食管神经内分泌癌患者原因别死亡竞争风险模型的建立与评价
|
摘要
目的采用Fine和Gray竞争风险模型分析影响食管神经内分泌癌(NEC)患者肿瘤特异性死亡和其他原因死亡的危险因素,并构建和评价预测模型。方法选取1998至2014年监察、流行病学和最终结果(SEER)数据库病理诊断为食管NEC的患者162例,采用竞争风险模型评估影响食管NEC患者肿瘤特异性死亡的独立危险因素,基于获取的危险因素建立预测模型预测食管NEC患者的原因别死亡风险。分别采用c指数、校正曲线、决策曲线分析对模型进行判别能力、预测效力及临床实用性的评价。结果纳入的162例患者中,121例死于原发肿瘤,17例死于其他原因。纳入研究的患者中,淋巴结转移、远处转移、手术及放疗的组间分布差异显著(P<0.05)。竞争风险的多因素分析显示,肿瘤位于中、下1/3段或其他位置者,其原因别死亡风险显著降低(HRs<0.41);合并转移者,其原因别死亡风险显著增高(HRs>2.26);手术、放疗和化疗均能显著降低其原因别死亡风险(HRs<0.56)。根据多因素分析结果建立食管NEC患者原因别死亡风险预测列线图。本预测模型的原因死亡与实际原因死亡一致性良好,显示出较为理想的预测效力(c指数为0.723)。决策曲线分析显示,当概率阈值为0.268～0.968时,该模型有明确的临床获益。结论本研究建立的食管NEC患者原因别死亡风险的预测模型具有较好的预测能力及临床实用性,有助于食管NEC患者原因别死亡高危人群的筛选及临床干预。
Objective To estimate probabilities of cause-specific mortality in patients with esophageal neuroendocrine carcinoma(NEC), and then to construct competing risk nomogram to predict prognosis in such patients using a large population-based cohort. Methods The surveillance, epidemiology and end results(SEER) database were used to identify a cohort of 162 patients diagnosed with esophageal NEC from 1998 to 2014. We estimated cumulative incident function(CIF) of cancer-specific mortality. We built the first nomogram for predicting cause-specific mortality with proportional subdistribution hazard model and validated it using a bootstrap cross-validation method and performed decision curve analysis to evaluate the clinical utility. Results One hundred of sixty-two patients enrolled in the study, 121 patients died of esophageal NEC and 17 patients died of other causes. There were significant differences in the distribution of lymph node metastasis, distant metastasis, surgery and radiotherapy among the patients enrolled in the study(P<0.05). Compared with patients with tumors in the upper third of the esophagus, patients with NEC involving the middle/lower third or other parts of the esophagus had substantially decreased risks of cause-specific mortality, with a subdistribution hazard ratio(HRs)<0.41. Patients with distant metastasis and patients whose metastasis status was unknown exhibited significantly increased risks compared with patients without metastasis, with an HRs>2.26. Relative to other patients, patients who underwent surgery, radiotherapy or chemotherapy had substantially decreased risks, with an HRs<0.56. The calibration plots demonstrated good concordance between the predicted and actual outcomes. The discrimination performance of the Fine-Gray model was evaluated using the c-index, which demonstrated 0.723 for cause-specific mortality in patients with esophageal NEC. The decision curve analysis yielded a range of threshold probability between 0.268 to 0.968 at which the clinical net benefit of the risk model was larger than those in hypothetical all-screening or no-screening scenarios. Conclusion We determined the probability of cause-specific mortality in patients with esophageal NEC on the basis of a large population-based cohort and developed the first nomogram based on a competing risks model. This nomogram will help clinicians identify who may be at higher risk of cause-specific mortality and assist in treatment and surveillance decision making.
Objective To estimate probabilities of cause-specific mortality in patients with esophageal neuroendocrine carcinoma(NEC), and then to construct competing risk nomogram to predict prognosis in such patients using a large population-based cohort. Methods The surveillance, epidemiology and end results(SEER) database were used to identify a cohort of 162 patients diagnosed with esophageal NEC from 1998 to 2014. We estimated cumulative incident function(CIF) of cancer-specific mortality. We built the first nomogram for predicting cause-specific mortality with proportional subdistribution hazard model and validated it using a bootstrap cross-validation method and performed decision curve analysis to evaluate the clinical utility. Results One hundred of sixty-two patients enrolled in the study, 121 patients died of esophageal NEC and 17 patients died of other causes. There were significant differences in the distribution of lymph node metastasis, distant metastasis, surgery and radiotherapy among the patients enrolled in the study(P<0.05). Compared with patients with tumors in the upper third of the esophagus, patients with NEC involving the middle/lower third or other parts of the esophagus had substantially decreased risks of cause-specific mortality, with a subdistribution hazard ratio(HRs)<0.41. Patients with distant metastasis and patients whose metastasis status was unknown exhibited significantly increased risks compared with patients without metastasis, with an HRs>2.26. Relative to other patients, patients who underwent surgery, radiotherapy or chemotherapy had substantially decreased risks, with an HRs<0.56. The calibration plots demonstrated good concordance between the predicted and actual outcomes. The discrimination performance of the Fine-Gray model was evaluated using the c-index, which demonstrated 0.723 for cause-specific mortality in patients with esophageal NEC. The decision curve analysis yielded a range of threshold probability between 0.268 to 0.968 at which the clinical net benefit of the risk model was larger than those in hypothetical all-screening or no-screening scenarios. Conclusion We determined the probability of cause-specific mortality in patients with esophageal NEC on the basis of a large population-based cohort and developed the first nomogram based on a competing risks model. This nomogram will help clinicians identify who may be at higher risk of cause-specific mortality and assist in treatment and surveillance decision making.
引文
[1] Hallet J, Law CH, Cukier M, et al. Exploring the rising incidence of neuroendocrine tumors: a population-based analysis of epidemiology, metastatic presentation, and outcomes [J]. Cancer, 2015,121(4): 589-597.
[2] Modlin IM, Oberg K, Chung DC, et al. Gastroenteropancreatic neuroendocrine tumours [J]. Lancet Oncol, 2008,9(1): 61-72.
[3] Chai SM, Brown IS, Kumarasinghe MP. Gastroenteropancreatic neuroendocrine neoplasms: selected pathology review and molecular updates [J]. Histopathology, 2018,72(1): 153-167.
[4] Ramezani M, Sadeghi M. Carcinoid tumour of the oesophagus: a systematic review [J]. Prz Gastroenterol, 2018,13(3): 196-199.
[5] Lee CG, Lim YJ, Park SJ, et al. The clinical features and treatment modality of esophageal neuroendocrine tumors: a multicenter study in Korea [J/OL]. BMC Cancer, 2014[2018-06-15]. https://www.ncbi.nlm.nih.gov/pubmed/25098730.
[6] Zhang Z, Geskus RB, Kattan MW, et al. Nomogram for survival analysis in the presence of competing risks [J/OL]. Ann Transl Med, 2017[2018-06-20]. http://atm.amegroups.com/article/view/16269.
[7] Wolbers M, Koller MT, Stel VS, et al. Competing risks analyses: objectives and approaches [J]. Eur Heart J, 2014,35(42): 2936-2941.
[8] van Walraven C, Hawken S. Competing risk bias in Kaplan-Meier risk estimates can be corrected [J/OL]. J Clin Epidemiol, 2016[2018-06-20]. http://europepmc.org/abstract/MED/26327491.
[9] Deng HY, Li G, Luo J, et al. The role of surgery in treating resectable limited disease of esophageal neuroendocrine carcinoma [J/OL]. World J Surg, 2018[2018-06-30]. https://www.ncbi.nlm.nih.gov/pubmed/29340724.
[10] Deng HY, Ni PZ, Wang YC, et al. Neuroendocrine carcinoma of the esophagus: clinical characteristics and prognostic evaluation of 49 cases with surgical resection [J]. J Thorac Dis, 2016,8(6): 1250-1256.
[11] Huang Q, Wu H, Nie L, et al. Primary high-grade neuroendocrine carcinoma of the esophagus: a clinicopathologic and immunohistochemical study of 42 resection cases [J]. Am J Surg Pathol, 2013,37(4): 467-483.
[12] Alese OB, Jiang R, Shaib W, et al. High-grade gastrointestinal neuroendocrine carcinoma management and outcomes: A National Cancer Database Study [J/OL]. Oncologist, 2018[2018-07-20]. https://www.ncbi.nlm.nih.gov/pubmed/30482824.
[13] Hong L, Zhang Y, Liu Z. Neuroendocrine carcinoma of esophageal and gastric cardia: clinicopathologic and immunohistochemistry study of 80 cases [J]. Oncotarget, 2018,9(12): 10754-10764.
[14] Yang M, Ke NW, Zeng L, et al. Survival analyses for patients with surgically resected pancreatic neuroendocrine tumors by World Health Organization 2010 Grading Classifications and American Joint Committee on Cancer 2010 Staging Systems [J/OL]. Medicine (Baltimore), 2015[2018-07-15]. https://journals.lww.com/md-journal/Fulltext/2015/12010/Survival_Analyses_for_Patients_With_Surgically.11.aspx.
[15] Chen WW, Wang F, Chen S, et al. Detailed analysis of prognostic factors in primary esophageal small cell carcinoma [J]. Ann Thorac Surg, 2014,97(6): 1975-1981.
[16] Ma Z, Cai H, Cui Y. Progress in the treatment of esophageal neuroendocrine carcinoma [J/OL]. Tumour Biol, 2017[2018-07-10]. http://europepmc.org/abstract/MED/28653897.
[2] Modlin IM, Oberg K, Chung DC, et al. Gastroenteropancreatic neuroendocrine tumours [J]. Lancet Oncol, 2008,9(1): 61-72.
[3] Chai SM, Brown IS, Kumarasinghe MP. Gastroenteropancreatic neuroendocrine neoplasms: selected pathology review and molecular updates [J]. Histopathology, 2018,72(1): 153-167.
[4] Ramezani M, Sadeghi M. Carcinoid tumour of the oesophagus: a systematic review [J]. Prz Gastroenterol, 2018,13(3): 196-199.
[5] Lee CG, Lim YJ, Park SJ, et al. The clinical features and treatment modality of esophageal neuroendocrine tumors: a multicenter study in Korea [J/OL]. BMC Cancer, 2014[2018-06-15]. https://www.ncbi.nlm.nih.gov/pubmed/25098730.
[6] Zhang Z, Geskus RB, Kattan MW, et al. Nomogram for survival analysis in the presence of competing risks [J/OL]. Ann Transl Med, 2017[2018-06-20]. http://atm.amegroups.com/article/view/16269.
[7] Wolbers M, Koller MT, Stel VS, et al. Competing risks analyses: objectives and approaches [J]. Eur Heart J, 2014,35(42): 2936-2941.
[8] van Walraven C, Hawken S. Competing risk bias in Kaplan-Meier risk estimates can be corrected [J/OL]. J Clin Epidemiol, 2016[2018-06-20]. http://europepmc.org/abstract/MED/26327491.
[9] Deng HY, Li G, Luo J, et al. The role of surgery in treating resectable limited disease of esophageal neuroendocrine carcinoma [J/OL]. World J Surg, 2018[2018-06-30]. https://www.ncbi.nlm.nih.gov/pubmed/29340724.
[10] Deng HY, Ni PZ, Wang YC, et al. Neuroendocrine carcinoma of the esophagus: clinical characteristics and prognostic evaluation of 49 cases with surgical resection [J]. J Thorac Dis, 2016,8(6): 1250-1256.
[11] Huang Q, Wu H, Nie L, et al. Primary high-grade neuroendocrine carcinoma of the esophagus: a clinicopathologic and immunohistochemical study of 42 resection cases [J]. Am J Surg Pathol, 2013,37(4): 467-483.
[12] Alese OB, Jiang R, Shaib W, et al. High-grade gastrointestinal neuroendocrine carcinoma management and outcomes: A National Cancer Database Study [J/OL]. Oncologist, 2018[2018-07-20]. https://www.ncbi.nlm.nih.gov/pubmed/30482824.
[13] Hong L, Zhang Y, Liu Z. Neuroendocrine carcinoma of esophageal and gastric cardia: clinicopathologic and immunohistochemistry study of 80 cases [J]. Oncotarget, 2018,9(12): 10754-10764.
[14] Yang M, Ke NW, Zeng L, et al. Survival analyses for patients with surgically resected pancreatic neuroendocrine tumors by World Health Organization 2010 Grading Classifications and American Joint Committee on Cancer 2010 Staging Systems [J/OL]. Medicine (Baltimore), 2015[2018-07-15]. https://journals.lww.com/md-journal/Fulltext/2015/12010/Survival_Analyses_for_Patients_With_Surgically.11.aspx.
[15] Chen WW, Wang F, Chen S, et al. Detailed analysis of prognostic factors in primary esophageal small cell carcinoma [J]. Ann Thorac Surg, 2014,97(6): 1975-1981.
[16] Ma Z, Cai H, Cui Y. Progress in the treatment of esophageal neuroendocrine carcinoma [J/OL]. Tumour Biol, 2017[2018-07-10]. http://europepmc.org/abstract/MED/28653897.