---

Background: Gastric cancer is a common and lethal disease throughout the world. This study was designed and carried out to determine the five-year survival rate of gastric cancer patients who had undergone surgical treatment at Taleghani Hospital of Tehran, and to assess its associated factors. Materials and Methods: In this historical-cohort study, information obtained from a total of 213 gastric cancer patients who underwent surgery at Taleghani Hospital of Tehran between 2003 and 2008 was included. In the analyses, Kaplan-Meier method, log-rank test, Cox proportional hazards model, and Lin-Ying additive hazards model were used. Results: The five-year survival rate and the median life expectancy in the studied patients were 14.6% and 29.6 months, respectively. Two covariates showed significant impacts on the gastric cancer patients’ data in both models: age at diagnosis and tumor size. We found that pathologic stage was significant under the Lin-Ying model (P=0.043) however, it was not significant under the Cox model (P=0.069). Other clinicopathological characteristics were not statistically significant (P>0.05). Conclusion: Since Cox and Lin-Ying models consider different aspects of the association between risk factors and the study outcome, it seems desirable to use both of them as complementary models but not as alternative models to obtain a more comprehensive understanding of data. This study showed that the early detection of patients in younger ages and in primary stages is important to decrease the risk of death in patients with gastric cancer and to increase the survival rate.

---

Background: Gastric cancer is the second leading cause of mortality due to cancer worldwide and it is the most common type of cancer in Iran. The aim of this study was to assess the effects of prognostic factors on the survival of patients with gastric cancer using the penalized spline in Cox model and compare it with Cox proportional hazards model. Materials and Methods: In this retrospective study, information obtained from a total of 216 gastric cancer patients who underwent surgery at Taleghani Hospital of Tehran between 2003 and 2008 years was included. Cox proportional hazards model and penalized spline in Cox model were used. R software was used for data analysis. The efficacy of these models was compared according to Akaike information criterion. Results: The five-year survival rate was 30% and the mean follow-up time was 16.60±15.28 months. Survival mean and median were 46 and 30 months, respectively. The analysis of Cox proportional hazards and penalized spline models resulted in age at diagnosis and tumor size as prognostic factors for survival time (P<0.05). Also, Akaike information criterion and survival curve for patients with a tumor size over 35 mm and age at diagnosis over 45 years were equal in both models. Conclusion: Cox proportional hazards and penalized spline models generated similar results thus, it is more appropriate to use Cox proportional hazards model because of its ease of interpretation and capability of modeling both continuous and discrete covariates. This study also showed if gastric cancer is diagnosed early, the relative risk of death will reduce.

---

Background: Colorectal cancer is the most common gastrointestinal cancer. Investigating the factors that predict survival time for these patients is important.The purpose of this study was comparison of parametric models by estimating the prediction error and also identifying the effective factors on predicted survival time of patients with colorectal cancer.

Materials and Methods: This cohort study was conducted with 600 patients who were suffered from colorectal cancer in Taleghani Hospital of Tehran between 2001 to 2005 and they were followed up for at least 5 years. For identifying the effective factors on survival time, of the patients we analyzed the data by some parametric models such as Weibull, Exponential and Log logistic and compared these models with the estimation of prediction error by apparent loss method.

Results: Among 600 patients there was 344 men (57.3%) and 256 wemon (47.7%). Of total, 151 patients were died that 62.3% of them were men. Univariate analysis showed that the effect of BMI, sex, staging of tumor, tumor site were significant but in multivariate model staging of tumor and BMI were significant. By the estimation of prediction error, the best model was Log logistic.

Conclusion: With respect to the importance of survival time prediction, we found that we can use the prediction error to compare the parametric models. In addition, because of effectiveness of tumor stages and BMI in the patients’ survival time, survival time could be increased by an on-time diagnosis and an appropriate controled diet.

---

Abstract

Background: 4cure models are a model to analyze survival data that these models exist for long term survivors. Cure models are a special type of survival model where it is assumed that there are a proportion of subjects who had never event, thus, survival curve will eventually reach a plateau. Therefore, standard survival models are not appropriate because they do not account for the possibility of cure.The aim of the present research is to apply non-mixture cure model to analyze survival of patients with colorectal cancer.

Materials and Methods: We studied 232 patients with colorectal cancer who were visited and treated at Taleghani Hospital Research Center for Gastroenterology and Liver Disease in Tehran. These patients were diagnosed from 1987 to 2012 and followed up until 2013. The Effect of age, gender, family history, body mass index and site of infection were studied. Kaplan-Meier and Non-Mixture cure Model were used for analzing data.

Results: The ten-year survival rate after diagnosis in the studied patients was 64 % .A total of 60 (25.8 %) deaths due to colorectal cancer were observed. The mean of age at the time of diagnosis was 51.6 years. Based on non-mixed cure model, the rangs of age was 45-65 years old and BMI were significant.

Conclusion: When the population is divided into two groups (susceptible and non- susceptible individuals), using Cox semi-parametric model is not appropriate. Therefore, we should use cure models.