疗。根据PSQI评分评估睡眠障碍情况，并随访2年预后。比较不同睡眠状态及睡眠障碍程度患者炎症因子水平，分析不良预后的影响因素，建立预测模型，并采用ROC曲线、校准曲线、H-L检验及DCA评价模型性能。 结果 乳腺癌睡眠障碍组血清IL-1β、IL-6、IL-8、TNF-α水平均高于无睡眠障碍组，IFN-γ水平低于无睡眠障碍组（P < 0.05）。重度睡眠障碍组血清IL-1β、IL-6、IL-8、TNF-α水平均高于中度组、轻度组，IFN-γ低于中度组、轻度组（P < 0.05）。预后不良组临床分期Ⅲ期、焦虑抑郁占比、PSQI评分、血清IL-1β、IL-6、IL-8、TNF-α水平均高于预后良好组，IFN-γ低于预后良好组（P < 0.05）。纳入临床分期、焦虑抑郁状态作为校正变量，以排除混杂因素对预后的影响，Logistic回归分析结果显示，临床分期、焦虑抑郁、PSQI评分、血清IL-1β、IL-6、IL-8、TNF-α、IFN-γ水平均为乳腺癌患者不良预后的影响因素（P < 0.05）。5项炎症因子联合检测预测乳腺癌患者不良预后的原始AUC为0.793，高于单项检测（Z = 3.388、3.268、3.708、4.439、3.390，P < 0.05），校正临床分期、焦虑抑郁后，校正AUC为0.789。列线图模型内部验证AUC=0.79（95%CI：0.70～0.88），时间验证集AUC=0.78（95%CI：0.67～0.89），校准曲线与理想曲线贴合良好。三种病理类型亚组中，炎症因子谱的变化趋势一致，联合预测AUC值介于0.776～0.801之间，亚组间AUC值差异无统计学意义（P > 0.05）。 结论 乳腺癌睡眠障碍患者血清IL-1β、IL-6、IL-8、TNF-α水平较高，IFN-γ水平较低，基于上述指标构建乳腺癌睡眠障碍患者不良预后的预警模型具有一定预测价值。 Abstract:Objective To explore the characteristics of serum inflammatory cytokine levels in breast cancer patients with sleep disorders and their prognostic predictive value for neoadjuvant chemotherapy combined with immunotherapy. Methods A total of 397 patients with hormone receptor-negative or HER2-negative breast cancer who were admitted to the Department of Oncology, Taiyuan Central Hospital, from May 2020 to May 2023 were retrospectively enrolled. Among them, 312 patients admitted between May 2020 and May 2022 served as the modeling set, and 85 patients admitted between June 2022 and May 2023 served as the temporal validation set. All patients underwent the Pittsburgh Sleep Quality Index （PSQI） assessment and serum inflammatory cytokine detection within 1 week after breast cancer diagnosis and before initiation of neoadjuvant chemotherapy combined with immunotherapy, followed by surgical treatment. Sleep disorders were evaluated according to PSQI scores, and patients were followed up for 2 years to assess prognosis. The levels of inflammatory cytokine were compared among patients with different sleep statuses and varying severities of sleep disorders. Risk factors associated with poor prognosis were identified, and a prognostic prediction model was developed. The performance of the model was evaluated using receiver operating characteristic （ROC） curves, calibration curves, the Hosmer–Lemeshow （H-L） goodness-of-fit test, and decision curve analysis （DCA）. Results Serum IL-1β, IL-6, IL-8 and TNF-α levels in the breast cancer sleep disorder group were significantly higher than those without sleep disorders, while IFN-γ levels were lower than those in patients without sleep disorders（P < 0.05）. The severe sleep disorder group had higher IL-1β, IL-6, IL-8, and TNF-α levels and lower IFN-γ levels than the moderate and mild groups, （P < 0.05）. The poor prognosis group had higher proportions of clinical stage III disease and anxiety/depression, higher PSQI scores, higher IL-1β, IL-6, IL-8, and TNF-α levels, and lower IFN-γ levels compared with the good prognosis group（P < 0.05）. After adjusting for clinical stage and anxiety/depression status, multivariate logistic regression showed that clinical stage, anxiety/depression, PSQI score, and levels of IL-1β, IL-6, IL-8, TNF-α, and IFN-γ were independent factors for poor prognosis in breast cancer patients （P < 0.05）. The original AUC of the combined detection of the five inflammatory cytokines combined for predicting poor prognosis in breast cancer patients was 0.793, which was significantly higher than that of single detection （Z = 3.388, 3.268, 3.708, 4.439, 3.390, P < 0.05）. After adjusting for clinical stage and anxiety/depression, the adjusted AUC was 0.789. The internal validation AUC of the nomogram model was 0.79（95%CI: 0.70-0.88）, and the time-validation set AUC was 0.78（95%CI: 0.67-0.89）. The calibration curve showed good agreement well with the ideal curve. In the three pathological subgroups, the trends of inflammatory cytokine changes was consistent, and the combined AUCs ranged from 0.776 to 0.801, with no significant differences between subgroups（P > 0.05）. Conclusion Breast cancer patients with sleep disorders have elevated serum IL-1β, IL-6, IL-8, and TNF-α levels and decreased IFN-γ levels. An early warning model for poor prognosis in breast cancer patients with sleep disorders constructed based on these indicators has predictive value. -
Key words:
- Breast cancer /
- Sleep disorders /
- Inflammatory factor spectrum /
- Immunotherapy /
- Poor prognosis /
- Predictive value
-
图 2 各项炎症因子预测乳腺癌患者不良预后的ROC曲线
Figure 2. ROC curves of individual inflammatory cytokines for predicting poor prognosis in breast cancer patients
图 3 乳腺癌患者不良预后的预警模型
Figure 3. Early warning model for poor prognosis of breast cancer patients
图 4 乳腺癌患者不良预后预警模型内部验证的校准曲线
Figure 4. Calibration curve for internal validation of the early warning model for poor prognosis in breast cancer patients
图 5 乳腺癌患者不良预后预警模型内部验证的ROC曲线
Figure 5. shows the ROC curve for internal validation of the early warning model for poor prognosis in breast cancer patients
图 6 乳腺癌患者不良预后预警模型内部验证的DCA曲线
Figure 6. shows the DCA curve for internal validation of the early warning model for poor prognosis in breast cancer patients
图 7 乳腺癌患者不良预后预警模型时间验证集验证的校准曲线
Figure 7. Calibration curve of the early warning model for poor prognosis in breast cancer patients in the temporal validation set
图 8 乳腺癌患者不良预后预警模型时间验证集验证的ROC曲线
Figure 8. ROC curve of the the early warning model for prognosis in breast cancer patients in the temporal validation set
图 9 乳腺癌患者不良预后预警模型时间验证集验证的DCA曲线
Figure 9. The DCA curve of the early warning model for poor prognosis in breast cancer patients in the temporal validation set
表 1 乳腺癌睡眠障碍组、无睡眠障碍组炎症因子比较（$ \bar x \pm s $）
Table 1. Comparison of inflammatory cytokines between the breast cancer sleep disorder group and the non-sleep disorder group （$ \bar x \pm s $）
组别 IL-1β（pg/mL） IL-6（pg/mL） IL-8（pg/mL） TNF-α（pg/mL） IFN-γ（ng/L） 乳腺癌睡眠障碍组（n = 198） 38.61 ± 3.10 120.99 ± 15.60 95.82 ± 9.12 25.12 ± 3.85 40.25 ± 5.82 乳腺癌无睡眠障碍组（n = 114） 31.09 ± 3.85 85.79 ± 8.02 65.05 ± 7.13 16.14 ± 2.14 45.31 ± 6.24 t 18.853 22.435 30.976 22.937 7.201 P <0.001** <0.001** <0.001** <0.001** <0.001** **P < 0.001。 
下载: 导出CSV
表 2 不同严重程度睡眠障碍患者炎症因子比较（$ \bar x \pm s $）
Table 2. Comparison of inflammatory cytokines among patients with different severities of sleep disorders （$ \bar x \pm s $）
组别 IL-1β（pg/mL） IL-6（pg/mL） IL-8（pg/mL） TNF-α（pg/mL） IFN-γ（ng/L） 轻度组（n = 68） 35.61 ± 3.10∆ 116.02 ± 13.62∆ 90.60 ± 10.31∆ 20.37 ± 4.37∆ 45.74 ± 6.78∆ 中度组（n = 92） 38.27 ± 3.85∆＃ 123.40 ± 13.14∆＃ 96.68 ± 9.10∆＃ 25.50 ± 3.18∆＃ 40.58 ± 5.47∆＃ 重度组（n = 38） 42.91 ± 4.86＃ 130.28 ± 10.26＃ 102.11 ± 10.36＃ 30.71 ± 3.35＃ 37.68 ± 5.15＃ F 44.187 15.858 17.887 6.086 26.409 P <0.001** <0.001** <0.001** <0.001** <0.001** **P < 0.001；与重度组比较，∆P < 0.05；与轻度组比较，＃P < 0.05。
下载: 导出CSV
表 3 预后良好组与预后不良组临床资料比较[（$ \bar x \pm s $）/n（%）]
Table 3. Comparison of clinical data between the good prognosis group and the poor prognosis group[（$ \bar x \pm s $）/n（%）]
项目 预后良好组（n = 261） 预后不良组（n = 51） χ2/t P 年龄（岁） 68.57 ± 4.52 68.94 ± 4.81 0.529 0.597 体质量指数（kg/m2） 21.67 ± 2.53 21.11 ± 2.12 1.482 0.139 糖尿病史 41（15.71） 10（19.61） 0.474 0.491 高血压史 30（11.49） 7（13.73） 0.203 0.652 乳腺癌家族史 28（10.73） 7（13.73） 0.385 0.535 孕产史 238（91.19） 46（90.20） 0.051 0.821 哺乳史 189（72.41） 39（76.47） 0.357 0.550 是否绝经 0.006 0.940 是 155（59.39） 30（58.82） 否 106（40.61） 21（41.18） 月经初潮年龄（岁） 14.72 ± 2.86 14.05 ± 2.13 1.588 0.113 肿瘤位置 0.109 0.742 左侧 147（56.32） 30（58.82） 右侧 114（43.68） 21（41.18） 病理类型 0.308 0.857 乳腺浸润性导管癌 102（39.08） 22（43.14） 乳腺导管内癌 90（34.48） 16（31.37） 乳腺浸润性小叶癌 69（26.44） 13（25.49） 临床分期 52.565 <0.001** Ⅰ～Ⅱ期 201（77.01） 13（25.49） Ⅲ期 60（22.99） 38（74.51） 焦虑抑郁 66.279 <0.001** 有 55（21.07） 40（78.43） 无 206（78.93） 11（21.57） 白细胞计数（×109/L） 4.20 ± 0.33 4.27 ± 0.55 1.221 0.224 红细胞计数（×1012/L） 4.38 ± 0.33 4.33 ± 0.40 0.940 0.348 血小板计数（×109/L） 198.58 ± 20.44 196.15 ± 20.58 0.772 0.441 血红蛋白（g/L） 127.74 ± 10.98 124.59 ± 11.47 1.871 0.062 PSQI评分（分） 10.58 ± 2.57 12.49 ± 3.78 4.454 <0.001** IL-1β（pg/mL） 35.05 ± 4.13 38.14 ± 4.14 4.885 <0.001** IL-6（pg/mL） 117.65 ± 15.31 126.69 ± 15.35 3.855 <0.001** IL-8（pg/mL） 90.31 ± 9.10 96.46 ± 9.18 4.408 <0.001** TNF-α（pg/mL） 22.72 ± 2.86 24.55 ± 2.53 4.255 <0.001** IFN-γ（ng/L） 39.24 ± 6.33 36.17 ± 6.32 3.172 0.002* *P < 0.05，**P < 0.001。
下载: 导出CSV
表 4 Logistic回归分析乳腺癌患者不良预后的影响因素
Table 4. Logistic regression analysis of factors the influencing poor prognosis in patients with breast cancer
因素 B SE Wald P OR 95%CI 临床分期 1.892 0.517 13.405 <0.001** 6.635 2.409～18.277 焦虑抑郁 1.539 0.441 12.194 <0.001** 4.660 1.963～11.064 PSQI评分 0.287 0.098 8.577 0.003* 1.333 1.100～1.615 IL-1β 0.166 0.044 14.482 <0.001** 1.181 1.084～1.287 IL-6 0.031 0.011 7.599 0.006* 1.032 1.009～1.055 IL-8 0.045 0.019 5.520 0.019* 1.046 1.007～1.086 TNF-α 0.136 0.063 4.770 0.029* 1.146 1.014～1.296 IFN-γ −0.244 0.821 7.487 0.006* 0.799 0.682～0.937 *P < 0.05，**P < 0.001。
下载: 导出CSV
表 5 各项炎症因子对乳腺癌患者不良预后的预测价值
Table 5. The predictive value of individual inflammatory cytokines for poor prognosis of breast cancer patients
指标 AUC 标准误 95%CI P Cut-off 约登指数 敏感度（%） 特异度（%） IL-1β 0.695a 0.039 0.641～0.746 <0.001** 37.62 0.364 66.67 69.73 IL-6 0.637a 0.042 0.581～0.691 <0.001** 121.89 0.264 64.71 61.69 IL-8 0.638a 0.043 0.582～0.691 <0.001** 100.90 0.246 37.25 87.36 TNF-α 0.604a 0.040 0.547～0.659 <0.001** 21.34 0.220 88.24 33.72 IFN-γ 0.654 0.043 0.598～0.707 <0.001** 40.29 0.268 49.02 77.78 联合检测（原始） 0.793 0.034 0.744～0.837 <0.001** 0.21 0.496 74.51 75.10 联合检测（校正） 0.789 0.035 0.738～0.835 <0.001** − 0.485 72.55 76.05 注：以预后不良组为阳性样本，预后良好组为阴性样本；a表示与联合相比，P < 0.05；**P < 0.001。多因素Logistic回归模型计算综合预测概率，将5项炎症因子作为自变量纳入模型，通过回归系数加权求和得到联合预测评分，再基于评分绘制ROC曲线。见图2。
下载: 导出CSV
表 6 睡眠障碍患者不同病理类型亚组炎症因子水平（$ \bar x \pm s $）
Table 6. Inflammatory cytokine levels in subgroups of different pathological types in patients with sleep disorders （$ \bar x \pm s $）
组别 亚组 IL-1β（pg/mL） IL-6（pg/mL） IL-8（pg/mL） TNF-α（pg/mL） IFN-γ（ng/L） 浸润性导管癌 预后良好（n = 52） 34.82 ± 3.95 116.87 ± 14.56 91.02 ± 8.85 21.85 ± 2.76 39.52 ± 6.08 预后不良（n = 12） 38.45 ± 4.02* 126.20 ± 15.12* 97.23 ± 9.02* 24.12 ± 2.98* 35.23 ± 5.85* 导管内癌 预后良好（n = 70） 34.52 ± 3.82 117.50 ± 14.82 90.35 ± 8.92 21.65 ± 2.68 39.85 ± 6.12 预后不良（n = 18） 37.98 ± 3.95* 126.11 ± 14.95* 96.52 ± 9.15* 23.85 ± 2.85* 36.02 ± 5.92* 浸润性小叶癌 预后良好（n = 37） 34.95 ± 3.88 118.91 ± 14.72 91.15 ± 8.95 21.95 ± 2.72 39.35 ± 6.05 预后不良（n = 9） 38.22 ± 4.05* 126.69 ± 15.05* 96.85 ± 9.25* 24.05 ± 2.92* 35.85 ± 5.88* 与同病理类型预后良好组比较，*P < 0.05。
下载: 导出CSV
表 7 睡眠障碍患者不同病理类型亚组炎症因子联合预测不良预后的效能
Table 7. The efficacy of combined inflammatory cytokines in predicting poor prognosis in different pathological type subgroups of patients with sleep disorders
组别 AUC 95%CI 敏感度 特异度 P 浸润性导管癌 0.801 0.682～0.920 75.00 78.85 0.312 导管内癌 0.785 0.685～0.885 72.22 77.14 0.348 浸润性小叶癌 0.776 0.652～0.900 77.78 75.68 0.286
下载: 导出CSV
-
[1] 李晨希, 查卓岑, 李娜, 等. 三阳性乳腺癌的强化辅助治疗方案选择[J]. 昆明医科大学学报, 2023, 44(10): 180-188. doi: 10.12259/j.issn.2095-610X.S20231020 [2] 李学涛, 王海波. 乳腺癌乳房重建术后患者报告结局评价[J]. 青岛大学学报(医学版), 2024, 60(6): 845-849. doi: 10.11712/jms.2096-5532.2024.60.193 [3] Yang G S, Starkweather A R, Lynch Kelly D, et al. Longitudinal analysis of sleep disturbance in breast cancer survivors[J]. Nurs Res, 2022, 71(3): 177-188. doi: 10.1097/NNR.0000000000000578 [4] Lan A, Li H, Shen M, et al. Association of depressive symptoms and sleep disturbances with survival among US adult cancer survivors[J]. BMC Med, 2024, 22(1): 225. doi: 10.1186/s12916-024-03451-7 [5] 钱芳, 孙永强, 张思涵, 等. 乳腺癌改良根治术后感染的病原学特征、影响因素及炎症因子分析[J]. 中国癌症杂志, 2025, 35(6): 563-569. doi: 10.19401/j.cnki.1007-3639.2025.06.005 [6] 夏羽, 陈杰. PD-1/PD-L1抑制剂在三阴性乳腺癌免疫治疗中的研究进展[J]. 重庆医学, 2024, 53(1): 169-171. [7] 洪晨忱, 姚峰. 肿瘤浸润性淋巴细胞在乳腺癌预后评估及免疫治疗中的应用进展[J]. 肿瘤防治研究, 2022, 49(11): 1180-1183. doi: 10.3971/j.issn.1000-8578.2022.22.0332 [8] 国家卫生健康委员会医政医管局. 乳腺癌诊疗指南(2022年版)[J]. 中国综合临床, 2024, 40(1): 1-30. doi: 10.3760/cma.j.cn112152-20230706-00281-1 [9] 孙振晓, 刘化学, 焦林瑛, 等. 医院焦虑抑郁量表的信度及效度研究[J/OL]. 中华临床医师杂志(电子版), 2017, 11(2): 198-201. [10] 路桃影, 李艳, 夏萍, 等. 匹兹堡睡眠质量指数的信度及效度分析[J]. 重庆医学, 2014, 43(3): 260-263. doi: 10.3969/j.issn.1671-8348.2014.03.002 [11] 老启芳, 雷奕, 李婛. 乳腺癌患者睡眠障碍的影响因素分析[J]. 重庆医学, 2024, 53(12): 1789-1792. doi: 10.3969/j.issn.1671-8348.2024.12.006 [12] Hwang Y, Knobf M T. Sleep health in young women with breast cancer: A narrative review[J]. Supportive Care Cancer, 2022, 30(8): 6419-6428. [13] 李华玉, 时萧寒, 张新蕊, 等. 203例胶质瘤患者睡眠障碍与炎症细胞因子的关联分析[J]. 山东大学学报(医学版), 2022, 60(12): 26-30+38. doi: 10.6040/j.issn.1671-7554.0.2022.0567 [14] Ciurescu S, Buciu V, Șerban D, et al. Role of cytokines in breast cancer: A systematic review and meta-analysis[J]. Biomedicines, 2025, 13(9): 2203. doi: 10.3390/biomedicines13092203 [15] Zwamel A H, Mousa H M, Abdul Azeez B A M, et al. The prediction power of thymidine phosphorylase and IL-6 in the relapse of breast cancer[J]. Pol Merkur Lekarski, 2025, 53(1): 88-93. doi: 10.36740/Merkur202501112 [16] Liao H, Li H, Song J, et al. Expression of the prognostic marker IL-8 correlates with the immune signature and epithelial-mesenchymal transition in breast cancer[J]. J Clin Lab Anal, 2023, 37(3): e24797. doi: 10.1002/jcla.24797 [17] 朱金成, 叶航, 王娅囡, 等. 多模态超声特征及血清IL-4、TNF-α检测预测三阴性乳腺癌术后复发转移的价值[J]. 放射学实践, 2023, 38(11): 1452-1457. doi: 10.13609/j.cnki.1000-0313.2023.11.015 [18] Heimes A S, Shehaj I, Almstedt K, et al. Prognostic impact of acute and chronic inflammatory interleukin signatures in the tumor microenvironment of early breast cancer[J]. Int J Mol Sci, 2024, 25(20): 11114. doi: 10.3390/ijms252011114 [19] Heimes A S, Härtner F, Almstedt K, et al. Prognostic significance of interferon-γ and its signaling pathway in early breast cancer depends on the molecular subtypes[J]. Int J Mol Sci, 2020, 21(19): 7178. doi: 10.3390/ijms21197178 [20] Zang H, Ni G, Gong L. Characterization of 5-inflammatory-gene signature to affect the immune status and predict prognosis in breast cancer[J]. Cent Eur J Immunol, 2022, 47(3): 218-233. doi: 10.5114/ceji.2022.121046 [21] 刘翠芳, 万冬桂. 乳腺癌术后继发失眠相关因素分析[J]. 中华中医药杂志, 2021, 36(10): 6188-6192. [22] 陈晓宇, 华琳, 吴建鑫. LINC00665联合CA153在乳腺癌诊断及预后评估中的价值研究[J]. 中国现代医生, 2025, 63(24): 17-20+94. doi: 10.3969/j.issn.1673-9701.2025.24.004 [23] 黄素珍, 李春华, 黄恩娥, 等. 乳腺癌靶向治疗患者睡眠障碍及其影响因素的研究[J]. 世界睡眠医学杂志, 2023, 10(4): 774-777. doi: 10.3969/j.issn.2095-7130.2023.04.022 [24] Shorofi S A, Nozari-Mirarkolaei F, Arbon P, et al. Depression and sleep quality among Iranian women with breast cancer[J]. Asian Pac J Cancer Prev, 2021, 22(11): 3433-3440. doi: 10.31557/apjcp.2021.22.11.3433 -
点击查看大图
计量
- 文章访问数: 3
- HTML全文浏览量: 3
- PDF下载量: 0
- 被引次数: 0
