Rasouli Mahboube, Ghadimi Mahmood Reza, Mahmoodi Mahmood, Mohammad Kazem, Zeraati Hojjat, Hosseini Mostafa
Department of Epidemiology and Biostatistics, School of Public Health, Tehran University of Medical Sciences, Tehran, Iran.
Asian Pac J Cancer Prev. 2011;12(9):2359-63.
Esophageal cancer is a major cause of mortality and morbidity in the Caspian littoral north-eastern part of Iran. The aim of this study was to calculate cure function as well as to identify the factors that are related to this function among patients with esophageal cancer in this geographical area.
Three hundred fifty nine cases of esophageal cancer registered in the Babol cancer registry during the period of 1990 to 1991 (inclusive) were followed up for 15 years up to 2006. Parametric cure model was used to calculate cure fraction and investigate the factors responsible for probability of cure among patients.
Sample of subjects encompassed 62.7% men and 37.3% women, with mean ages of diagnosis was 60.0 and 55.3 years, respectively. The median survival time reached about 9 months and estimated survival rates in 1, 3, and 5 years following diagnosis were 23%, 15% and 13%, respectively. Results show the family history affects the cured fraction independently of its effect on early outcome and has a significant effect on the probability of uncured. The average cure fraction was estimated to be 0.10.
As the proportionality assumption of Cox model does not meet in certain circumstances, a parametric cure model can provide a better fit and a better description of survival related outcome.
食管癌是伊朗东北部里海沿岸地区死亡和发病的主要原因。本研究的目的是计算治愈函数，并确定该地理区域食管癌患者中与该函数相关的因素。
对1990年至1991年（含）期间在巴博尔癌症登记处登记的359例食管癌病例进行了长达15年的随访，直至2006年。采用参数治愈模型计算治愈比例，并调查患者中治愈概率的相关因素。
研究对象样本中男性占62.7%，女性占37.3%，平均诊断年龄分别为60.0岁和55.3岁。中位生存时间约为9个月，诊断后1年、3年和5年的估计生存率分别为23%、15%和13%。结果表明，家族史独立于其对早期结局的影响而影响治愈比例，且对未治愈概率有显著影响。估计平均治愈比例为0.10。
由于Cox模型的比例假设在某些情况下不成立，参数治愈模型可以提供更好的拟合以及对生存相关结局的更好描述。
