Nomogram模型用于预测胃癌D2根治术后个体化预后
Nomogram Model Is Used to Predict the Individualized Prognosis of Gastric Cancer after D2 Radical Operation
DOI: 10.12677/ACM.2020.108237, PDF, HTML, XML, 下载: 593  浏览: 2,946 
作者: 郑 寒, 刘勇峰:延安大学附属医院,陕西 延安
关键词: 胃癌生存率预后列线图Gastric Cancer Survival Rate Prognosis Nomogram
摘要: 背景与目的:目前延安市尚未有针对胃癌患者预后的预测模型,本研究拟构建一种个体化预测模型预测延安市患者胃癌预后的列线图模型。方法:回顾性收集2010年1月1日至2014年11月1日在YAUAH胃肠外科(705例)行胃癌手术的患者。并采用COX比例风险回归模型进行多因素生存分析,并构建Nomogram模型。结果:多因素分析显示,年龄、肿瘤直径、T分期、N分期、LODDS和化疗次数是影响总生存率(OS)的独立危险因素,并依据此构建了Nomogram模型。结论:本研究构建的Nomogram模型能够预测胃癌患者3年、5年生存率,并有良好的临床应用前景。
Abstract: Background and Purpose: At present, there is no prediction model for the prognosis of gastric cancer in Yan’an City. This study intends to build a personalized prediction model to predict the prognosis of gastric cancer in Yan’an City. Method: The patients who underwent gastric cancer surgery in YAUAH Gastrointestinal Surgery (705 cases) were collected retrospectively from January 1, 2010 to November 1, 2014. The COX proportional hazards regression model was used for multi-factor survival analysis, and the Nomogram model was constructed. Results: Multivariate analysis showed that age, tumor diameter, T stage, N stage, LODDS, and the number of chemotherapy were independent risk factors affecting overall survival (OS), and based on this, the Nomogram model was constructed. Conclusion: The Nomogram model constructed in this study can predict the 3-year and 5-year survival rates of gastric cancer patients, and has good clinical application prospects.
文章引用:郑寒, 刘勇峰. Nomogram模型用于预测胃癌D2根治术后个体化预后[J]. 临床医学进展, 2020, 10(8): 1587-1593. https://doi.org/10.12677/ACM.2020.108237

1. 引言

胃癌是全球第五大恶性肿瘤,也是第三大癌症相关死亡原因,在东亚地区发病率尤其高 [1]。世界上每年有近一半的胃癌新发病例发生在中国,但近90%的胃癌患者在初次诊断时已进展或远处转移,特别是在非发达地区 [2]。准确预测预后对选择治疗策略和有效的医患沟通是至关重要的。

根治性胃癌切除联合局部淋巴结切除术是治疗无远处转移胃癌唯一有效的方法 [3]。GC患者生存预测的准确性是术后治疗决策和监测的关键。近年来,淋巴结分期方法一直都是国际关注的热点,每个分期都有其各自的优势与局限性。目前,被广泛接受的淋巴结分期是美国癌症联合会(American Joint Committee on Cancer, AJCC)/国际抗癌联盟(Union for International Cancer Control, UICC)的N分期,pN分期是胃癌独立预后因素 [4],但以pN作为基础的淋巴结分期的可靠性因其可导致淋巴结分期产生偏移而受到广泛质疑 [5]。鉴于pN分期的局限性,相关学者提出了阳性淋巴结对数比率(log odds of positive lymph nodes, LODDS)分期,并且在肺癌 [6]、乳腺癌 [7]、胰腺癌 [8]、食管癌 [9] 和结直肠癌 [10] 等几种癌症中,尤其是在根治术中淋巴结清扫不足的情况下,LODDS分期与AJCC-pN分期相比,在患者的生存率方面显示出明显的优越性。

2. 材料与方法

2.1. 一般资料

收集2010年1月1日至2014年11月1日在YAUAH胃肠外科(705例)行胃癌手术的患者。纳入标准:原发性胃癌;无合并恶性肿瘤;无术前化疗;无远处转移;R0切除(无肉眼或显微镜下残留肿瘤);无一个或多个缺失值。数据集中包括患者的人口统计学(年龄和性别)、病理特征(位置、直径、分化、组织学、浸润深度、转移淋巴结数目和检查淋巴结数目)、辅助化疗和随访数据(随访时间和生存率)。肿瘤的位置分为胃的上三分之一、中三分之一、下三分之一或超过胃体一半。胃食管交界处腺癌被归类为上三分之一胃癌。肿瘤大小定义为最大直径长度。分化程度分为高分化、中分化、低分化和其他。侵犯深度分为粘膜下层、固有肌、浆膜下、浆膜下及邻近器官侵犯。转移淋巴结数目按(AJCC) TNM第八版 [11] 分类(0, 1~2, 3~6, 7~15, ≥16)的淋巴结分组进行分类。LODDS通过log(pnod + 0.5)/(tnod-pnod + 0.5)估计(pnod:阳性淋巴结数;tnod:总淋巴结切除数) [5],0.5是为了避免无穷多个节点。LODDS值按X-tile分为:LODDS < 0.70;−0.70 ≤ LODDS < 0.27;LODDS ≥ 0.27。辅助化疗分为无、1~5次或≥6次。从医院记录中收集随访数据。随访资料收集自手术时间至最后一次随访时间。

2.2. LODDS的计算公式

阳性淋巴结对数比率(log odds of positive lymph nodes, LODDS)。定义为当淋巴结被获取时,阳性淋巴结数和阴性淋巴结比率的对数,其公式为log(pnod + 0.5)/(tnod-pnod + 0.5) (pnod:阳性淋巴结数;tnod:总淋巴结切除数) [12],分子和分母各加0.5的目的是为了避免出现被除数为0以及出现众多LODDS为0的情况。

2.3. 统计分析

YAUAH患者(n = 705)评估其临床病理特征。被认为与临床相关或与结果呈单变量关系的变量将被纳入多变量Cox比例风险回归模型。多变量模型包括单变量分析P < 0.2的候选变量。在多元Cox回归模型中,变量采用反向逐步选择法。在确定预后因素的预测模型的基础上,构建了预测3年和5年总生存率的列线图。

3. 结果

3.1. 临床特征

训练集中患者的临床特征见表1。在训练和验证数据集中,死亡率分别为55.2%。在训练集中,LN的平均数分别为15。大多数病人接受了辅助化疗。中位生存期分别为54个月。中位随访期分别为87个月。

Table 1. Characteristics of patients of radical resection of GC in the dataset

表1. 胃癌根治术患者的特征数据集

3.2. 训练集单因素分析的独立预后因素

单因素分析显示性别与生存率无明显相关性(P = 0.286)。多因素分析显示,年龄、肿瘤直径、T分期、N分期、LODDS和化疗次数是影响总生存率(OS)的独立危险因素(表2)。

Table 2. Predictors for prognosis of patients following surgical

表2. 术后患者预后的预测因素

3.3. OS的预后列线图

图1显示了数据集中OS所有重要独立因素的预后列线图。

4. 讨论

在本研究中,我们对701例胃癌根治术患者进行了评估,结果表明:1) LODDS与(AJCC) pN分期都是独立的预后因素;2) 建立了一个诺模图,并在预测胃癌患者的生存率方面表现良好。

本研究构建了延安地区胃癌患者预后的Nomogram模型。发现LODDS定量分析对淋巴结转移的胃癌患者的预后判断是有一定价值的,胃癌淋巴结转移对治疗决策和生存评估有很大的临床影响。(AJCC) pN分期被广泛应用,但其一个主要缺陷是预后的准确性受淋巴结总数的影响 [12]。最新的AJCC指南建议使用D2淋巴结切除术切除足够数量的淋巴结(不少于16个);然而,并非所有机构都能达到这一标准。一些研究报告称,在美国 [13] 和中国 [14] [15],只有29%和60.2%的病例获得了≥16个淋巴结。此外,在一些研究中,避免分期偏移的总淋巴结的最佳数目仍不清楚 [16] [17]。因此,在淋巴结总数不足的情况下,可能出现分期偏移现象。在我们的研究中,我们根据训练组评估了独立的危险因素,包括年龄、肿瘤直径、T分期、pN分期、LODDS和化疗次数。正如Antoni M. Szczepanik所认为的,由于胃癌的异质性,可能无法通过单一的测量来预测疾病过程。与其他研究人员相比,本研究通过随访后,将患者术后化疗次数也纳入到了本研究的模型中,其对患者的预后更有意义。

Figure 1. Prognostic nomogram of all important independent factors of OS in dataset

图1. 集中OS所有重要独立因素的预后列线图

综上所述,目前的研究表明LODDS是胃癌根治术患者的一个有用的预后指标。本研究构建了延安地区胃癌根治术后的个体化预后预测模型,该Nomogram模型优于传统TNM分期,有着良好的应用前景。

参考文献

[1] 彭智. GLOBCAN胃癌流行病学数据解读[J]. 肿瘤综合治疗电子杂志, 2018, 4(4): 63-65.
[2] Wang, W., Li, Y.F., Sun, X.W., et al. (2010) Prognosis of 980 Patients with Gastric Cancer after Surgical Resection. Chinese Journal of Cancer, 29, 923-930.
https://doi.org/10.5732/cjc.010.10290
[3] Shiozaki, H., Shimodaira, Y., Elimova, E., et al. (2016) Evolution of Gastric Surgery Techniques and Outcomes. Chinese Journal of Cancer, 35, Article No. 69.
https://doi.org/10.1186/s40880-016-0134-y
[4] Kwon, S.J. and Kim, G.S. (1996) Prognostic Significance of Lymph Node Metastasis in Advanced Carcinoma of the Stomach. British Journal of Surgery, 83, 1600-1603.
https://doi.org/10.1002/bjs.1800831136
[5] Kong, S.H., Lee, H.J., Ahn, H.S., et al. (2012) Stage Migration Effect on Survival in Gastric Cancer Surgery with Extended Lymphadenectomy: The Reappraisal of Positive Lymph Node Ratio as a Proper N-Staging. Annals of Surgery, 255, 50-58.
https://doi.org/10.1097/SLA.0b013e31821d4d75
[6] Deng, W., Xu, T., Wang, Y., Xu, Y., Yang, P., Gomez, D. and Liao, Z. (2018) Log Odds of Positive Lymph Nodes May Predict Survival Benefit in Patients with Node-Positive Non-Small Cell Lung Cancer. Lung Cancer, 122, 60-66.
https://doi.org/10.1016/j.lungcan.2018.05.016
[7] Wen, J.H., Ye, F., He, X.F., Li, S.J., Huang, X.J., Xiao, X.S. and Xie, X.M. (2016) Development and Validation of a Prognostic Nomogram Based on the Log Odds of Positive Lymph Nodes (LODDS) for Breast Cancer. Oncotarget, 7, 21046-21053.
https://doi.org/10.18632/oncotarget.8091
[8] Agalar, C., Aysal, A., Unek, T., et al. (2019) The Role of Log Odds of Positive Lymph Nodes in Predicting the Survival after Resection for Ampullary Adenocarcinoma. Pathology & Oncology Research, 26, 467-473.
https://doi.org/10.1007/s12253-019-00584-6
[9] Cao, J., Yuan, P., Ma, H., et al. (2016) Log Odds of Positive Lymph Nodes Predicts Survival in Patients after Resection for Esophageal Cancer. The Annals of Thoracic Surgery, 102, 424-432.
https://doi.org/10.1016/j.athoracsur.2016.03.030
[10] Amin, M.B., Edge, S., Greene, F., et al. (2016) AJCC Cancer Staging Manual. 8th Edition, Springer, New York.
https://doi.org/10.1007/978-3-319-40618-3_2
[11] Occhionorelli, S., Andreotti, D., Vallese, P., Morganti, L., Lacavalla, D., Forini, E. and Pascale, G. (2018) Evaluation on Prognostic Efficacy of Lymph Nodes Ratio (LNR) and Log Odds of Positive Lymph Nodes (LODDS) in Complicated Colon Cancer: The First Study in Emergency Surgery. World Journal of Surgical Oncology, 16, 186.
https://doi.org/10.1186/s12957-018-1483-6
[12] Twine, C.P., Roberts, S.A., Rawlinson, C.E., et al. (2010) Prognostic Significance of the Endoscopic Ultrasound Defined Lymph Node Metastasis Count in Esophageal Cancer. Diseases of the Esophagus, 23, 652-659.
https://doi.org/10.1111/j.1442-2050.2010.01072.x
[13] Reid-Lombardo, K.M., Gay, G., Patel-Parekh, L., Ajani, J.A. and Donohue, J.H. (2007) Treatment of Gastric Adenocarcinoma May Differ among Hospital Types in the United States, a Report from the National Cancer Data Base. Journal of Gastrointestinal Surgery: Official Journal of the Society for Surgery of the Alimentary Tract, 11, 410-419.
https://doi.org/10.1007/s11605-006-0040-8
[14] Strong, V.E., Song, K.Y., Park, C.H., Jacks, L.M., Gonen, M., Shah, M., Coit, D.G. and Brennan, M.F. (2010) Comparison of Gastric Cancer Survival Following R0 Resection in the United States and Korea Using an Internationally Validated Nomogram. Annals of Surgery, 251, 640-646.
https://doi.org/10.1097/SLA.0b013e3181d3d29b
[15] Peeters, K.C.M.J., Hundahl, S.A., Kranenbarg, E.K., Hartgrink, H. and van de Velde, C.J.H. (2005) Low Maruyama Index Surgery for Gastric Cancer: Blinded Reanalysis of the Dutch D1-D2 Trial. World Journal of Surgery, 29, 1576-1584.
https://doi.org/10.1007/s00268-005-7907-9
[16] Sun, Z., Xu, Y., Li, D.M., et al. (2010) Log Odds of Positive Lymph Nodes: A Novel Prognostic Indicator Superior to the Number-Based and the Ratio-Based N Category for Gastric Cancer Patients with R0 Resection. Cancer, 116, 2571-2580.
https://doi.org/10.1002/cncr.24989
[17] Lee, J.W., Ali, B., Park, C.H., et al. (2016) Different Lymph Node Staging Systems in Patients with Gastric Cancer from Korean. Medicine, 95, e3860.
https://doi.org/10.1097/MD.0000000000003860